Month: August 2024

You open your computer on a Monday morning, and you have a few Slack messages about a campaign you're launching on Tuesday.

After you‘ve answered those, you check your inbox and see you’ve been tagged in some slides for that same campaign.

Once you're done responding, you hop on a Zoom call to chat with stakeholders about last-minute tasks that need to be completed for launch. A few of the stakeholders would like you to email a follow-up from the meeting, so you do.

But others would rather you tag them in the appropriate Google docs, so you do that, too.

Suddenly it‘s 1pm, and you’ve done nothing substantial on your to-do list to get this project launched. Your entire day has been hopping in and out of various messaging apps, slide decks, and Zoom calls, just trying to get everyone aligned.

Sound familiar?

I spoke with Jake Cerf, Head of Corporate Marketing at Asana, to untangle the biggest challenges most teams face when it comes to productivity in 2024 – and how you can solve them.

What Teams Get Wrong When It Comes to Productivity

Jake empathizes with the chaos that can ensue when you don't focus on creating efficient processes for team-wide productivity.

“It can get chaotic,” he told me, adding, “Before I joined Asana, I reflected back on how I spent my time coordinating with folks — and it was a mess. We would be on email, Slack, and Google docs, and slides. And you never really knew who was doing what, and when, and it was too easy to lose sight of the objective we were all after.”

Which sounds painfully relatable. Fortunately, he has some tried-and-true tips for cleaning up your team's processes and creating more scalable options to improve cross-functional collaboration.

1. Each team leader needs to know how their work ladders up to corporate objectives — and they need to make it clear in their workflows.

People always want to know how their work connects to broader strategic initiatives. They want to feel seen, valued, and know they are making an impact. So much of a leader's job is about making sure people are working on the right priorities, and aligning to goals that move the needle.

That’s what makes a product like Asana so crucial. Jake has an easy time ensuring he isn‘t micro-managing his team on specific tasks, and that’s because in Asana he can see how each sub-task his team is responsible for ladders up to the company's key objectives for 2024.

Additionally, to solve for conflicting cross-department goals, it can be helpful to use one centralized productivity tool that highlights the top-down priorities for the company.

“As a leader, so much of our job is making sure people are working on the right things, helping unblock team members and enabling them to have a North star. It's good for productivity because when folks feel like they're working on things that matter, they do better work,” Jake says.

He adds, “You don't have to be as in-the-weeds on the details. You can tell team members the what and the why, and they can figure the rest out. But being clear about big picture objectives unlocks productivity up, down, and across the organization.”

If you‘re dealing with productivity issues, start by ensuring each leader is aligned on the major company objectives for 2024 – and then task them with demonstrating how all of their team’s projects ladder up to that ultimate goal. If a task doesn‘t fit, it’s time to consider re-focusing on the activities that do.

2. Assign your AI a "role" to uplevel your team's productivity.

There's been plenty of conversation surrounding AI over the past two years, but people are still skeptical about the improvements it can make to their daily lives.

In fact, 62% of marketers globally believe people should use some AI in their roles. For Jake, AI has proven much more useful as a teammate rather than just a tool.

"My life changed drastically when I stopped prompting AI with generic requests like, 'Please write this blog post‘, and instead honed in on who I wanted AI to be: ’Please write this blog post as if you're a tech writer at a large-scale SaaS company.'"

Jake highly recommends assigning AI a “role” when leveraging AI for productivity.

“When teams are working on an important initiative, and you give each AI bot its own specific role, the output is much greater. Let's say you're writing a blog post — you can assign AI to be the editor, the fact-checker, or the content strategist.”

“Or,” He adds, “if you use tools like Asana, you’ll have access to AI that is one of the world's greatest project managers. It can help you unblock issues and triage requests and make sure people are working on the right things.”

Ideally, the productivity tools you leverage already have AI capabilities built-in. If not, look into which plug-ins or external tools you might use to increase efficiency.

3. Leverage AI to minimize busywork.

The antithesis of productivity is busywork.

If your team is bogged down by menial tasks, they likely don‘t have the energy or time to focus on the big picture objectives that account for most of your team’s impact.

That's a major roadblock – and one that can be solved with AI.

Jake offers the example of repurposing content as one opportunity for increased productivity. He says, “With AI, you can take a keynote presentation and ask AI to draft a blog post on the keynote. Or, you can take your keynote script and ask AI to design the presentation itself.”

He continues, “Finding new avenues to increase the longevity and impact of your content is one of the best ways to use AI.”

Additionally, Jake encourages marketers to leverage AI for content creation, as well as more creative outputs like manager reviews, sending feedback to teammates, riffing on ideas, role playing scenarios, and more.

4. Have one centralized workspace for teams to work cross-functionally.

Finally, none of this is possible without creating a strong foundation for efficient, scalable cross-functional collaboration.

Remember those slide decks and Google docs and Slack messages and emails I mentioned earlier? Why not try to put more of your work in one centralized place?

“Productivity comes down to visibility,” Jake says. “Your team needs to be rowing in the same direction. Having a tool like Asana has been super helpful for our team productivity — you need a place where you can set your goals and then track all of the team's work and hold people accountable.”

“Plus,” he adds, “It's crucial you use the same centralized workspace when you're setting strategy so that you have alignment around the tasks and initiatives that will help you achieve your goals.”

In other words – jumping between 30 different messaging and content creation apps and tools isn‘t conducive to long-term productivity. As a leader, it’s your job to figure out how to centralize as much as you can in one place – and then use AI to supercharge it all.

To learn more about how HubSpot and Asana are helping marketers drive productivity, take a look at the HubSpot and Asana integration available today.

Email marketing is integral to any marketing strategy because it’s a great way to generate leads and convert audiences.

Whether you’re creating your first strategy or looking to modify your process, AI email marketing tools can help you save time, optimize your strategy, and meet your email goals.

In this piece, I’ll go over how AI email marketing tools work, new data about how marketers currently use AI for email marketing, and a list of tools you can leverage in your role.

Table of Contents

• What is AI email marketing?
• Why should you use AI email marketing?
• What are the challenges of AI in email marketing?
• How Marketers Are Using AI in Email Marketing [New Research]
• 15 AI Email Marketing Tools

What is AI email marketing?

AI email marketing is a machine-learning-powered process that helps marketers create email campaigns that reach the right audiences at the right time with the right messages.

AI email marketing tools use data (like your historical performance data) to help you optimize your email strategy, automation to help you save time on repetitive tasks (like triggering an email workflow), and generative AI to help you create email content.

When using AI in email marketing, you can do things like:

• Analyze past email performance to identify how to optimize your email strategies, like the best time to send your emails or the subject lines that get the most clicks.
• Compile email analytics so you understand the health of your campaigns.
• Trigger email workflows after people take a specific action.
• Clean up your email lists to improve deliverability.
• Write compelling copy that speaks to your audience.
• Personalize email content to specific audience segments.

Some tools have one specific function, like a generative AI email tool, while others offer multiple features.

Why should you use AI in email marketing?

I've found that the most significant benefit of using AI in email marketing is that it saves time while improving performance. The routine processes you spend time on can happen instantly, and you can launch your optimized campaigns faster.

Most AI email marketing tools are also powered by machine learning, meaning that they use data points (from your business and sometimes your industry) to help you optimize your email strategy.

You won’t be left to guess what works best because the AI can look at your past emails, and you can benchmark your performance against competitors to see where you can improve.

If you’re interested in learning more about the impact of AI on marketing, check out this Marketing Against The Grain episode about marketing opportunities that AI unlocks for business.

Click here to listen to the full episode.

What are the challenges with AI in email marketing?

Of course, using AI in email marketing is not without its challenges. AI tools are still relatively new in terms of technology, so they’re not perfect.

Here are a few challenges marketers face when using AI in email marketing and how they solve those issues.

Tone

Anyone who’s asked ChatGPT to write an email for them knows the first output will usually sound formal. Tone is undoubtedly the biggest challenge marketers face when using generative AI for email marketing.

Most AI tools need to be trained to match your brand’s voice and tone, and even then, you’ll probably still need to edit it to sound exactly how you want.

Meg O'Neill, co-founder of Intuitive Marketing Collective, gets around this challenge by using clear and specific prompts.

“I want [my emails] to sound like I’m talking to a friend,” says O’Neill. “I’ve added this requirement to my prompt, and it’s helped a lot. Sometimes, I’ll give [the AI tool] the name of a famous business person and ask it to write in a similar tone.”

Quality

Whether you use AI to write the body copy of your email, generate subject line ideas, or outline an email funnel structure, the quality of the overall content isn’t going to be perfect the first time around.

“The quality is not always there,” says Jeanne Jennings, Founder and Chief Strategist for Email Optimization Shop, an email marketing consultancy.

To overcome this challenge, Jennings takes a collaborative approach by “micro-managing the AI tool at each stage to get the quality content I need. Without my collaborative approach, the output is usually junk,” she states.

Data Accuracy

Ben Schreiber, Head of ecommerce at Latico Leathers, says the biggest challenge he’s faced using AI in email marketing is data accuracy and integration.

“Good quality data is paramount to the success of using AI systems since any inaccuracies may lead to wrong output results,” says Schreiber.

“We have had issues with outdated or incomplete data, which directly affects how well or poorly our campaigns perform.”

How Marketers Are Using AI in Email Marketing [New Research]

Our State of AI in Marketing report surveyed 1,062 U.S. marketing and advertising professionals about how they’re currently using AI.

For starters, AI usage for marketing has increased significantly since 2023.

74% of marketers who responded to the survey said they use at least one AI marketing tool.

While chatbots are the most popular marketing tool used by marketers, 25% of marketers use AI through existing CRM and marketing tools with AI-enhanced features.

Here are a few specific ways marketers are using AI for email marketing.

Content Creation

By far, the most common use case for AI tools among marketers is content creation. Of respondents who report using AI, 43% say they use it for content creation.

While marketers use AI for everything from creating images to creating outlines, the most popular type of content to create is written content.

There’s no denying AI’s ability to generate text — both long-form and short-form — in an instant, making it a great tool for email campaigns.

In fact, 47% of marketers use AI to create email marketing content such as newsletters or campaigns.

Testing

From subject lines to body copy to design elements, testing is essential to increase engagement and ultimately improve your email marketing performance.

Of marketers, 27% say they use AI tools for brainstorming, and it’s safe to say testing email content falls into that category.

Not sure how to use AI for experimenting or brainstorming?

“Ask AI to provide you with specific testing ideas so you always have a fantastic list to choose from to continue improving your email performance,” recommends email marketer Bethany Fiocchi Root, CEO and founder of Oceanview Marketing.

Data Collection

Email marketing relies on data, but data collection can be a time-consuming process for busy marketers. That’s where AI comes in.

Not only can AI tools help automate data collection — decreasing time spent on these tasks significantly — they can be more precise with the information.

A majority of marketers (including those who don’t currently use AI) agree that AI can help their organizations share data more effectively.

From a leadership standpoint, 39% of marketing directors agree that AI and automation tools help employees make data-driven decisions.

And the more data you have, the more you can personalize your email marketing.

In fact, 69% of marketers agree that AI tools can help them personalize the experience their customers get.

Automation

Finally, marketers use generative AI to automate their processes. With AI, everything from scheduling email campaigns to email data entry is taken care of.

Of the marketers surveyed, 75% say using AI for automation helps them reduce time spent on manual tasks and more time on critical or creative tasks.

In other words, AI helps them focus on the aspects of the job they enjoy rather than on administrative tasks.

15 AI Email Marketing Tools

1. HubSpot AI Tools

Click here to learn more about HubSpot AI tools.

HubSpot has multiple email marketing tools and features to leverage to drive clicks and conversions.

AI Features

• Email Marketing Software that helps you easily create email workflows and triggers to reach your target audiences with the right messages at all stages of their journey.
• Inbox automation tool that scans your emails and recommends tasks based on email content and can auto-populate contact properties (like name and phone number) from every first-time email to create a customer profile.
• Content assistant uses generative AI to help you write high-quality email content, and you can ask ChatSpot to quickly write things like professional follow-up emails or thank you notes to prospects.

Price: Free tools are available. Starter plans cost $20 per month. Professional plans cost $890 per month. Enterprise plans cost $3,600 per month.

2. Mailchimp

Mailchimp’s email automation software helps ecommerce businesses create automated email workflows that reach audiences at the best possible time.

AI Features

• Its Content Optimizer compares your email data to industry benchmarks to give you recommendations for optimizing your campaigns and email content.
• You can choose from different versions of AI-generated content that match your intent and brand tone.
• Its Creative Assistant leverages your brand assets to create unique email designs you can personalize to different contacts.

Price: Free forever plans are available. Essential plans cost $13 per month. Standard plans cost $20 per month. Premium plans cost $350 per month.

3. Sendgrid

Sendgrid helps you create an automated email marketing process with custom workflows and triggers.

AI Features

• Its real-time API scans your email lists and removes junk or undeliverable email addresses to lower your bounce rate and ensure you reach more people.
• Get data-driven insights and recommendations for improvement based on your historical metrics and email performance.
• AI paces your email send and monitors your reputation with ISPs.

Price: Free trials are available. Basic plans cost $15 per month. Advanced plans cost $60 per month.

4. Phrasee

Phrasee uses AI to help you create effective email campaigns and content to share with your audience.

AI Features

• Its deep learning model and language insights leverage your historical data to tell you what works best with your audience and what inspires clicks for an optimized campaign.
• The Magic Button helps you generate email content (like subject lines or in-email CTAs) that will resonate with your audience.
• It always uses your custom guidelines and messaging to ensure everything you create is on-brand.

Price: Contact for pricing.

