Technology Archives

GA4 Segments

Part 4 of many about GA4 and what is coming after UA

We will pick up this installment of the Analytics after UA here at the end of the last article GA4 Configuration by digging into segments and audiences.

With its robust audience builder and advanced generation tools, GA4 stands head and shoulders above Universal Analytics (UA). However, GA4 uses a unique approach to track screen views, unlike UA. Instead of treating mobile and web views separately, it combines data from both channels into one property. As a result, GA4 gives you a complete picture of a user’s journey across different devices, allowing for more comprehensive user profiles.

What Are Audiences?

Think of audiences as specific groups of users (‘user segments”) that you can use for analysis and advertising.

Groups are formed based on a range of criteria, such as dimensions, page visits, user attributes, and events (or their sequences) relevant to your company.

For instance, you could compare the behavior of US users against all users or users originating from your app against mobile users and desktop users.

It’s worth noting that audiences are re-evaluated as new user data is collected in GA4 automatically.

What Are Segments?

Segments and audiences may seem similar, but they have some key differences. Segments are data sections that meet particular requirements and can be analyzed independently.

For example, you can compare the purchase behavior of your local users to the conduct of all users or mobile users to desktop users to gain deeper insights.

Segments allow you to segregate certain groups of people and compare their behavior to the rest of the data rather than analyzing data from all visitors.

GA4 Segments and Audiences

You can opt to develop an audience based on a segment while creating a segment, but these two capabilities are distinct. Segments cannot be used in regular reports, although audiences may.

Exploration reports in GA4 include segments but do not incorporate audiences. However, despite being present in these reports, audiences, as opposed to segments, can be utilized for remarketing campaigns. Audiences are created based on segments and are shared across Google’s advertising products, which makes them suitable for targeted ad campaigns. Conversely, while they can be created and applied within the Explore section of GA4, segments are not used directly in Google’s advertising products like Google Ads.

Most significantly, segments are retroactive, whereas audiences are not because they begin gathering information only from the moment they are created.

How To Create Segments

In GA4, navigate to Explorations > Create a New Exploration.

There are three types of segments you can create:

1. - User segment: Includes all events associated with users who satisfy a specific set of criteria.
  - Sessions segment: Includes all events associated with those sessions that match the set of criteria.
  - Event segment: Includes only specific events that meet a set of criteria.

Let’s dive into how each segment type works with a simplified model.

Let’s say the user who visited the website performed the following actions during the first session:

1. - Page view.
  - Submitted form.
  - Made a purchase.

On the second session the next day, the same user performed these actions:

1. - App click through.
  - Page view.
  - Made a purchase.
  - Made a purchase.

If we create a user segment with the condition that the user submitted a form, it will contain data from all sessions because we tell GA4 to include all data that belongs to that user.

If we build the sessions segment with the condition that the user submitted a form during the session, it will contain only data from the first session because we told GA4 to retain data only from sessions where one has submitted a form.

Building a session segment of when someone watched a video will have data from both sessions because the user made a purchase during both sessions.

And finally, if we create an event segment for times when the user clicks a link, it will contain only click event data from the second session.

Segment Builder UI In GA4

User segment builder with time constraints

Here are some of the features to know about within segment builder:

1. - Sequence is available only in user segments and isn’t configured for session or event segments. You can use it to know how many users subscribe after visiting your home page.
  - You can apply time constraints to your segments. An example use case would be to know how many users visit your subscribe page and sign up within five minutes or how many users visit your product page and add it to the cart within five minutes.
  - Group level scoping: applies to all conditions within the group and is either across all sessions or within the same session within the same event. Please note that you can use specific sequence level scoping “across all sessions” and “within the same session.” Note that “across all sessions” scoping is available only in user segments.
  - “Exclusion group” is a valuable tool that allows you to refine your segment or audience by excluding specific users. For instance, you can create an audience comprising individuals who clicked the ‘Add to cart’ button but exclude those who have already purchased. An exclusion group enables you to develop a targeted audience for your remarketing campaign in Google Ads, specifically catering to those who expressed interest in your product but still need to complete the purchase.
  - OR/AND conditional logic operators can apply multiple criteria simultaneously in the groups.
  - By checking the “Build Audience” checkbox, you can also create an audience from that segment.

Example Of An Event Segment

Event segments are subsets of events triggered on your website or app.

Why would you need event segments if you already have users and session segments? Event segments, on the other hand, enable a more targeted and specific analysis than session-based or user-based segments.

They enable you to investigate people’s actions on your website and optimize the user experience.

As an example, let’s make a scroll-depth event segment. You may identify interesting sites by analyzing users that achieve specific scroll depth criteria, optimize content layout, and enhance overall user engagement.

Event Segment - Scroll depth greater than 50% — Event Segment – Scroll depth greater than 50%

In our case, we want to view data from people who went 50% of the way down the page and noted that only “within the same event” remained.

Granular scroll depth monitoring necessitates a specialized configuration, which we will discuss in the following steps. If you don’t have it configured, you may use the “scroll” event without the special parameter “percent_scrolled.”

Here are some examples of when you might want to use event segments instead of users or session segments.

- - Clicks Events: Make a segment from click events like “Add to Cart” or “Submit Form.” A click event enables you to analyze sessions or users who actively engage with your call-to-action buttons and quantify the effectiveness of individual call-to-action buttons.
  - Show Page Engagement:
    - Interaction with Dynamic Content. Analyze interactions with dynamically loaded or personalized content pieces, such as image carousels, actor bios, or press releases.
    - Video Engagement: A segment can be created depending on events such as “video_start,” “video_progress,” or “video_complete.” This segment lets you examine sessions or users interacting with video content on your website or app.
    - Buy Tickets: Set up custom event monitoring to generate segments based on unique interactions important to your business, such as buy ticket button clicks (to your event ticketing site), to assist you in identifying and analyzing problematic sessions or individuals.
  - Error Tracking: Form validation errors, broken links, and server-side problems can all be used to create a segment. The benefit: error-focused event segments allow you to discover and analyze sessions or users that faced issues, enabling you to improve error handling, fix broken elements, and improve the overall website or app operation.

Example Of A User Segment

At DWG, we have sign-ups, and we are interested in how long it takes a user to sign up for our newsletter after landing on our article pages.

For that, we should build a segment of users who visited the homepage for our shows, followed by a “buy_tickets” event within 5 minutes.

For that, we choose “first_visit” as an event.

Add subsequent events & time constraints

If you want to apply a time constraint to a specific step, you can utilize the time constraint option within the sequence. However, in our case, we have chosen a sequence global time constraint, which applies to all steps within the sequence.

We’ve found from this segment that 7.5K out of 12K subscribers (50%) subscribe within 5 minutes when they visit the homepage.

If you run a shop, you can create a segment of customers who buy within a day of browsing a product page.

There are unlimited ways to use this, and I’d like to emphasize that GA4’s new segment or audience builder is far more flexible and powerful than Universal Analytics.

Here are a few ideas of segments you may need to use for a users segment:

1. - User Acquisition: Understand user behavior from different acquisition channels (organic search, paid ads, social media, email marketing, DSP). Tracking the user across to the event ticketing site across sessions with event tracking across all sessions. Get a full ROI of all your marketing efforts, based on revenue, order value, and purchase frequency.
  - Demographic Analysis: Segment users by age, gender, location, or interests to personalize marketing strategies.
  - Churn Analysis: Identify and re-engage users who have churned via remarketing campaigns.
  - Cohort Analysis: Compare new and returning users’ engagement, retention, and conversion rates.
  - Customer Lifetime Value (CLTV): Identify high-value users based on revenue, order value, and purchase frequency.

Example Of A Session Segment

While user segments help analyze long-term behavior and user attributes, session segments help analyze user interactions within a single session.

For content engagement analysis reports, session segments are the best option. For example, you can use session segments to analyze how visitors interact with various types of content on your website or to study user behavior based on referral sources.