5. Drift

Drift offers an AI-powered inbox management tool that helps you clean up your email lists and improve deliverability.

AI Features

• Its Email Bots leverage machine learning to interpret emails and help you reply with engaging, conversational emails that inspire responses.
• AI can qualify a lead as ready for sales and automatically introduce the prospect to the right salesperson for seamless marketing to sales handoff.
• Use different Email Bots for your unique business need, like the follow-up email bot, abandoned chat email bot, and webinar email bot.

Price: $2,500 per month.

6. GetResponse

Use Get Response to design behavior-based email workflows to engage with audiences at key moments with content personalized to their needs.

AI Features

• Share keywords or phrases, email goals, and tone with the GPT-powered email generator that leverages industry data to produce emails most likely to increase your conversions.
• Display different images, text, or AI-driven product recommendations in each email.
• The AI subject line generator helps you test subject lines and learn what stands out in your subscribers’ inboxes.

Price: A free 30-day trial is available; paid plans start at $19 per month.

7. Levity

Levity’s software helps you manage your inbox, understand your email health, and save time.

AI Features

• Build an AI tool unique to your business by uploading your data that it will learn from and use to make human-level decisions.
• Create different AI blocks for every email workflow you want to run (like a workflow for responding to emails).
• Share unique categorization criteria with your AI to automatically sort emails as soon as you receive them.

Price: A 30-day free trial is available. Startup plans cost $49 per month. Business plans cost $139 per month.

8. Superhuman

Superhuman is an AI-powered inbox management tool that helps you streamline your processes. Best for teams that use Gmail or Outlook.

AI Features

• Immediately sort incoming emails into a split inbox based on your custom rules so you can sort spam from genuine humans and focus on what needs attention.
• Use its Snippets tool to create pre-built templates for phrases, paragraphs, or entire emails that you can quickly add to emails to automate responses.
• Set reminders for email tasks, like following up on unanswered emails or a reminder to respond to a message you snoozed for later.

Price: Starter plans cost $30 per month. Growth plans cost $45 per month. Enterprise pricing is available.

Most of the tools listed above have multiple AI features, like email writing help to automated inbox sorting. Below, we’ll go over AI email marketing tools that only offer generative features.

9. Hive

Hive offers an easy-to-use and time-saving tool for your email marketing. Simply share a brief prompt of what you’re looking for with its Notes AI, and it’ll help you generate a perfect response.

Price: There is a free forever plan. Teams plans cost $12 per user a month. Enterprise pricing is available.

10. ChatGPT

ChatGPT is a generative AI tool that you can use to write your marketing emails, and all you have to do is enter a descriptive prompt into the chat. It’s a conversational tool, so you can ask it to rewrite the email until you’re satisfied.

Price: There is a free research preview. ChatGPT Plus costs $20 per month.

11. Zapier

Zapier runs on Zaps, automated workflows you can customize to your needs. You can create an email-based Zap to generate email copy with an API key from OpenAI.

Whenever you receive an email matching your Zaps rules, it’ll prompt GPT-3 to write an appropriate response.

Price: A free forever plan is available. Professional plans cost $19.99 per month. Team plans cost $69 per month. Contact Zapier for Enterprise pricing.

12. Copy.ai

Copy.ai is an email copywriting tool you can use to create high-converting emails. It can write email content for you, suggest subject lines, and help you stay on track with suggestions to improve email quality.

Price: A free forever plan is available. Starter plans cost $36 per month. Advanced plans cost $186 per month. Enterprise pricing is available.

13. Compose.ai

Compose.ai is powered by GPT-3 and helps you write personalized and on-brand emails.

Its autocomplete feature suggests how you can finish what you’re writing, and its suggestions and generations are always tone- and brand-relevant because it learns your unique brand voice.

It’s an always-free Chrome extension, so you can easily use it on your favorite sites.

Price: A free forever plan is available. Premium plans cost $9.99 per month. Ultimate plans cost $29.99 per month. Enterprise pricing is available.

14. Grammarly

Grammarly’s machine-learning copy-editing tool recognizes in-text errors and suggests how to fix them. I use this feature all the time to assist in my content writing.

GrammarlyGo extracts the context from short prompts and helps you instantly generate appropriate email replies. Leverage the tools on its website, as a Chrome extension, or within your favorite email client.

Price: A free forever plan is available. Premium plans cost $12 per month. Business plans cost $15 per month.

15. Jasper

Jasper Commands helps you create effective marketing emails quickly with machine learning algorithms.

Use it to write entire emails or email subject lines, and its outputs always match your business’ unique writing style and tone for brand consistency.

Price: Free trial is available. Creator plans cost $39 per month. Pro plans cost $59 per month. Custom business pricing is available.

Leveling Up Your Emails With AI

From writing copy to generating subject lines and even collecting and organizing data to improve personalization, a majority of marketers agree that using AI for email marketing makes their job easier.

When AI tools are used as just that — tools — they can improve your email marketing campaigns and give you more time to spend on creative tasks that make you better at your job.

It often surprises people to learn just how unfunny making comedy can be. I worked with this week’s master of marketing some years ago out of The Onion’s HQ, so we’ve both been behind the scenes. A business is still a business, and marketing is still marketing.

Which isn’t to say it can’t be a helluva lot of fun.

I talked to Hassan S. Ali, the creative director of brand at Hootsuite, where he describes his job as “leading a team of creatives to ruffle B2B marketing feathers for an equally feather-ruffling product.”

Case in point: His team recently produced a (mostly) SFW commercial that promises to “uncover social media insights” by repositioning a local green space as a nudist park.

Lesson 1: Comedy begins with empathy.

Since I last saw him, Ali’s had stints as the brand creative director for Potbelly’s and now Hootsuite. At both places, he’s brought his sometimes wry, sometimes absurdist humor into play.

I ask him to spill his secrets. What can I tell our readers that will make them funnier marketers?

His answer is no joke: If you want to successfully use humor in marketing, start by building trust and practicing empathy. He gives me this example:

Say you’ve got an idea for a hilarious new ad campaign, but you keep hearing that the stakeholders “don’t want to have fun.” (Cyndi Lauper weeps.)

Ali asks, “Is it that, or is it that they’re kind of worried that they’re going to spend money on this,” and if it flops, they’ll be reprimanded — or worse?

“That’s a very human emotion. So if we go into these conversations with, ‘Listen, I hear this might be a little outside of your norm,’” you’re immediately showing empathy, even if the person hasn’t voiced their fears.

Lesson 2: Data can make you funnier.

“Data helps inform and persuade and build that trust,” Ali says. He’s “definitely gotten a CEO who’s shifted in their chair a little bit” during a pitch, so he knows something about persuading the risk-averse.

When you’re asking stakeholders to work outside their comfort zones, you “oftentimes need the data to show to them that this is actually what surveyed people want.” Ali points me to Hootsuite’s 2024 social media consumer report: 55% of the 6000+ respondents enjoy brand content that “makes me laugh.”

Image Source

A practical tip ties this all together: Ali will sometimes shoot a funny version and a straighter version of an ad, and test both. Building trust means showing “that you’re able to communicate the needs of the business in a way your audience cares about.”

Lesson 3: Use the peanut butter method.

“Everyone hates advertising, but they're okay being sold to,” Ali says.

It’s like using peanut butter to sneak your dog a pill. “If people are willing to be sold to, pitch the pill in something yummy. People will watch it.” (Let’s ignore for a moment that we are all the hapless dogs in this analogy.)

“I often think that the best ads are ones we can't measure, because they're shared in a group chat with friends.” I sincerely hope nobody is working on a pixel that can track my group chats, but it’s true that if somebody shares an ad, it’s because it’s both funny and emotionally resonant.

Maybe you see a funny ad for diapers. Your sister’s just had a baby, and you share the ad in the family group chat. “All of a sudden, there’s a bond formed through this piece of advertising.” And it goes beyond “here, buy this thing,” Ali says.

Without that (hopefully imaginary) group-chat tracking pixel, traditional marketing metrics won’t necessarily be of much use.

“But what did you solve for the customer?” Ali asks. “Those are the real results.” The more we can focus on that, “the better we’ll be as marketers.”

Lingering Questions

Each person we interview gives us a question for our next master of marketing. Last week, Wistia CEO Chris Savage asked:

What’s something you’re doing that’s working so well, you’re afraid to tell others about it?

Ali: I have to say that the creative brand team at Hootsuite is working so well that it‘s like a secret. Just to watch the collaboration and the teamwork that occurs here — it’s something I’ve never experienced before.

And Ali’s question for our next master in marketing:

What advice would you give yourself when you were first starting out?

Come back next Monday for the answer!

On August 20, 2024, OpenAI enabled GPT-4o fine-tuning in the OpenAI playground and the OpenAI API. The much-awaited feature is free for fine-tuning 1 million daily tokens until September 23, 2024.

In this article, I will show you how to fine-tune the OpenAI GPT-4o model for text classification and summarization tasks.

It is important to note that in my previous articles I have already demonstrated results obtained for zero-shot text classification and zero-shot text summarization using default GPT-4o model. In this article, you will see that fine-tuning a GPT-4o model improves text classification and text summarization performance significantly.

So, let's begin without an ado.

The following script installs the Python libraries you need to run codes in this article.

!pip install openai
!pip install rouge-score
!pip install --upgrade openpyxl
!pip install pandas openpyxl

The script below imports the required libraries into your Python application.

import os
import json
import time
import pandas as pd
from rouge_score import rouge_scorer
from sklearn.metrics import accuracy_score
from openai import OpenAI

In a previous article, I explained the process of fine-tuning GPT-4o mini and GPT-3.5 turbo models for zero-shot text classification.

The process remains the same for fine-tuning GPT-4o.
We will first import the text classification dataset, which in this article is the Twitter US Airline Sentiment Dataset.

The following script imports the dataset.

dataset = pd.read_csv(r"D:\Datasets\Tweets.csv")
dataset.head()

Output:

Next, we will write the preprocess_data() function, which takes in a dataset, the start index n, and the number of records as parameters. It then divides the dataset by sentiment category and returns the number of records beginning at the specified index. This approach ensures we have an equal number of records for each sentiment category.

We will fetch 600 records (200 positive, negative, and neutral) for training and 99 records (33 for each category) for testing. You can use more number of records for fine-tuning if you want.

def preprocess_data(dataset, n, records):

    # Remove rows where 'airline_sentiment' or 'text' are NaN
    dataset = dataset.dropna(subset=['airline_sentiment', 'text'])

    # Remove rows where 'airline_sentiment' or 'text' are empty strings
    dataset = dataset[(dataset['airline_sentiment'].str.strip() != '') & (dataset['text'].str.strip() != '')]

    # Filter the DataFrame for each sentiment
    neutral_df = dataset[dataset['airline_sentiment'] == 'neutral']
    positive_df = dataset[dataset['airline_sentiment'] == 'positive']
    negative_df = dataset[dataset['airline_sentiment'] == 'negative']

    # Select records from Nth index
    neutral_sample = neutral_df[n: n +records]
    positive_sample = positive_df[n: n +records]
    negative_sample = negative_df[n: n +records]

    # Concatenate the samples into one DataFrame
    dataset = pd.concat([neutral_sample, positive_sample, negative_sample])

    # Reset index if needed
    dataset.reset_index(drop=True, inplace=True)

    dataset = dataset[["text", "airline_sentiment"]]

    return dataset

The following script creates training and test sets.

training_data = preprocess_data(dataset, 0, 200)
print("Training data value counts:\n", training_data["airline_sentiment"].value_counts())
print("===========================")
test_data = preprocess_data(dataset, 600, 33)
print("Test data value counts:\n", test_data["airline_sentiment"].value_counts())

Output:

Next, we convert our dataset into the JSON format required to fine-tune OpenAI models.

# JSON file path
json_file_path = r"D:\Datasets\airline_sentiments.json"

# Function to create the JSON structure for each row
def create_json_structure(row):
    return {
        "messages": [
            {"role": "system", "content": "You are a Twitter sentiment analysis expert who can predict sentiment expressed in the tweets about an airline. You select sentiment value from positive, negative, or neutral."},
            {"role": "user", "content": row['text']},
            {"role": "assistant", "content": row['airline_sentiment']}
        ]
    }

# Convert DataFrame to JSON structures
json_structures = training_data.apply(create_json_structure, axis=1).tolist()

# Write JSON structures to file, each on a new line
with open(json_file_path, 'w') as f:
    for json_structure in json_structures:
        f.write(json.dumps(json_structure) + '\n')

print(f"Data has been written to {json_file_path}")

To fine-tune the OpenAI model, you need to upload training files to the OpenAI server. To do so, create a client object of the OpenAI class and pass the JSON file to the files.create() method of the client object.

client = OpenAI(
    # This is the default and can be omitted
    api_key = os.environ.get('OPENAI_API_KEY'),
)


training_file = client.files.create(
  file=open(json_file_path, "rb"),
  purpose="fine-tune"
)

Finally, as shown in the script below, you can start fine-tuning using the client.fine_tuning.jobs.create() method. Here, you must pass GPT-4o model id gpt-4o-2024-08-06 to the model attribute.

fine_tuning_job_gpt4o = client.fine_tuning.jobs.create(
  training_file=training_file.id,
  model="gpt-4o-2024-08-06"
)

You can see fine-tuning events for your fine-tuning job using the following script:

# List up to 10 events from a fine-tuning job
print(client.fine_tuning.jobs.list_events(fine_tuning_job_id = fine_tuning_job_gpt4o.id,
                                    limit=10))

Once fine-tuning is completed, you will receive an email with the fine-tuned model ID. Alternatively, you can retrieve the fine-tuned model ID using the following script.

ft_model_id = client.fine_tuning.jobs.retrieve(fine_tuning_job_gpt4o.id).fine_tuned_model

Once you have the fine-tuned model ID, you can use it like any default OpenAI model. The following script defines the find_sentiment() function, which uses the fine-tuned model ID to predict the sentiments of the tweets in the test set and finally prints the overall fine-tuned model accuracy.

def find_sentiment(client, model, dataset):
    tweets_list = dataset["text"].tolist()

    all_sentiments = []


    i = 0


    while i < len(tweets_list):

        try:
            tweet = tweets_list[i]
            content = """What is the sentiment expressed in the following tweet about an airline?
            Select sentiment value from positive, negative, or neutral. Return only the sentiment value in small letters.
            tweet: {}""".format(tweet)

            response = client.chat.completions.create(
                model=model,
                temperature=0,
                max_tokens=10,
                messages=[
                    {"role": "user", "content": content}
                ]
            )

            sentiment_value = response.choices[0].message.content

            all_sentiments.append(sentiment_value)
            i += 1
            print(i, sentiment_value)

        except Exception as e:
            print("===================")
            print("Exception occurred:", e)

    accuracy = accuracy_score(all_sentiments, dataset["airline_sentiment"])
    print(f"Accuracy: {accuracy}")

find_sentiment(client,ft_model_id, test_data)

Output:

The above output shows that the fine-tuned model achieved an accuracy of 92.92%, significantly better than the accuracy achieved via the default GPT-4o model in a previous article.