In the sessions segment, everything is the same, but it no longer has the following:

1. - “Across all sessions” scope.
  - Sequences.
  - Time constraints.

Let’s build a segment of users who arrived at our website from social.

Session Segment - Social Traffic — Session Segment – Social Traffic

.*(?:facebook\.com|instagram\.com|twitter\.com|linkedin\.com|pinterest\.com|youtube\.com|t\.co|lnkd\.in).*

Here are a few ideas for how you can use session segments vs. users:

1. - Content Engagement: Analyze sessions where users interacted with specific content types.
  - Campaign Performance: Evaluate sessions originating from specific marketing campaigns or channels to understand your marketing efforts’ effectiveness.
  - Conversion Funnels: Analyze sessions that progress through crucial steps of your conversion funnel, from initial engagement to goal completion, helping you optimize the user journey.

As we learned how to use segment builder in UI, we can create an audience quickly. In the next part we will focus on the Audiences and get to the bottom of how to create audiences and multiple examples.

Dream Warrior Group, a Los Angeles Based web design and digital marketing Company, providing solutions for your online marketing needs. Our expertise includes Search Engine Optimization (SEO), Social Media Posts & Marketing & Google PPC campaigns. Call us now at 818.610.3316 or click here.

GA4 Configuration

Part 3 of many about GA4 and what is coming after UA

This installment of the Analytics after UA deals with detailed work you need to do to ensure you have completed the basic setup of GA4 by performing a proper GA4 configuration. We will pick up here at the end of the last article:

Next steps with your new GA4 property

After completing the process with the wizard, you’ll see “You have successfully connected your properties” at the top of your Google Analytics 4 Property Setup Assistant page. If your Universal Analytics property name is “Example property (UA-nnnnnnn),” your GA4 property name will be “Example property – GA4 (xxxxxxx)”, without a “UA-” prefix, and where xxxxxxx is a new property number.

Set up the GA4 Configuration tag

Step 1: Create a GA4 Configuration tag

Start by creating a Google Analytics: GA4 Configuration tag to send data to your Analytics property.

1. 1. 1. 1. In Google Tag Manager, click Tags > New.
      2. Enter a name for the tag at the top (e.g., “GA4 Configuration – example.com”).
      3. In the Tag Configuration box, select Google Analytics: GA4 Configuration.
      4. Configure your tag:
        
        In the field Measurement ID, enter your “G-” ID. Keep the Send a page view event when this configuration loads option checked to automatically send page view events.
      5. Optional settings:
        
        Server-side tagging: Select Send to server container to send all GA4 events to a Tag Manager server container instead of Google Analytics. Specify the Server Container URL of your Tag Manager server container. Learn more about Server-Side Tagging.
        
        Parameters: Add any parameters you’d like to configure in Fields to Set. Use recommended event parameter names so that Google Analytics can populate dimensions and metrics for you.
        Example 1: To set a user ID, add a row to Fields to Set. Set the Field Name to user_id, and the Value to a Tag Manager Variable that returns the user ID.
      6. Example 2: Use Fields to Set to configure cookie field settings:
      7. Custom properties: Add any custom user properties that you’d like to configure in User Properties.
        Note: Analytics automatically collects some user dimensions so you don’t have to define user properties for them. You can set up to 25 additional user properties per Google Analytics 4 property.

Step 2: Create a trigger

Next, set up a trigger to load the Google Analytics: GA4 Configuration when someone loads your website.

1. 1. To ensure that the Google Analytics 4 Configuration tag fires before other triggers, click Triggering and use the Initialization – All pages trigger. Learn more about Page triggers.
  2. Save the tag configuration.