In the next section, you will see how to fine-tune GPT-4o for text summarization.

We will use the News Articles with Summary dataset to fine-tune the GPT-4o model.

The following script imports the dataset.

dataset = pd.read_excel(r"D:\Datasets\dataset.xlsx")
dataset = dataset.sample(frac=1)
dataset['summary_length'] = dataset['human_summary'].apply(len)
average_length = dataset['summary_length'].mean()
print(f"Average length of summaries: {average_length:.2f} characters")
print(dataset.shape)
dataset.head()

Output:

The rest of the process remains the same as text classification. We will filter a subset of data for fine-tuning (in this case, records 101 to 200) and convert the dataset into OpenAI-compliant JSON format.

selected_data = dataset.iloc[101:201]

# Function to create the JSON structure for each row
def create_json_structure(row):
    return {
        "messages": [
            {"role": "system", "content": "You are analyzing news articles. Use the provided content to generate a concise summary."},
            {"role": "user", "content": row['content']},
            {"role": "assistant", "content": row['human_summary']}
        ]
    }

# Convert selected DataFrame rows to JSON structures
json_structures = selected_data.apply(create_json_structure, axis=1).tolist()

# JSON file path
json_file_path = r"D:\Datasets\news_summaries.json"

# Write JSON structures to file, each on a new line
with open(json_file_path, 'w') as f:
    for json_structure in json_structures:
        f.write(json.dumps(json_structure) + '\n')

print(f"Data has been written to {json_file_path}")

Next, upload the training file to OpenAI servers.

training_file = client.files.create(
  file=open(json_file_path, "rb"),
  purpose="fine-tune"
)

Finally, you can start fine-tuning using the following script.

fine_tuning_job_gpt4o_ts = client.fine_tuning.jobs.create(
  training_file=training_file.id,
  model="gpt-4o-2024-08-06"
)

Once the model is fine-tuned, retrieve the model ID using the following script.

ft_model_id = client.fine_tuning.jobs.retrieve(fine_tuning_job_gpt4o_ts.id).fine_tuned_model

We will use the ROUGE scores to evaluate the text summarization performance of the fine-tuned model. The following script defines the calculate_rouge() function that allows you to calculate ROUGE1, ROUGE2, and ROUGEL scores.

# Function to calculate ROUGE scores
def calculate_rouge(reference, candidate):
    scorer = rouge_scorer.RougeScorer(['rouge1', 'rouge2', 'rougeL'], use_stemmer=True)
    scores = scorer.score(reference, candidate)
    return {key: value.fmeasure for key, value in scores.items()}

Finally, the following script demonstrates how we generate the summaries of the first 20 articles in our dataset using the fine-tuned model.

%%time

results = []

i = 0

for _, row in dataset[:20].iterrows():
    article = row['content']
    human_summary = row['human_summary']

    i = i + 1
    print(f"Summarizing article {i}.")

    prompt = f"Summarize the following article in 1150 characters. The summary should look like human created:\n\n{article}\n\nSummary:"

    response = client.chat.completions.create(
        model= ft_model_id,
        messages=[{"role": "user", "content": prompt}],
        max_tokens=1150,
        temperature=0.7
    )
    generated_summary = response.choices[0].message.content
    rouge_scores = calculate_rouge(human_summary, generated_summary)

    results.append({
    'article_id': row.id,
    'generated_summary': generated_summary,
    'rouge1': rouge_scores['rouge1'],
    'rouge2': rouge_scores['rouge2'],
    'rougeL': rouge_scores['rougeL']
    })

The following script prints average ROUGE scores.

results_df = pd.DataFrame(results)
mean_values = results_df[["rouge1", "rouge2", "rougeL"]].mean()
print(mean_values)

Output:

rouge1    0.579758
rouge2    0.417515
rougeL    0.431266
dtype: float64

The above script shows that the fine-tuned GPT-4o model achieves significantly higher ROUGE scores than the default GPT-4o model.

Fine-tuning can significantly improve a model's performance on a specific task. This article explains how to fine-tune the OpenAI GPT-4o model for text classification and text summarization. The results show that the fine-tuned GPT-4o model significantly outperforms the default GPT-4o model on both tasks.

Giant kudos to Scott Jehl on releasing his new Web Components De-Mystified online course! Eight full hours of training from one of the best in the business.

Quick Hit #12 originally published on CSS-Tricks, which is part of the DigitalOcean family. You should get the newsletter.

Eric gifting us with his research on all the various things that anchors (not links) do when they are in :focus.

Turns out, there’s a lot!

That’s an understatement! This is an incredible amount of work, even if Eric calls it “dry as a toast sandwich.” Boring ain’t always a bad thing. Let me simply drop in a pen that Dave put together pulling all of Eric’s findings into a table organized to compare the different behaviors between operating systems — and additional tables for each specific platform — because I think it helps frame Eric’s points.

That really is a lot! But why on Earth go through the trouble of documenting all of this?

All of the previously documented behavior needs to be built in JavaScript, since we need to go the synthetic link route. It also means that it is code we need to set aside time and resources to maintain.

That also assumes that is even possible to recreate every expected feature in JavaScript, which is not true. It also leaves out the mental gymnastics required to make a business case for prioritizing engineering efforts to re-make each feature.

There’s the rub! These are the behaviors you’re gonna need to mimic and maintain if veering away from semantic, native web elements. So what Eric is generously providing is perhaps an ultimate argument against adopting frameworks — or rolling some custom system — that purposely abstract the accessible parts of the web, often in favor of DX.

As with anything, there’s more than meets the eye to all this. Eric’s got an exhaustive list at the end there that calls out all the various limitations of his research. Most of those notes sound to me like there are many, many other platforms, edge cases, user agent variations, assistive technologies, and considerations that could also be taken into account, meaning we could be responsible for a much longer list of behaviors than what’s already there.

And yes, this sweatshirt is incredible. Indeed.

Basic keyboard shortcut support for focused links originally published on CSS-Tricks, which is part of the DigitalOcean family. You should get the newsletter.

In the world of search engine marketing (SEM), I’m seeing a growing number of marketers turn to PPC campaigns. That’s how Google has significantly grown its advertising revenue every year.

In 2023, 77.8% of Google’s revenue came from advertising on Google properties and YouTube.

Why? Well-run pay-per-click or PPC campaigns help you generate leads by nearly guaranteeing ad placement in search engine result pages.

And if your ads tool is tightly integrated with your CRM, you can use insights from your ad data to nurture leads across their buying journey.

In this post, I’ll walk you through what makes a successful campaign and the best practices for optimizing your PPC campaigns.

Table of Contents

• What is a PPC campaign?
• How to Build a PPC Campaign Strategy
• What is PPC management?
• PPC Campaign Management Template
• How to Manage a PPC Campaign
• Effective PPC Campaign Management Strategies
• How to Optimize Your PPC Campaign
• PPC Campaign Management by Platform

So, how can you win a PPC campaign? Success starts by using research, data, and insights to create an effective strategy.

If you jump into a PPC campaign without an effective strategy, it’s easy to waste money, attract the wrong leads, and set yourself up for frustration without much to show for it.

Here are a few ways I’ve seen marketers go wrong with PPC campaign management:

• Building one basic campaign without using Google Ads’ Ad Groups tool.
• Coming up with keywords by relying on gut feeling instead of doing research.
• Not adding “negative keywords” or monitoring campaigns to avoid wasting budget.
• Using unengaging landing pages or a homepage that generates no leads for the campaign.
• Creating campaigns, setting budget caps, and going live without informing internal or external stakeholders.

Building an excellent campaign structure can help you avoid these obstacles. Luckily, building an excellent PPC campaign structure can help you avoid these obstacles.

PPC Campaign Management Template

To help you succeed with your next campaign, we’ve created a free PPC campaign management template.

What I love about this template is that it allows you and your clients to set up a full-funnel campaign structure that follows PPC best practices.

Once you do that, you’ll be better positioned to maximize the return on your PPC investment. The template is broken into two sections: Ads Planner and Ads Results.

Ads Planner Template

In this section of the PPC template, you’ll record your ad campaign information.

The first three columns contain your campaign name, keywords, and negative keywords (keywords you don’t want ads shown for).

Enter your ad variations in the next section. There’s room for multiple headlines, descriptions, and URL paths to help you keep track of your running ads.

Download this Template

Ads Results Template

This part of the PPC template allows you to track relevant campaign metrics. Use it to record your total ad cost and analyze the performance of impressions, conversions, cost-per-click information, and more.

Image Source

Now that you’re familiar with the template’s components, let’s look at managing your PPC campaign.

How to Manage a PPC Campaign

Image Source

I’m going to show you how to use the PPC template in this blog post — so download it now and follow along with me.This template acts like a checklist that can help you manage the steps along the way.

Pro tip: Here’s what you need to know to make using this template easy:

• You’ll want to clear out the example data in the template. That includes placeholder keywords, campaign and ad group names, ads, and final URLs.
• Be careful not to alter the variables in columns F, J, and N of the ads planner. They contain the number of characters required by Google Ads for headlines, descriptions, URL paths, and Final URLs.

Now, let’s dive into PPC campaign management.

Step 1: Choose your PPC campaign management tools and software.

There are several platforms for managing your PPC campaign. I recommend starting with one platform and mastering it to keep your costs low in the initial stages of PPC planning.

To simplify managing your campaign, you can natively manage it in the platform you’re already using for running the ads, rather than paying for an external campaign management tool.

However, as you expand your strategy to include more sites, I’ve found that I’ve needed the features of a more robust PPC campaign management software.

These more advanced tools allow you to keep track of each platform, budget, and set of ad creatives in one place.

Pro tip: Here are some of my favorite tools for the job:

• Marin Software: AI-powered management and analytics to monitor and optimize cross-channel campaigns.
• WordStream Advisor: Analyzes Google and Facebook ad performance to improve reach, results, and ROI.
• SpyFu: Analyzes your competitor’s campaigns to help you build a well-rounded strategy.

• Optmyzr: Offers one-click optimizations, custom reporting, and rule-based automations for managing PPC campaigns across multiple platforms.

Step 2: Understand PPC campaign structure.

Don’t fall into the trap of setting up an account, creating an ad, directing the ad to a website’s home page, picking some keywords, and hitting go.

Running a successful PPC campaign is all about organizing your campaigns and ad groups effectively, choosing the right keywords, and targeting your audience with precision.

You’ll want to manage your budget wisely, craft relevant ad copy, and continuously optimize based on performance data.

And remember, PPC isn’t a one-time setup; it’s an ongoing process where staying updated and making tweaks will keep your campaigns driving results.

Chances are, the account you’re managing will have more than one PPC campaign. Your campaigns may contain several ad groups, each based on a set of related keyword groupings.

Each ad group may contain multiple ads with similar targeted keywords.

Managing multiple campaigns allows you to set daily budget caps, ad schedules, and select geo-targeted regions at the campaign level. If you're bidding on generic and branded keywords, you’ll want to put these in separate campaigns.

The parameters around these two types of keywords will likely be different.

As you’ll see below, your template reflects these best practices, providing cells for several campaigns, ad groups, and ad variations within those ad groups.

Download this Template Free

Step 3: Identify your landing pages.

A landing page is the web page where your PPC traffic goes after they click on your paid ad.

Don’t drive PPC visitors to your home page or blog, hoping they’ll see and fill out a lead generation form. That’s the role of organic search. Instead, drive your audience to a landing page with messaging tailored to your campaign goal.

It should contain a form that collects their contact information in exchange for an offer or asset. Be sure to add a tracking token or UTM parameter to the page so you can identify your lead sources.

Pro tip: Keep in mind that the final URL within an ad group will be the same regardless of the keyword or ad. You can create another ad group if you want to drive a keyword to a different landing page.

If you want to get even more specific, create another campaign for that keyword.

Step 4: Build your keyword strategy.

Always select keywords that are relevant to your landing page and offer. Relevant keywords are important because they increase the chance of the right prospect seeing your ad, clicking on it, and visiting the landing page.

It’s tempting to want to rank for multiple keywords, but you shouldn’t try to do this using a single landing page. Why? An effective landing page has focused messaging that addresses only a few targeted keywords.

If a keyword is irrelevant to your landing page messaging, you’ll end up wasting your ad budget, as visitors are much less likely to convert.

It’s better to create another offer and landing page that specifically addresses those additional keywords with relevant messaging.

Pro tip: To understand search volumes and costs around keywords you want, use tools like the Google AdWords Tool or the HubSpot keywords tool.

If it’s your first time managing a PPC campaign, I recommend checking out this article on how to design a keyword strategy. For Google Ads, you’ll want to learn more about keyword quality scores, too.

Step 5: Create your ads.

This is the fun part.

Most PPC campaign management software allows you to create one or more ads for each ad group (hence the “group” terminology).

After creating multiple ads, these platforms display the ad variations to your audience to test which version leads to a higher clickthrough rate (CTR). This is called A/B testing.

While running A/B tests is optional, I highly recommend taking advantage of it. I’ve seen time and again that A/B testing insights significantly improve campaign performance and ROI.

Pro tip: One of the most effective A/B tests is testing variations of your ad’s call to action (CTA), the phrase that encourages your prospect to take a specific action, like downloading a resource or signing up for more information.

If you need more detail, read how the HubSpot team A/B tests our own CTAs.

Writing Your Headline

When writing your ad, give special attention to your headline. The headline has the greatest influence on an ad’s CTR. Be sure to include a keyword in your headline to draw a user's attention.

An even better practice is using dynamic keyword insertion to make your ads even more relevant to people searching for what you offer.

Keep in mind that you have 30 characters for each ad headline, 30 characters for the display URL (the URL that's displayed in the ad, not to be confused with the final URL), and 90 characters for each line of copy.

If you're using this template, you can easily keep track of these values.

Download this Template Free

Choosing Your Display URL

Finally, there’s the tricky matter of the display URL. You‘re only allowed 30 characters here, but it’s unlikely that your final URL, the actual URL for your landing page, will be that short, especially after adding UTM parameters.

For this reason, search engines like Google allow you to create a display URL, which may not be an actual URL on your website. However, the domain in your display URL must be the same as the domain in your final URL.

This allows users to know they’re in the right place when they click your ad and visit your landing page.

Step 6: Share the completed template with stakeholders.

Your completed template needs to align with stakeholders’ expectations and the elements of a productive PPC campaign.

If you’re a PPC campaign stakeholder, this template will help you with two things.

First, the ads planner template gives you a bird‘s-eye view of what the person running the ads is doing. Second, the ad results template shows how much you’re spending on PPC.

With this, you can reallocate and swiftly modify your budget as you respond to changes in the marketplace.

Effective PPC Campaign Management Strategies

When I first started managing PPC campaigns, I quickly realized that there’s a lot more to it than just setting up ads and hoping for the best. You need to fine-tune every element to get the best results.

Over the years, I’ve honed my approach, and I’m excited to share my top recommendations for effective PPC campaign management strategies that can drive serious revenue, results, and ROI.

Set Clear Campaign Objectives and Goals

Clear, measurable objectives ensure every dollar you spend is working toward driving tangible business outcomes, like boosting sales or generating qualified leads.

When your goals are laser-focused, you’re setting yourself up to maximize ROI by putting your efforts into what truly matters for revenue growth.

Nail Your Audience Targeting

By honing in on precise audience segments, you can run hyper-targeted ads that minimize wasted spend and maximize conversions.

When you reach the right people at the right time, your traffic quality improves, leading to more conversions and better ROI.

Master Keyword Research and Selection

Keywords are the backbone of any PPC campaign. Choosing high-intent keywords ensures your ads appear when people are actively searching for your products or services.

This not only increases your chances of conversions but also helps drive more sales or leads, giving you the best bang for your PPC buck. If you need help, check out our video on using ChatSpot AI for keyword research.

Write Ad Copy That Converts

Developing compelling copy that speaks directly to your audience’s needs can skyrocket your clickthrough rates and conversions.

Effective messaging doesn’t just attract clicks — it persuades potential customers to take action, which means more revenue and better ROI.

Optimize Your Landing Pages

When your landing pages are relevant and persuasive, they guide visitors smoothly toward conversion, lowering bounce rates and increasing sales or leads — ultimately giving you a bigger return on your ad spend.

Manage Your Budget Wisely

Allocating your spend to the highest-performing campaigns minimizes waste and maximizes ROI. When you manage your budget efficiently, you can scale successful campaigns, driving more revenue over time.

Continuous Optimization

PPC isn’t a set-it-and-forget-it game.

Ongoing optimization is the secret to getting better results over time. Regularly testing and tweaking your targeting, bids, and ad creatives help you refine your campaigns, driving more results and revenue as your efforts compound.

Analyze Competitors Regularly

By keeping tabs on what others in your industry are doing, you can spot gaps in the market and seize opportunities they’re missing. This intel helps you outperform competitors, capture more market share, and drive higher revenue and ROI.

Scale Your Campaigns Strategically

Once you’ve found a winning formula, it’s time to scale.

Expanding your reach and increasing ad spend on high-performing campaigns can amplify your results without compromising performance. Strategic scaling leads to substantial revenue growth and maximizes ROI.

How to Optimize Your PPC Campaign

PPC campaign management isn’t a one-time thing. You’ll need to adjust your methods continually for optimized results. Keep these factors in mind to maximize the performance of your PPC campaigns.

Location

Geographic targeting, or geotargeting, allows you to focus your ads on specific locations. This can be as broad as an entire country or as specific as a single city or even a particular ZIP code.

Track the performance of your ads by location to identify the most high-performing areas where you may need to adjust your strategy.

You might reallocate the budget to more profitable locations or test new geotargeted ad copy to improve performance.

For example, if you own a bike shop, the ad performance data may show that targeting urban and densely populated areas may be better than targeting rural areas where most folks residents rely on a car to get around.

Performance by Device

In my experience, conversion rates can vary significantly by device. Campaigns that are effective for desktop users may not perform so well with mobile users.

Consider targeting each group separately and note any differences in conversion rates, cost per click, or return on ad spend (ROAS) for each device type.

If a campaign works better on mobile versus desktop, allocate funds towards your mobile efforts while you try a different campaign for desktop users. This way, you ensure you’re spending on campaigns that guarantee the highest ROI.

Evaluating Keywords

When running campaigns, every keyword you choose won’t prove useful. You’ll need to evaluate and remove the low performers. These could be keywords that:

• Are not converting.
• Are converting at a very high cost.
• Have a "below average" quality score rating.

Remove these keywords and stick with those that are performing well. This allows you to get maximum value from your ad budget.

Examine Keyword Bids

When bidding for keywords, you’ll want to determine how much you can pay for each conversion and still make a profit. To do this, I recommend using Google Ads tools to optimize your bids, like:

• Bid simulator: This allows you to see how higher or lower bids can affect your ad’s performance.
• First-page bid estimates: This shows how much you likely need to bid to get your ads on the first page of Google search results.

After determining the maximum you can pay for a keyword, these tools will help you make the most of your budget.

Performance by Day and Time

Campaign performance will fluctuate depending on the time of day or day of the week. So, observe your campaigns to see when they perform well.

If they aren’t performing during a specific timeframe, adjust your campaign so that you are only bidding on the most profitable times.

Now, let’s explore the platforms available to run your PPC campaigns.

PPC Campaign Management by Platform

Besides figuring out the ads that work best for your business, understanding where your audience spends most of their time online is key.

That’s why you need to familiarize yourself with the different platforms available for running your PPC campaigns.

Let’s look at some of the most popular ad platforms, like Google, Microsoft (Bing), Facebook, X (formerly Twitter), and YouTube.

Google PPC Campaign Management

Google has been the dominant player in the search engine space for over 20 years, and it still produces some of the most innovative ad experiences in the market.

Google PPC management offers several key advantages compared to other platforms, like larger reach and audience, advanced targeting options like intent-based targeting and remarketing, and native integration with other Google tools.

Here's a look at two of the most popular ways to serve ads on Google.

Image Source

Google Search Ads

One of the most popular types of Google Ads is the search ad. These ads appear at the top and bottom of the search results for specific keywords you bid on.

The typical goal for running Google search ad campaigns is to drive traffic to a specific webpage — like a landing or product page.

Google Display Ads

If you’ve ever visited a website that advertised on the banner, sidebar, or footer of a web page, you've probably seen a Google display ad. These ads are typically visual, featuring colorful graphics, videos, and occasionally audio.

Google display ads are helpful for retargeting customers who visit your website without taking your desired action.

Microsoft Ads (Formerly Bing Ads) PPC Campaign Management

Overall, Microsoft Ads works almost like Google Ads. Both offer similar ad formats, like text ads, responsive ads, and dynamic search ads.

However, Microsoft PPC management can be considered easier to use than Google PPC management for digital marketers starting out, with fewer field requirements and lower thresholds to get to advanced features.

There are a few more distinctions that’ll help you get the most out of your PPC campaign.

Image Source

Microsoft Ads Keyword Planner

The bulk of your PPC efforts will likely live in Google Ads. When you decide to bid on Microsoft Ads, avoid using your Google Ads keywords for your Microsoft Ads.

Google and Bing are different search engines, meaning the search volume for your Google keywords may not be the same in Bing.

Microsoft Bing’s keyword research and suggestion tool provides more accurate search volumes for your keywords.

So, while you can use your keyword list from Google, use this tool to verify whether you should bid on the same keywords versus other variants with more traffic.

Lower CPC

One upside of Microsoft Ads is the possibility of having lower CPC, which decreases your ad spend.

A WordStream test of both Google and Microsoft found that the average CPC of running ads on Microsoft was 33% lower.

The implication? Bidding on Microsoft Ads is less competitive than Google, meaning you may likely spend less when paying for Microsoft Ads keywords.

This may be especially true for specific industries, as you’ll see in this table:

Image Source

For a deeper dive into Microsoft Ads, check out this resource portal.

Facebook PPC Campaign Management

Facebook Ads Manager is a platform that connects 1.6 billion people to businesses on Facebook.

Where Facebook PPC management differs from Google PPC management is in its ability to target specific demographics and interests, which is great for brand awareness campaigns.

Image Source

Some of the most popular ads you can use for Facebook campaigns are:

Story Ads

With more users spending time on social media platforms’ story features, these ads are one way to reach your audience.

Like personal stories, you can share story ads as a video with a link. You can also use a series of photos to explain your ad and entice your audience to take a specific action.

Keep in mind: You can only post stories for 24 hours. These ads are best used for promotions like time-limited offers.

Playable Ads

Gamification is an innovative way to catch a lead’s attention. Facebook’s playable ads allow you to create a brief interactive version of a game or app so users can get a feel for your product.

You’ll want to keep the functionality simple, so you don’t deter potential customers. And, of course, make it fun.

Messenger Ads

If you’ve ever used Facebook’s messenger tool, you’ve probably seen an ad in your conversations. Messenger ads are useful because potential customers can decide to connect with your business directly from their messages.

If you have a customer service team that connects with people via chat, this is a great way to establish an instant connection. You can also send a lead to your site or a specific landing page from the ad.

To get a comprehensive deep dive into building Facebook ad campaigns, check out HubSpot’s Facebook Ads Training Course.

X (Formerly Twitter) PPC Campaign Management

X / Twitter Ads Manager makes it easy to plan your X ad while providing reporting on campaign performance.

Image Source

EMarketer reports that X is still a supportive environment for native brand content, with its ability to connect directly to consumers and interact in real time.

But while you may have an excellent chance of success with X, your ads need to be catchy enough to stop someone mid-scroll. Some ads you can include in your X ad campaigns are:

Promoted Ads

X’s Promoted Ads can display your message as a single Image Ad, a Video Ad, a Carousel Ad with up to six horizontally swipeable images or videos, or Text Ads for native advertising content.

This approach allows your business to convert users, or gain new followers. That can help build your brand’s awareness.

Vertical Video Ads

Grab the attention of your prospect with X’s full-screen vertical video Ads. Vertical video is the fastest-growing segment of X, capturing roughly 20% of viewers’ total time spent on the platform.

X reports that users are up to seven times more likely to follow, repost, like, and click the URLs of vertical video ads, as compared to the same ads on their home feed.

X Amplify/X Takeovers/X Live

For even more ways to align ads with premium video content, X offers ways to target ad display based on content categories, feed placement on Timelines or the Explore tab, or even livestream content.

Learn more about X Advertising for your business, and get to posting!

YouTube PPC Campaign Management

Consider adding YouTube PPC to your campaign, especially if you’re already using Google PPC management.

YouTube is one part of the Google Display Network with over 2.4 billion monthly users. This high number of users suggests that including YouTube in your ad campaign strategy makes sense.

If your business can create ads that’ll interest your audience and keep them from hitting “skip” you’re already winning.

Let’s look at some of the different YouTube ads.

Skippable In-Stream Ads

These are likely the ads you are most familiar with. These ads have a little button that says “skip.” Clicking the button allows you to start or continue viewing a video on YouTube.

Often, users have to wait five seconds before they can skip. Five seconds isn’t much time to convince someone to stick around, so ensure your ad’s hook can capture your audience’s attention.

However, you shouldn't worry if users skip your ad within the first five seconds. When a user skips your ad in the first 5 seconds, you won’t have to pay for such views or clicks.

Non-Skippable In-Stream Ads (Including Bumper Ads)

Since many people opt to skip ads on YouTube, advertisers have the option of creating non-skippable ads. If you think your creative is captivating enough to resonate with your target audience, this option may work.

Ensure you measure the results from your non-skippable ads to make the best use of your budget. If the results aren’t in your favor, revert to a skippable ad.

Video Discovery Ads (Formerly Known as In-display Ads)

Discovery ads are what users see in the search results. As the second largest search engine, people watch over 1 billion hours of YouTube videos daily. You’ll want those ads appearing in search results too.

These ads will include a thumbnail and a few lines of text as a description. Since many people prefer visuals over text, this is an opportunity to get your audience to view your video instead of reading a competitor’s text resource.

Start Your PPC Campaign Today

PPC campaign management is all about researching, budgeting, testing, reporting, building on what works, and optimizing for the best ROI.

You don‘t have to do it alone. With the right tools and instructions I’ve outlined in this guide, you’ll be able to implement a PPC campaign that yields results for your business.

Keep in mind that there are a few things that will help you make the most of your efforts:

• Clearly defined goals, so you can focus your efforts on reaching them.
• Focus on your audience, so you can reach the right people with the right message at the right time.

• Write good ad copy by testing your message and iterating on it to lockdown your conversions.
• Make data-informed decisions, so that you’re adapting your strategy based on numbers and not a gut check.
• Always be optimizing to meet a changing market.

By keeping these tips in mind, you’ll be well on your way to running PPC campaigns that don’t just drive traffic — they drive meaningful growth for your business.

Editor's note: This post was originally published in May 2019 and has been updated for comprehensiveness.

Developing a consistent brand starts with creating a brand style guide. These branding rule books help graphic designers, marketers, web developers, community managers, and even product packaging departments present a unified vision of the brand to the public.

The best brands stick in our brains because their presence is defined by the repetition of the same logo, fonts, colors, and images. Once we see them enough, they become instantly recognizable. All of this is possible when each member of your team adheres to a cohesive brand style guide.

So, what is a brand style guide? In this article, I'll go over the elements of a style guide and share some amazing examples of them in action to help inspire your next branding project or website redesign.

Table of Contents

• What are brand guidelines?
• The Elements of a Brand Style Guide
• Brand Style Guide Examples
• Branding Guidelines Tips

Picture the most recognizable brands you can think of.

Chances are, you've learned to recognize them due to one of the following reasons:

• There's a written or visual consistency across the messaging.
• The same brand colors are reflected across every asset.
• The language sounds familiar.
• It‘s all very organized and, while not rigid, it’s cohesive.

But before you sit down to create your branding guidelines, I'd recommend taking a step back and define your brand’s mission statement and buyer personas.

These strategic elements will help you dive into the tactical components of your brand style guide later.

Brand Guidelines Mission Statement

To me, your mission statement is the compass of your brand style guide. It‘s an action-oriented statement declaring your organization’s purpose.

This statement ensures that all your content is working toward the same goal and connecting with your audience. It can also guide your blog and paid content, ad copy, visual media, and slogan.

Pro tip: You can either include your mission statement within your style guide, create a separate document for reference, or distill your mission statement into a slogan that you can place at the head of your document.

Brand Guidelines Buyer Persona

A buyer persona is a fictional representation of your ideal customer. It includes details on your customer's job title, age, gender, and professional challenges — therefore stipulating for whom your brand publishes content.

Your buyer persona guides your blog content, ad copy, and visual media, which can attract valuable leads and customers to your business.

Pro tip: Download our free resource below on how to create your own style guide with brand guidelines templates to follow. Creating a consistent style guide isn't easy, but with these tools you can build an unforgettable one with ease.

The Elements of a Brand Style Guide

A brand style guide encompasses much more than just a logo (although that’s important, too). It visually encompasses everything your brand is about — down to your business' purpose.

Here are some key elements that I believe make or break a brand style guide.

Logo

Your logo might seem like the simplest aspect of your branding guidelines, but in reality, I‘d argue it’s one of the most complex and most important parts.

In your guide, you should:

• Include a visual of your logo.
• Explain the design details of your logo.
• Describe how your logo can be used by external and internal publishers.

You should also include wrong usages — i.e, you might advise against rotating the design or curving the font. That way, whether you or someone else is publishing information about your company, your logo looks consistent everywhere.

Pro tip: If your brand is well-known and many outlets publish information about you, you also might want to provide an entire document outlining acceptable use policies for your logo.

Color Palette

In my opinion, the color palette is probably one of the most distinctive and recognizable parts of a company’s branding guidelines.

It’s the group of colors your company uses to design its brand assets, guiding every piece of visual content created.

These color combinations often follow HEX or RGB color codes, and govern your logo, web design, printed ads, and event collateral.

Pro tip: A brand color palette should not only include your primary color, but also a wide variety of secondary, tertiary, and neutral colors. This will allow you to come up with more dynamic and varied designs in the content creation stage.

If you don’t define an array of options, you can run the risk of having your team create content with random secondary colors, which can look inconsistent.

Typography

Typography is a visual element of your brand style guide that goes beyond the font you use in your company logo. It supports your brand’s design down to the links and copy on your website — even your tagline.

I recommend specifying a primary and secondary font, with a mixture of serifs and font weights for different use cases.

Remember, the goal of your branding guidelines is to empower your people and external stakeholders to create consistent but varied collateral on behalf of your brand. You don’t want to limit them with a single font option.

For instance, HubSpot’s primary font is Lexend Deca (sans-serif), while our secondary font is Queens (serif). They’re both integrated in our very own Content Hub, and our design tool, Canva, where we can use them to create assets.

Pro tip: Don’t forget that typography also plays a major role in your website's user experience. You want to make sure it is visually appealing while also being accessible and easy to read.

Imagery and Iconography

You may be able to only include your logo, colors, and fonts in your guidelines.

However, if you’d like to create a stronger style guide, consider including approved imagery, pre-designed icons, and custom symbols for your company to use across your website and print collateral.

If your budget is smaller, you can recommend photographic styles (i.e candid versus staged, etcetera), and then direct content creators to your preferred stock photo provider (i.e. Shutterstock, Unsplash).

Alternatively, you can commission a company photoshoot at a studio and make the resulting photography available for creative use.

Pro tip: Symbols and icons can also be a great addition to your branding guidelines. As with photos, you can always find free icons online and recommend what to use versus what not to use (e.g., outlines only vs. full color).

You can also commission custom icons from a freelance graphic designer.

Brand Voice

If your company visuals are the flesh and bones of your style guide, I'm going to say your brand voice is the beating heart.

The importance of your brand voice can’t be overstated.

Maybe you want your company’s personality to be friendly and casual, or you may prefer a more distant and formal voice.

Either way, you want to make it easy for marketers, salespeople, and content creators on your team to know how to represent your brand online. This will ensure consistent messaging across all channels.

For example, if your content marketing strategy mainly focuses on blogs, you could use our Blog Topic Generator to streamline the content creation process and help maintain a consistent tone.

Besides helping you generate content ideas, the tool can also create and edit blog posts based on your selected tone of voice. 

You can also include a full editorial style guide. The job of an editorial style guide is to commit an editorial stylebook on how to phrase certain products, list topics the brand can and cannot write about, and other companies it can mention.

Your editorial style guide can guide your blog content, video scripts, website and landing page copy, PR talking points, and knowledge base articles.

As you can see, the purpose of the brand style guide is to form and maintain all of the various elements of a company that, when combined, spell out the entire brand as it's recognized.

Ready to get started? HubSpot's Brand Kit Generator can help you create all of these key branding and style guide elements with ease (and for free).

1. Medium

See the full brand guide here.

Image Source

What I like: Medium‘s simple brand style guide emphasizes usage of its logo, wordmark, and symbol. Medium’s logo is the brand's primary graphic element and was created to feel “confident, premium, timeless, and modern.”

2. Walmart

See the full brand guide here.

Image Source

What I like: The guide includes the brand‘s logo, photography, typography, illustrations, iconography, voice, editorial style, and more. Walmart’s color palette is so integral to its brand identity that its primary color is called “Walmart Blue.”

3. Asana

See the full brand guide here.

Image Source

What I like: Asana‘s simple style guide highlights its logo and color palette. It also explains how to properly use the brand’s assets.

4. Skype

See the full brand guide here.

Image Source

What I like: Everyone's favorite video chat platform also has a squeaky-clean style guide for its brand. Skype, now owned by Microsoft, focuses primarily on its product phrasing and logo placement.

5. Barre & Soul

See the full brand guide here.

Image Source

What I like: Barre & Soul's brand style guide includes variations of its logo, logo spacing, secondary logos, supporting imagery, and a five-color color palette.

6. Spotify

See the full brand guide here.

Image Source

What I like: Spotify‘s color palette includes three color codes, while the rest of the company’s branding guidelines focus on logo variation and album artwork. The style guide even allows you to download an icon version of its logo, making it easier to represent the company without manually recreating it.

7. Starbucks

See the full brand guide here.

Image Source

What I like: Starbucks' interactive brand style guide includes details about how to use its core elements such as the iconic Siren logo and green color palette. Plus, the guide features a visual spectrum of how their creative assets can be used across different channels.

8. Paris 2024

See the full brand guide here.

Image Source

What I like: Paris 2024's brand identity pays homage to the 1924 Olympic Games through Art Deco inspired design. Best of all, designers applied eco-branding methods to reduce the amount of ink and paper needed for physical materials as well as limit the power and data consumption on digital elements.

9. Urban Outfitters

See the full brand guide here.

Image Source

What I like: Photography, color, and even tone of voice appear in Urban Outfitters‘ California-inspired brand guidelines. Plus, the company isn’t shy to include information about its ideal consumer and what the brand believes in.

10. Love to Ride

See the full brand guide here.

Image Source

What I like: Love to Ride, a cycling company, is all about color variety in its visually pleasing style guide. The company's brand guidelines include nine color codes and tons of detail about its secondary logos and imagery.

11. Barbican

See the full brand guide here.

Image Source

What I like: Barbican, an art and learning center in the United Kingdom, sports a loud yet simple style guide focusing heavily on its logo and supporting typefaces.

12. I Love New York

See the full brand guide here.

Image Source

What I like: Despite its famously simple t-shirts, I Love New York has a brand style guide. The company begins its guidelines with a thorough explanation of its mission, vision, story, target audience, and tone of voice. Only then does the style guide delve into its logo positioning on various merchandise.

13. TikTok

See the full brand guide here.

Image Source

What I like: TikTok‘s style guide isn’t just a guide — it's an interactive brand book. First, it provides an in-depth look into how it brings its brand to life through design. Then, it gives an overview of its logo, co-branding, color, and typography.

14. University of the Arts Helsinki

See the full brand guide here.

Image Source

What I like: The style guide of the University of the Arts Helsinki is more of a creative branding album than a traditional marketing guide. It shows you dozens of contexts in which you‘d see this school’s provocative logo, including animations.

15. Ivy Lane Events

See the full brand guide here.

Image Source

What I like: Ivy Lane Events‘ bold style guide is reflective of the edgy events the company produces. In it, you’ll find a mood board with dark, romantic visuals inspired by “victorian gothic style and vintage book art.”

16. Western Athletic Conference

See the full brand guide here.

Image Source

What I like: The Western Athletic Conference's brand style guide includes extensive information about its history, mission, and vision. It also highlights its member universities and athletic championships and awards it is involved with.

17. Discord

See the full brand guide here.

Image Source

What I like: Discord‘s brand guide is as colorful and playful as the communities it serves. The brand’s motion elements are based on the dot, which represents the Discord user interacting with others in the communities it belongs to.

18. Netflix

See the full brand guide here.

Image Source

What I like: As far as its public brand assets are concerned, Netflix is focused primarily on the treatment of its logo. The company offers a simple set of rules governing the size, spacing, and placement of its famous capitalized typeface.

19. Scrimshaw Coffee

See the full brand guide here.

Image Source

What I like: Featuring a six-code color palette, this “laid back,” “cool,” and “eclectic” brand has a number of secondary logos it embraces in various situations.

20. NASA

See the full brand guide here.

Image Source

What I like: NASA‘s "Graphics Standards Manual" is as official and complex as you think it is. At 220 pages, the guide describes countless logo placements, color uses, and supporting designs. And yes, NASA’s space shuttles have their own branding rules.

21. New York City Transit Authority

See the full brand guide here.

Image Source

What I like: Like NASA, the NYCTA has its own Graphics Standards Manual, and it includes some fascinating typography rules for the numbers, arrows, and public transit symbols the average commuter takes for granted every day.

Branding Guidelines Tips

If you want to take your branding style guide to the next level, let HubSpot's Brand Kit Generator do some of the heavy lifting for you.

I'd also recommend following the best practices below, which the HubSpot Creative team has used to disseminate branding information to the rest of the HubSpot Marketing team.

This has not only made my job as a blogger easier, but also makes our branding feel well thought-out and cohesive.

1. Make your guidelines a branded document.

Whether you’re publishing your branding guidelines online or creating an internal presentation, consider making the guidelines themselves a branded document.

Ensure the published document follows your established brand voice, uses the symbols and imagery you’ve created, and employs the colors and typography that makes your brand feel like you.

Insights from HubSpot's Creative Team

When our Creative team rolled out a visual identity refresh for the HubSpot brand, we all received access to a branded playbook that summarized all the changes and described how we should represent HubSpot online moving forward.

Not only was I a huge fan of the refresh, but also of the way it was presented to our team in a branded document.

You can do the same, regardless of your budget. Our Creative team actually used a free tool, Google Slides — so it’s totally doable for a small or freelance brand!

2. Name your brand's colors.

You’ve already chosen your color palette — why not go as far as naming the colors?

Giving your colors unique names (aside from “blue” or “orange”) can help you tie the tactical elements of your branding into an overall theme or ethos.

Not to mention that it’s awesome to be able to refer to company colors by a unique name. Imagine if we called Solaris, HubSpot’s primary brand color, “HubSpot Orange” — that simply doesn’t have the same ring.

Insights from HubSpot's Creative Team

In our visual identity refresh, our Creative team brightened and intensified our color palette, then renamed the individual hues.

They wrote, “Every color, tint, and shade is based on central themes. [...] Whether it’s a subway line in Paris, or a flower-lined street in Japan, the secondary color names are a veritable tour of important cultural and geographical touchstones from HubSpotters all over the world.”

Think about what makes your brand unique, and why you chose the colors that you did. For instance, if you work at a law firm that specializes in car accident cases, you might choose red as one of the brand colors and call it “Stop Light.”

3. Create easy-to-use branded templates.

Alongside your branding guidelines should be templates to empower your team to easily design branded assets, even if they’re not designers.

Insights from HubSpot's Creative Team

At HubSpot, we keep all of our templates in our team’s Canva account. There, anyone (myself included) can edit pre-made designs for any number of use cases.

As a writer on the HubSpot blog, I have to create graphics to supplement the information I’m sharing.

The branded templates made by our Creative team have made my work a great deal easier, and I can imagine that it’s the same for our Social Media team.

Not everyone is a designer, but with templates, you can ensure your brand looks professional no matter who creates an asset.

4. Ensure your branding is optimized for all channels.

Your branding guidelines should include different specifications for different channels.

Or, alternatively, you should have assets and designs that can be adjusted for various channels and mediums. Not only for sizing purposes, but for accessibility purposes, too.

Image Source

For instance, if you primarily market your brand over Instagram and on your website, then your branding should have web accessible colors, as well as Instagram-friendly designs and sizes.

However, you don’t want to significantly change your branding from channel to channel. It should work relatively well no matter where you’re marketing your brand.

Build a Memorable Style Guide of Your Own

Once you build your unique brand style guide, customers will recognize your brand and associate it with all the visual cues you want them to.

I hope you were inspired by our list of amazing brand style guides and wish you luck in creating a timeless style of your own.

Editor's note: This post was originally published in January 2017 and has been updated for comprehensiveness.

Every time I work with a new brand, my first order of business is to conduct a competitive analysis. 

A competitive analysis report helps me understand the brand’s position in the market, map competitors’ strengths/weaknesses, and discover growth opportunities. 

In this article, I’ll break down the exact steps I follow to conduct competitor analysis and identify ways to one-up top brands in the market. 

We’ll cover:

• What is competitive analysis?
• What is competitive market research?
• Competitive Analysis in Marketing
• How To Conduct Competitive Analysis in 5 Steps
• How to Do a Competitive Analysis (the Extended Cut)
• Competitive Product Analysis
• Competitive Analysis Example
• Competitive Analysis Templates
• Competitive Analysis FAQs

Competitive analysis gives you a clearer picture of the market landscape to make informed decisions for your growth. 

That said, you have to remember that competitive analysis is an opportunity to learn from others. It isn’t:

• Copying successful competitors to the T.
• Trying to undercut others’ pricing.
• A one-and-done exercise.

Let’s look at how this exercise can help your business before breaking down my 5-step competitive analysis framework.

Four Reasons to Perform Competitive Analysis 

If you’re on the fence about investing time and effort in analyzing your competitors, know that it gives you a complete picture of the market and where you stand in it.

Here are four main reasons why I perform a competitive analysis exercise whenever working with a brand for the first time:

• Identify your differentiators. Think of competitor analysis as a chance to reflect on your own business and discover what sets you apart from the crowd. And if you’re only starting out, it helps you brainstorm the best opportunities to differentiate your business.
• Find competitors’ strengths. What are your competitors doing right to drive their growth? Analyzing the ins and outs of an industry leader will tell you what they did well to reach the top position in the market.
• Set benchmarks for success. A competitor analysis gives you a realistic idea of mapping your progress with success metrics. While every business has its own path to success, you can always look at a competitor’s trajectory to assess whether you’re on the right track.
• Get closer to your target audience. A good competitor analysis framework zooms in on your audience. It gives you a pulse of your customers by evaluating what they like, dislike, prefer, and complain about when reviewing competing brands.

The bottom line: Whether you’re starting a new business or revamping an existing one, a competitive analysis eliminates guesswork and gives you concrete information to build your business strategy.

What is competitive market research?

Competitive market research is a vital exercise that goes beyond merely comparing products or services. It involves an in-depth analysis of the market metrics that distinguish your offerings from those of your competitors.

A thorough market research doesn't just highlight these differences but leverages them, laying a solid foundation for a sales and marketing strategy that truly differentiates your business in a bustling market.

In the next section, we’ll explore the nuts and bolts of conducting a detailed competitive analysis tailored to your brand.

Essential Aspects to Cover in Competitive Analysis Research 

Before we walk through our step-by-step process for conducting competitor analysis, let’s look at the main aspects to include for every competitor:

• Overview. A summary of the company — where it’s located, target market, and target audience.
• Primary offering. A breakdown of what they sell and how they compare against your brand.
• Pricing strategy. A comparison of their pricing for different products with your pricing.
• Positioning. An analysis of their core messaging to see how they position themselves.
  Customer feedback: A curation of what customers have to say about the brand.

Now, it’s time to learn how to conduct a competitive analysis with an example to contextualize each step. 

Competitive Analysis in Marketing

Every brand can benefit from regular competitor analysis. By performing a competitor analysis, you'll be able to:

• Identify gaps in the market.
• Develop new products and services.
• Uncover market trends.
• Market and sell more effectively.

As you can see, learning any of these four components will lead your brand down the path of achievement.

Next, let's dive into some steps you can take to conduct a comprehensive competitive analysis.

How to Conduct Competitive Analysis in 5 Quick Steps

As a content marketer, I’ve performed a competitive analysis for several brands to improve their messaging, plan their marketing strategy, and explore new channels. Here are the five steps I follow to analyze competitors.

1. Identify and categorize all competitors.

The first step is a simple yet strategic one. You have to identify all possible competitors in your industry, even the lesser-known ones. The goal here is to be aware of all the players in the market instead of arbitrarily choosing to ignore a few.

As you find more and more competitors, categorize them into these buckets:

• Direct competitors. These brands offer the same product/service as you to the same target audience. People will often compare you to these brands when making a buying decision. For example, Arcade and Storylane are direct competitors in the demo automation category.
• Indirect competitors. These businesses solve the same problem but with a different solution. They present opportunities for you to expand your offering. For example, Scribe and Whatfix solve the problem of documentation + internal training, but in different ways.
• Legacy competitors. These are established companies operating in your industry for several years. They have a solid reputation in the market and are a trusted name among customers. For example, Ahrefs is a legacy competitor in the SEO industry.
• Emerging competitors. These are new players in the market with an innovative business model and unique value propositions that pose a threat to existing brands. For example, ChatGPT came in as a disruptor in the conversational AI space and outperformed several brands. 

Here’s a competitive matrix classifying brands in the community and housing space:

Image Source

Testing It Out

To help you understand each step clearly, we’ll use the example of Trello and create a competitor analysis report using these steps.

Here’s a table of the main competitors for Trello:

able of the main competitors for Trello:

2. Determine each competitor’s market position.

Once you know all your competitors, start analyzing their position in the market. This step will help you understand where you currently stand in terms of market share and customer satisfaction. It’ll also reveal the big guns in your industry — the leading competitors to prioritize in your analysis report.

Plus, visualizing the market landscape will tell you what’s missing in the current state. You can find gaps and opportunities for your brand to thrive even in a saturated market.

To map competitors’ market positions, create a graph with two factors: market presence (Y-axis) and customer satisfaction (X-axis). Then, place competitors in each of these quadrants:

• Niche. These are brands with a low market share but rank high on customer satisfaction. They’re likely targeting a specific segment of the audience and doing it well.
• Contenders. These brands rank low on customer satisfaction but have a good market presence. They might be new entrants with a strong sales and marketing strategy.
• Leaders. These brands own a big market share and have highly satisfied customers. They’re the dominant players with a solid reputation among your audience.
• High performers. These are another category of new entrants scoring high on customer satisfaction but with a low market share. They’re a good alternative for people not looking to buy from big brands.

This visualization will tell you exactly how crowded the market is. But it’ll also highlight ways to gain momentum and compete with existing brands.

Testing It Out

Here’s a market landscape grid by G2 documenting all of Trello’s competitors in the project management space. For a leading brand like Trello, the goal would be to look at top brands in two quadrants: “Leaders” and “High Performers.” 

Image Source

3. Extensively benchmark key competitors.

Step 2 will narrow down your focus from dozens of competitors to the few most important ones to target. Now, it’s time to examine each competitor thoroughly and prepare a benchmarking report.

Remember that this exercise isn’t meant to find shortcomings in every competitor. You have to objectively determine both the good and bad aspects of each brand.

Here are the core factors to consider when benchmarking competitors:

• Quality. Assess the quality of products/services for each competitor. You can compare product features to see what’s giving them an edge over you. You can also evaluate customer reviews to understand what users have to say about the quality of their offering.
• Price. Document the price points for every competitor to understand their pricing tactics. You can also interview their customers to find the value for money from users’ perspectives.
• Customer service. Check how they deliver support — through chat, phone, email, knowledge base, and more. You can also find customer ratings on different third-party platforms.
• Brand reputation. You should also compare each competitor’s reputation in the market to understand how people perceive the brand. Look out for anything critical people say about specific competitors.  
• Financial health. If possible, look for performance indicators to assess a brand's financial progress. You can find data on metrics like revenue growth and profit margins. 

This benchmarking exercise will involve a combination of primary and secondary research. Invest enough time in this step to ensure that your competitive analysis is completely airtight.

Check out this example of a competitor benchmarking report for workforce intelligence tools:

Image Source

Testing It Out

Here’s how I benchmarked Asana based on these criteria using the information I could find:

4. Deep dive into their marketing strategy.

While the first few steps will tell you what you can improve in your core product or service, you also need to find how competitors market their products.

You need to deep-dive into their marketing strategies to learn how they approach buyers. I analyze every marketing channel, then note my observations on how they speak to their audience and highlight their brand personality.

Here are a few key marketing channels to explore:

• Website. Analyze the website structure and copy to understand their positioning and brand voice.
• Email. Subscribe to emails to learn their cadence, copywriting style, content covered, and other aspects.
• Paid ads. Use tools like Ahrefs and Semrush to find if any competitor is running paid ads on search engines.
• Thought leadership. Follow a brand’s thought leadership efforts with content assets like podcasts, webinars, courses, and more.
• Digital PR. Explore whether a brand is investing in digital PR to build buzz around its business and analyze its strategy.
• Social media. See how actively brands use different social channels and what kind of content is working best for them.
• Partnerships. Analyze high-value partnerships to see if brands work closely with any companies and mutually benefit each other.

You can create a detailed document capturing every detail of a competitor’s marketing strategy. This will give you the right direction to plan your marketing efforts. 

5. Perform a SWOT analysis.

The final step in a competitive analysis exercise is creating a SWOT analysis matrix for each company. This means you‘ll take note of your competitor’s strengths, weaknesses, opportunities, and threats. Think of it as the final step to consolidate all your research and answer these questions:

• What is your competitor doing well?
• Where do they have an advantage over your brand?
• What is the weakest area for your competitor?
• Where does your brand have the advantage over your competitor?
• In what areas would you consider this competitor a threat?
• Are there opportunities in the market that your competitor has identified?

You can use tools like Miro to visualize this data. Once you visually present this data, you’ll get a clearer idea of where you can outgrow each competitor. 

Image Source

Testing It Out

Here’s a SWOT analysis matrix I created for Asana as a competitor of Trello:

To run a complete and effective competitive analysis, use these ten templates, which range in purpose from sales to marketing to product strategy.

Featured Resource: 10 Competitive Analysis Templates

Download Now

1. Determine who your competitors are.

First, you‘ll need to figure out who you’re competing with to compare the data accurately. What works in a business like yours may not work for your brand. Divide your “competitors” into two categories: direct and indirect.

Keep these brands on your radar since they could shift positions at any time and cross over into the direct competitor zone. Using our example, Stitch Fix could start a workout line, which would certainly change things for Fabletics.

This is also one of the reasons why you‘ll want to routinely run a competitor analysis. The market can and will shift at any time, and if you’re not constantly scoping it out, you won‘t be aware of these changes until it’s too late.

2. Determine what products your competitors offer.

You‘ll want to analyze your competitor’s complete product line and the quality of the products or services they're offering. You should also take note of their pricing and any discounts they're offering customers.

Some questions to consider include:

• Are they a low-cost or high-cost provider?
• Are they working mainly on volume sales or one-off purchases?
• What is their market share?
• What are the characteristics and needs of their ideal customers?
• Are they using different pricing strategies for online purchases versus brick-and-mortar?
• How does the company differentiate itself from its competitors?
• How do they distribute their products/services?

3. Research your competitors' sales tactics and results.

Running a sales analysis of your competitors can be a bit tricky.

You'll want to track down the answers to questions such as:

• What does the sales process look like?
• What channels are they selling through?
• Do they have multiple locations, and how does this give them an advantage?
• Are they expanding? Scaling down?
• Do they have partner reselling programs?
• What are their customers' reasons for not buying? For ending their relationship with the company?
• What are their revenues each year? What about total sales volume?
• Do they regularly discount their products or services?
• How involved is a salesperson in the process?

These helpful pieces of information will give you an idea of how competitive the sales process is and what information you need to prepare your sales reps with to compete during the final buy stage.

For publicly held companies, you can find annual reports online, but you'll have to do some sleuthing to find this info from privately owned businesses.

You could find some of this information by searching through your CRM and reaching out to those customers who mentioned they were considering your competitor.

Find out what made them choose your product or service over others out there.

When a competitor is identified, have your sales team dive deeper by asking why they are considering switching to your product. If you've already lost the deal, be sure to follow up with the prospect to determine why you lost to your competitor.

4. Take a look at your competitors' pricing, as well as any perks they offer.

There are a few major factors that go into correctly pricing your product — and one major one is understanding how much your competitors are charging for a similar product or service.

If you feel your product offers superior features compared to those of a competitor, you might consider making your product or service more expensive than industry standards.

However, if you do that, you'll want to ensure your sales reps are ready to explain why your product is worth the additional cost.

Alternatively, perhaps you feel there‘s a gap in your industry for affordable products. If that’s the case, you might aim to charge less than competitors and appeal to prospects who aren't looking to break the bank for a high-quality product.

Of course, other factors go into correctly pricing a product, but it‘s critical you stay on top of industry pricing to ensure you’re pricing your product in a way that feels reasonable to prospects.

Additionally, take a look at any perks your competitors offer and how you might match those perks to compete. For instance, perhaps your competitors offer a major referral discount or a month-long free trial version.

These perks could be the reason you‘re losing customers, so if it feels reasonable for your brand, consider where you might match those perks — or provide some unique perks of your own if competitors don’t offer any.

5. Ensure you're meeting competitive shipping costs.

Did you know expensive shipping is the number one reason for cart abandonment?

Nowadays, free shipping is a major perk that can attract consumers to choose one brand over another. If you work in an industry where shipping is a major factor — like e-commerce — you‘ll want to take a look at competitors’ shipping costs and ensure you're meeting (if not exceeding) those prices.

If most of your competitors offer free shipping, you‘ll want to look into the option for your own company. If free shipping isn’t a practical option for your business, consider how you might differentiate in other ways — including loyalty programs, holiday discounts, or giveaways on social media.

6. Analyze how your competitors market their products.

Analyzing your competitor's website is the fastest way to gauge their marketing efforts. This is a great way to see how accessible and engaging their assets are, and if you can, try experimenting with A/B testing your landing pages or website as well. Take note of any of the following items and copy down the specific URL for future reference:

• Do they have a blog?
• Are they creating whitepapers or e-books?
• Do they post videos or webinars?
• Do they have a podcast?
• Are they using static visual content such as infographics and cartoons?
• What about slide decks?
• Do they have a FAQs section?
• Are there featured articles?
• Do you see press releases?
• Do they have a media kit?
• What about case studies?
• Do they publish buying guides and data sheets?
• What online and offline advertising campaigns are they running?

7. Take note of your competition's content strategy.

Then, take a look at the quantity of these items. Do they have several hundred blog posts or a small handful? Are there five white papers and just one e-book?

Next, determine the frequency of these content assets. Are they publishing something new each week or once a month? How often does a new e-book or case study come out?

Chances are, if you come across a robust archive of content, your competitor has been publishing regularly. Depending on the topics they're discussing, this content may help you hone in on their lead-generating strategies.

From there, you should move on to evaluating the quality of their content. After all, if the quality is lacking, it won‘t matter how often they post since their target audience won’t find much value in it.

Choose a small handful of samples to review instead of tackling every single piece to make the process more manageable.

Your sampler should include content pieces covering a variety of topics, so you'll have a fairly complete picture of what your competitor shares with their target audience.

When analyzing your competitor's content, consider the following questions:

• How accurate is their content?
• Are spelling or grammar errors present?
• How in-depth does their content go? (Is it at the introductory level that just scratches the surface, or does it include more advanced topics with high-level ideas?)
• What tone do they use?
• Is the content structured for readability? (Are they using bullet points, bold headings, and numbered lists?)
• Is their content free and available to anyone, or do their readers need to opt in?
• Who is writing their content? (In-house team? One person? Multiple contributors?)
• Is there a visible byline or bio attached to their articles?

As you continue to scan the content, pay attention to the photos and imagery your competitors are using.

Do you quickly scroll past generic stock photos, or are you impressed by custom illustrations and images? If they're using stock photos, do they at least have overlays of text quotes or calls-to-action that are specific to their business?

If their photos are custom, are they sourced from outside graphic professionals, or do they appear to be done in-house?

When you have a solid understanding of your competitor‘s content marketing strategy, it’s time to find out if it's truly working for them.

8. Learn what technology stack your competitors use.

Understanding what types of technology your competitors use can be critical for helping your own company reduce friction and increase momentum within your organization.

For instance, perhaps you‘ve seen positive reviews about a competitor’s customer service — as you're conducting research, you learn the customer uses powerful customer service software you haven't been taking advantage of.

This information should arm you with the opportunity to outperform your competitors' processes.

To figure out which software your competitors use, type the company's URL into Built With, an effective tool for unveiling what technology your competitors' site runs on, along with third-party plugins ranging from analytics systems to CRMs.

Alternatively, you might consider looking at competitors' job listings, particularly for engineer or web developer roles. The job listing will likely mention which tools a candidate needs to be familiar with — a creative way to gain intel into the technology your competitors use.

9. Analyze the level of engagement on your competitor's content.

To gauge how engaging your competitor‘s content is to their readers, you’ll need to see how their target audience responds to what they're posting.

Check the average number of comments, shares, and likes on your competitor's content and find out if:

• Certain topics resonate better than others.
• The comments are negative, positive, or mixed.
• People are tweeting about specific topics more than others.
• Readers respond better to Facebook updates about certain content.
• Don't forget to note if your competitor categorizes their content using tags and if they have social media follow and share buttons attached to each piece of content.

10. Observe how they promote their marketing content.

From engagement, you‘ll move right along to your competitor’s content promotion strategy.

• Keyword density in the copy itself
• Image ALT text tags
• Use of internal linking

The following questions can also help you prioritize and focus on what to pay attention to:

• Which keywords are your competitors focusing on that you still haven't tapped into?
• What content of theirs is highly shared and linked to? How does your content compare?
• Which social media platforms are your target audience using?
• What other sites are linking back to your competitor's site but not yours?
• Who else is sharing what your competitors are publishing?
• Who is referring traffic to your competitor's site?
• For the keywords you want to focus on, what is the difficulty level? There are several free (and paid) tools that will give you a comprehensive evaluation of your competitor's search engine optimization.

11. Look at their social media presence, strategies, and go-to platforms.

The last area you‘ll want to evaluate when it comes to marketing is your competitor’s social media presence and engagement rates.

How does your competition drive engagement with their brand through social media? Do you see social sharing buttons with each article? Does your competitor have links to their social media channels in the header, footer, or somewhere else? Are these clearly visible? Do they use calls-to-action with these buttons?

If your competitors are using a social network that you may not be on, it's worth learning more about how that platform may be able to help your business, too.

To determine if a new social media platform is worth your time, check your competitor's engagement rates on those sites. First, visit the following sites to see if your competition has an account on these platforms:

• Facebook.
• Twitter.
• Instagram.
• Snapchat.
• LinkedIn.
• YouTube.
• Pinterest.

Then, take note of the following quantitative items from each platform:

• Number of fans/followers.
• Posting frequency and consistency.
• Content engagement. (Are users leaving comments or sharing their posts?)
• Content virality. (How many shares, repins, and retweets do their posts get?)

With the same critical eye you used to gauge your competition's content marketing strategy, take a fine-toothed comb to analyze their social media strategy.

What kind of content are they posting? Are they more focused on driving people to landing pages, resulting in new leads? Or are they posting visual content to promote engagement and brand awareness?

How much of this content is original? Do they share curated content from other sources? Are these sources regular contributors? What is the overall tone of the content?

How does your competition interact with its followers? How frequently do their followers interact with their content?

After you collect this data, generate an overall grade for the quality of your competitor's content. This will help you compare the rest of your competitors using a similar grading scale.

12. Perform a SWOT Analysis to learn their strengths, weaknesses, opportunities, and threats.

As you evaluate each component in your competitor analysis (business, sales, and marketing), get into the habit of performing a simplified SWOT analysis at the same time.

This means you‘ll take note of your competitor’s strengths, weaknesses, opportunities, and threats any time you assess an overall grade.

Some questions to get you started include:

• What is your competitor doing well? (Products, content marketing, social
• Where does your competitor have the advantage over your brand?
• What is the weakest area for your competitor?
• Where does your brand have the advantage over your competitor?
• What could they do better with?
• In what areas would you consider this competitor a threat?
• Are there opportunities in the market that your competitor has identified?

You‘ll be able to compare their weaknesses against your strengths and vice versa. By doing this, you can better position your company, and you’ll start to uncover areas for improvement within your own brand.

Competitive Product Analysis

Product analysis drills down to discover key differences and similarities in products that share the same general market.

If you have a competitor selling products in a similar market niche to your own — you want to ensure that you aren’t losing market share to the competition.

Leveraging the example above, we can drill down and discover some of the key differentiators in product offerings.

1. Assess your current product pricing.

The first step in any product analysis is to assess current pricing.

Nintendo offers three models of its Switch console: The smaller lite version is priced at $199, the standard version is $299, and the new OLED version is $349.

Sony, meanwhile, offers two versions of its PlayStation 5 console: The standard edition costs $499, and the digital version, which doesn’t include a disc drive, is $399.

2. Compare key features.

Next is a comparison of key features. In the case of our console example, this means comparing features like processing power, memory, and hard drive space.

3. Pinpoint differentiators.

With basic features compared, it’s time to dive deeper with differentiators. While a glance at the chart above seems to indicate that the PS5 is outperforming its competition, this data only tells part of the story.

Here’s why: The big selling point of the standard and OLED Switch models is that they can be played as either handheld consoles or docked with a base station connected to a TV. What’s more, this “switching” happens seamlessly, allowing players to play whenever, wherever.

The Playstation offering, meanwhile, has leaned into market-exclusive games that are only available on its system to help differentiate them from their competitors.

4. Identify market gaps.

The last step in a competitive product analysis is looking for gaps in the market that could help your company get ahead.

When it comes to the console market, one potential opportunity gaining traction is the delivery of games via cloud-based services rather than physical hardware.

Companies like Nvidia and Google have already made inroads in this space, and if they can overcome issues with bandwidth and latency, it could change the market at scale.

Competitive Analysis Example

How do you stack up against the competition? Where are you similar, and what sets you apart? This is the goal of competitive analysis.

By understanding where your brand and competitors overlap and diverge, you’re better positioned to make strategic decisions that can help grow your brand.

Of course, it’s one thing to understand the benefits of competitive analysis, and it’s another to actually carry out an analysis that yields actionable results. Don’t worry — we’ve got you covered with a quick example.

Sony vs. Nintendo: Not all fun and games.

Let’s take a look at popular gaming system companies Sony and Nintendo.

Sony’s newest offering — the Playstation 5 — recently hit the market but has been plagued by supply shortages.

Nintendo’s Switch console, meanwhile, has been around for several years but remains a consistent seller, especially among teens and children.

This scenario is familiar for many companies on both sides of the coin; some have introduced new products designed to compete with established market leaders, while others are looking to ensure that reliable sales don’t fall.

Using some of the steps listed above, here’s a quick competitive analysis example.

1. Determine who your competitors are.

In our example, it’s Sony vs Nintendo, but it’s also worth considering Microsoft’s Xbox, which occupies the same general market vertical.

This is critical for effective analysis; even if you’re focused on specific competitors and how they compare, it’s worth considering other similar market offerings.

2. Determine what products your competitors offer.

PlayStation offers two PS5 versions, digital and standard, at different price points, while Nintendo offers three versions of its console.

Both companies also sell peripherals — for example, Sony sells virtual reality (VR) add-ons, while Nintendo sells gaming peripherals such as steering wheels, tennis rackets, and differing controller configurations.

3. Research your competitors' sales tactics and results.

When it comes to sales tactics and marketing, Sony and Nintendo have very different approaches.

In part thanks to the recent semiconductor shortage, Sony has driven up demand via scarcity — very low volumes of PS5 consoles remain available. Nintendo, meanwhile, has adopted a broader approach by targeting families as its primary customer base.

This effort is bolstered by the Switch Lite product line, which is smaller and less expensive, making it a popular choice for children.

The numbers tell the tale: Through September 2021, Nintendo sold 14.3 million consoles, while Sony sold 7.8 million.

4. Take a look at your competitors' pricing, as well as any perks they offer.

Sony has the higher price point: Their standard PS5 sells for $499, while Nintendo’s most expensive offering comes in at $349. Both offer robust digital marketplaces and the ability to easily download new games or services.

Here, the key differentiators are flexibility and fidelity. The Switch is flexible — users can dock it with their television and play it like a standard console or pick it up and take it anywhere as a handheld gaming system.

The PS5, meanwhile, has superior graphics hardware and processing power for gamers who want the highest-fidelity experience.

5. Analyze how your competitors market their products.

If you compare the marketing efforts of Nintendo and Sony, the difference is immediately apparent: Sony’s ads feature realistic in-game footage and speak to the exclusive nature of their game titles.

The company has managed to secure deals with several high-profile game developers for exclusive access to new and existing IPs.

Nintendo, meanwhile, uses brightly lit ads showing happy families playing together or children using their smaller Switches while traveling.

6. Analyze the level of engagement on your competitor's content.

Engagement helps drive sales and encourage repeat purchases.

While there are several ways to measure engagement, social media is one of the most straightforward: In general, more followers equates to more engagement and greater market impact.

When it comes to our example, Sony enjoys a significant lead over Nintendo: While the official Playstation Facebook page has 38 million followers, Nintendo has just 5 million.

Competitive Analysis Templates

Competitive analysis is complex, especially when you’re assessing multiple companies and products simultaneously.

To help streamline the process, we’ve created 10 free templates that make it possible to see how you stack up against the competition — and what you can do to increase market share.

Let’s break down our SWOT analysis template. Here’s what it looks like:

Download Free Templates

Strengths. Identify your strengths. These may include specific pieces of intellectual property, products that are unique to the market, or a workforce that outperforms the competition.

Weaknesses. Here, it’s worth considering potential issues around pricing, leadership, staff turnover, and new competitors in the market.

Opportunities. This part of the SWOT analysis can focus on new market niches, evolving consumer preferences, or new technologies being developed by your company.

Threats. These might include new taxes or regulations on existing products or an increasing number of similar products in the same market space that could negatively affect your overall share.

Competitive Analysis: FAQs

What is a competitive analysis framework?

A competitive analysis framework is a structured approach used to evaluate potential competitors and understand their strengths, weaknesses, opportunities, and threats.

This framework serves as a guide for businesses to identify competitive advantages, understand market positioning, and inform strategic decisions.

Depending on the industry and the company's objectives, this framework might focus on areas like product features, market share, pricing, customer feedback, and more.

How do you do a good competitive analysis?

Conducting a thorough competitive analysis involves several steps:

• Identify competitors. Start by listing key competitors in your market, both direct and indirect.
• Evaluate their products/services. Analyze what they're selling and how it compares to your offerings.
• Analyze market position. Determine their market share, brand perception, and unique value proposition.
• Check their marketing strategies. Observe their advertising, content marketing, PR efforts, and online presence.
• Assess their financial health. If available, review financial statements, annual reports, or investor presentations.
• Gather customer feedback. Reviews, testimonials, and surveys can give insights into competitor strengths and weaknesses.
• Regularly review and update. The market evolves, and competitors change strategies, so it's vital to keep your analysis current.

What are the 5 parts of a competitive analysis?

The five key components of a competitive analysis include:

1. Company overview. A brief snapshot of the competitor, including its history, size, and mission.
2. Product/service analysis. An examination of their key products or services and how they compare to yours.
3. Marketing strategy. Insights into their promotional tactics, target audience, and unique selling propositions.
4. Operational analysis. An understanding of their supply chain, distribution, and customer service practices.
5. Strengths and weaknesses. A clear breakdown of where the competitor excels and where they might be vulnerable.

What are the 3 C's in a competitive analysis?

The 3 C's refer to a strategic model that considers three main factors when understanding the broader business environment:

1. Company. Understanding your own strengths, weaknesses, opportunities, and threats.
2. Customers. Knowing who your target audience is, what they value, and how they behave.
3. Competitors. Analyzing direct and indirect competitors to determine market dynamics and potential threats.

Is SWOT analysis a competitive analysis?

Yes, SWOT analysis is a type of competitive analysis. SWOT stands for Strengths, Weaknesses, Opportunities, and Threats. It is a strategic planning tool used to identify and analyze these four elements of a business.

While a SWOT analysis can be focused internally on a company‘s own attributes, when used as a competitive analysis tool, it assesses a competitor’s SWOT to understand where your business has advantages or might be vulnerable.

How does your business stack up?

Before you accurately compare your competition, you need to establish a baseline. This also helps when it comes time to perform a SWOT analysis.

Take an objective look at your business, sales, and marketing reports through the same metrics you use to evaluate your competition. Record this information just like you would with a competitor and use this as your baseline to compare across the board.

Editor's Note: This post was originally published prior to July 2018 but has been updated for comprehensiveness.

A gem from Chris Ferdinandi that details how to use custom events to hook into Web Components. More importantly, Chris dutifully explains why custom events are a better fit than, say, callback functions.

With a typical JavaScript library, you pass callbacks in as part of the instantiate process. […] Because Web Components self-instantiate, though, there’s no easy way to do that.

There’s a way to use callback functions, just not an “easy” way to go about it.

JavaScript provides developers with a way to emit custom events that developers can listen for with the Element.addEventListener() method.

We can use custom events to let developers hook into the code that we write and run more code in response to when things happen. They provide a really flexible way to extend the functionality of a library or code base.

Don’t miss the nugget about canceling custom events!

Callbacks on Web Components? originally published on CSS-Tricks, which is part of the DigitalOcean family. You should get the newsletter.

There’s no debate that when it comes to succeeding in digital content, mastering SEO is a must. When I first became a content writer, most of the advice I received was about keywords, backlinks, relevancy, and improving user experience.

However, as technology has advanced, SEO has changed. We’re entering a new era of search — the AI age — and, with it, a new type of digital content technique.

Enter: Generative Engine Optimization (GEO).

How we obtain information now has changed dramatically as a result of AI. More than ever, people rely on generative engines for information, making learning how to optimize for AI essential.

If we want our content to reach the right people, we need to adapt. There’s no need to throw out the SEO principles you’ve spent decades learning. GEO is an extension of these techniques — and it’s not as scary as it sounds.

Let’s take a look at what we know about GEO so far.

Table of Contents

• What is Generative Engine Optimization?
• GEO vs. SEO
• How does Generative Engine Optimization work?
• Tips for Navigating the Generative SEO Landscape
• Generative Engine Optimization FAQ
• What’s Next for GEO

What is Generative Engine Optimization?

Generative Engine Optimization (GEO) is a new technique for maximizing your digital content’s visibility. Instead of traditional SEO practices that focus on ranking on SERPs, GEO is about being used by generative AI engines.

Generative engines pull information directly from web content to deliver responses to user queries. They use large language models (LLMs) to make sense of the info they’ve scraped and provide coherent, relevant answers.

ChatGPT, Perplexity AI, and Google AI Search are all examples of generative engines.

Generative engines work by:

1.  Interpreting a user’s query.
2.  Leveraging personal data it may have on the user, such as preferences or conversation history.
3.  Searching to find relevant answers to the query
4.  Synthesizing information from these documents into a straightforward response

GEO vs. SEO

GEO and SEO are similar in many ways, as both find relevant, credible content to answer user inquiries. But they also have three key differences.

Focus

GEO focuses on making content discoverable to AI, while SEO is about improving SERP rank.

While SEO is generally related to Google and Bing, there are lots of generative engines, so it’ll be interesting to see if different engines use different qualities to determine their sources.

Emphasis

The primary techniques for SEO emphasize things like backlinks and keywords, while GEO techniques emphasize structure.

An AI bot’s job is easier when it can pull clear, concise snippets that are easily synthesized, so it makes sense why structure would be an emphasis.

Output

The key difference is the output of the engines.

GEO optimizes content for AI engines, which produce a summary as the output.

SEO, on the other hand, optimizes content for traditional search engines, which produce a ranked list of sources as the output.

How is GEO impacting SEO?

SEO experts worldwide have cited generative AI as the number one disruptor to SEO. So it’s worth understanding more about what experts suspect the impact will be.

I spoke with an expert on SEO, Nick Baird, to hear about his thoughts on GEO and how he thinks it will change SEO and digital marketing.

“First, top-of-funnel informational terms, which are the industry's bread and butter, are going to take a huge hit. Since answers will come right in the SERP, why would people click away to websites?” Baird notes.

Secondly, Baird says, informational searches will be tough to win. However, local SEO won't change because of generative AI.

“When people search locally, they‘re hoping to find a plumber to fix their toilet or a mechanic to fix their car. ChatGPT can’t help with that. Google will continue pointing people to local service businesses in this case,” Baird says.

His thoughts about TOFU terms are the most interesting to me. I see his point and, as a user, I’m definitely engaging less with sites when I’m looking for a short answer. I’m interested to see how high-ranking topics evolve.

How does Generative Engine Optimization work?

Learning to use GEO is simpler than it sounds and, in many ways, is overlapped with best SEO practices. AI tools respond well to clear, well-structured information that it can easily synthesize.

This means you should:

1. Ensure your content is easy to read and understand.
2. Incorporate credible sources, quotes, and statistics to enhance the content’s richness and authority.
3. Structure your writing to align with the patterns used by generative engines.

When I’m writing an article, I focus on clear headings, concise paragraphs, lists, and well-sourced information. The same is true for best practices when it comes to SEO — so don’t worry about reinventing the wheel.

Other ways to improve GEO are using AI-friendly structured data, focusing on user intent, using easy-to-read/conversational language, and using unique words to make the content stand out.

These best practices aren’t all that different from SEO best practices. And, as with SEO, there’s no indication that AI-created content is ranking poorly.

If your content is high quality, you shouldn’t experience any sort of penalty for leveraging AI content tools, like the ones offered by HubSpot.

Image Source

Tips for Navigating the Generative SEO Landscape

Succeeding at GEO isn’t too different from traditional SEO practices. Here are a few tips to keep in mind as you start writing content for generative AI.

Cite sources and use statistics.

I also asked Baird what digital marketers who are hoping to create AI-scannable content should do, and he recommended starting with strong EEAT (Experience, Expertise, Authoritativeness, and Trustworthiness).

These four factors are currently used by Google’s algorithm to determine how pieces rank in search.

“The more proof of a human author, rather than AI, the better. Domain authority becomes even more important as well,” Baird says.

While you can still use AI tools to write your content, it’s vital that it reads as useful, credible content. Incorporating EEAT is a great way to increase credibility, and that serves to improve your entire domain’s authority.

So what exactly does EEAT look like? I’ve been navigating the shift first hand for my HubSpot posts.

I start by looking at my own personal experience. Do I have lived experience in a subject matter? Can I include any personal anecdotes about when I’ve encountered the topic?

Beyond that, I try to showcase my authority on a subject, noting how long I’ve worked in the field.

Obviously, I am not an expert in every topic I want to cover. In that case, I do extensive research. I find original statistics with verified information. I talk to experts whose quotes I can showcase throughout my work.

These elements enhance the credibility and uniqueness of my content.

Optimize for readability.

Think of AI like a busy student frantically looking for information to use in a paper.

They don’t have time to decipher complicated sentences. They want information that’s clear and easy to understand from the get-go. Which brings me to my next tip: Use clear, concise language and scannable paragraphs.

I’ll be honest: Writing short, clear sentences may not always be my first instinct. When I’m discussing a complicated topic, my first draft often includes lengthy explanations. I then use Hemingway, an app designed to help you write clearly.

Hemingway lets me know which sentences are lengthy, confusing, or overly complicated. I can’t always get every sentence to green, but I make an effort to correct phrases marked as “very hard to read.”

Image Source

Focus on content quality.

In the past, having the right keywords was enough to win in search. Let’s take a classic example: recipe blogs.

I remember searching for a brownie recipe for a party and finding a promising, top-ranking article.

When I clicked on it, I had to scroll past huge chunks of text about what goes into a brownie, when brownies became popular, how the author's kids loved the brownies, so on and so forth.

Why? The author knew having the word “brownie recipe” as many times as possible would help her rank.

Well, that may no longer be the case. In today’s landscape, knowing that the recipe has been passed down from older generations and won an award in a local competition would be enough to show credibility.

TL;DR: Make sure your content is relevant to potential search engines, and avoid keyword stuffing. Your audience and AI care more about the quality of your post.

Monitor trends and track your results.

Stay on top of AI engine evolution. GEO is a new technique, so expect best practices to emerge over time. Trends in the search landscape are continuously changing.

Right now, Google is prioritizing EEAT, but that may change as AI overviews take over the scene.

Keep an eye on the traffic and conversions of posts you’ve written using GEO best practices. You can also leverage HubSpot’s new AI Search Grader, which lets you know how in line your posts are with GEO.

Generative Engine Optimization FAQ

What is GEO?

Generative engine optimization (GEO) is a method of improving your content’s visibility to AI generative engines, increasing its reach.

How do I structure my content for GEO?

Ensure your content is clear, well-organized, and has credible sources. Take advantage of lists and H2s and incorporate quotes and statistics when possible.

How do I check if my content is optimized for GEO?

Because GEO is so new, there aren’t many tools to measure how successful your content will be with AI generative engines. HubSpot’s AI Search Grader App is the only tool on the market that can scan your content for its GEO performance.

All you have to do is drop your URL into the grader. From there, you’ll have custom suggestions on what areas you can change to optimize your AI search performance.

That may include including more authority and personal experience or shifting the focus area of your page.

What is AI looking for in GEO content?

Large language models (LLMs) are looking for clear, well-structured information that they can pull and summarize to respond to user queries.

AI scanning works best for content that:

• Uses headers and lists
• Incorporates expert quotes
• Cites sources
• Uses simple language and scannable paragraphs

Is GEO going to replace SEO?

No — search engines aren’t going anywhere, so SEO isn’t, either. It’s best to consider GEO as an extension of SEO practices, as opposed to a replacement.

Best practices (like using H2s and credible citations) are shared between GEO and SEO, as are worst practices. Keyword stuffing, for example, has a negative impact on both SEO and GEO.

How can you measure GEO success?

Liam Carnahan, an SEO coach and content strategist, has a recommendation for how you can leverage existing tools to check if your article is getting picked up by AI engines.

"Of all the popular search-oriented options out there right now, Perplexity does the best job of citation," Carnahan says.

He continues, "So when I‘m trying to understand whether LLMs are ‘enjoying’ my content, I’ll go there first, and type in questions and prompts I imagine people might ask, using keywords I know my content is ranking for, to see how often it shows up in citations there."

Carnahan also notes that this isn‘t the most elegant solution, but “for now, it can give me a good idea about which of my clients’ content is ranking in AI results, and which content pieces are missing the mark.”

What’s Next for GEO

My biggest takeaway from exploring GEO is that it emphasizes different things, but in many ways, is similar to SEO. While SEO focuses on keyword optimization and backlinks, GEO focuses more on content structure.

Content writers are going to have to balance both but, thankfully, they click together well. I think we’re going to see clearer, more helpful content as a result of these two strategies — which is something I’m really excited about.

AI isn’t going anywhere. While we’re still learning how to use and measure GEO, it’s clear that keeping an eye on emerging techniques is going to be the key to success in the digital content world going forward.