Result

Your tag configuration should look like this:

Step 3: Verify your tag works

To make sure your tag works as intended:

1. 1. In Google Tag Manager, click Preview. The Tag Assistant opens.
  2. Enter your site’s URL.
  3. Check if the Google Analytics: GA4 Configuration fired when the page loaded.
    ✅ If your tag fired successfully, the Tag Assistant UI look like this:❌ If your tag didn’t fire, check your tag’s trigger settings.
  4. When you are done with debugging, close Preview mode.
  5. In your Workspace, click Submit to publish your changes.

Set up events

To set up an event using Google Tag Manager, you will configure a Google Analytics: GA4 Event tag and then create a trigger that specifies when you want to send the event.

The following steps show you how to send a custom event to a Google Analytics 4 property when a user clicks a button to sign up for your newsletter. The steps show you how to implement the event using Tag Manager and don’t require you to implement a data layer object.

Step 1: Create a GA4 Event tag

Start by creating a Google Analytics: GA4 Event tag for the new custom event.

1. 1. In Google Tag Manager, click Tags > New.
  2. Enter a name for the GA4 Event tag at the top (e.g., “GA4 Event – Signup newsletter”).
  3. Select Google Analytics: GA4 Event.
  4. In the Configuration Tag, select your Google Analytics: GA4 Configuration tag.
  5. In Event Name, enter a name for the event (e.g. signup_newsletter). This will create a new custom event and the name will appear in your GA4 reports. To create a recommended event, use one of the predefined event names.

Step 2: Create a trigger

Next, create a trigger to send the event when someone clicks the button.

1. 1. Click the Triggering box in your GA4 Event tag.
  2. Click + on the top right.
  3. Enter a name for the trigger (e.g., “Trigger – Signup newsletter”).

You can choose the conditions for sending the event. The following example sends the event based on the button label:

1. 1. Click the Trigger Configuration box in your trigger.
  2. Choose All Elements.
  3. Click Some Clicks.
  4. Set the following trigger condition: “Click Text contains Sign up for the newsletter”.
  5. Save all your changes.

If you want the event to trigger when someone views a page (e.g., on a confirmation page), you could use a Page View trigger instead.

Step 3: Preview your changes

Before you publish your new event in Tag Manager, click Preview to see the data that’s recorded when you click the “Sign up for the newsletter” button.

You can use preview mode to test changes to your container before you publish those changes to your website.

See your events in Analytics

You can see your events and their parameters using the Realtime and DebugView reports. Note that the DebugView report requires some additional configuration before you can use the report.

Analytics after UA

Part 1 of many about GA4 and what is coming after UA

Hey there! Google Universal Analytics (UA) will sunset before you know it, and your analytics after UA will be dominated by Google Analytics 4 (GA4). I’m thrilled you’re joining me as we dive into the world of SEO analytics with GA4. GA4 is already emerging as the key to gaining comprehensive and user-centric insights into customer journeys across multiple devices and platforms. So, let’s dive in and uncover how GA4 can help us elevate our SEO game.

We might still be learning, but with GA4, we’ll be able to answer two critical questions: How do people find our website, and what do they do once they get there? The most significant change in GA4 is its user-centric approach to tracking and analysis. The new approach means we’ll look at the entire user experience rather than fragmented experiences on individual devices or mediums.

GA4 Events

Before we proceed, let’s familiarize ourselves with events in GA4. The platform is all about event-based tracking, simplifying the process of creating and customizing events, tracking conversions, and building reports. So, Events dominate the landscape of GA4. For example, in the table below, you can see that so many properties that we took for granted in UA are now specific events in GA4:

UA to GA4 — Google table describing some UA properties that are Events in GA4

Now, let’s talk about setting up events. GA4 automatically collects specific event data, but we should set up event tracking for the events that aren’t captured natively. This gives us significantly more data flexibility and empowers us by making it easy to design and personalize events.

event-based-vs-session-based — The difference between UA session-based model and GA4 event-based model. (Source: Rittman Analytics)

To make the most of GA4, we should start by auditing what our GA4 account is already tracking, understand the parameters we’d like to collect and think about the audiences and segments we’d like to build. We should also work backward to understand which events to track and consider event-tracking limitations when finalizing the list of custom events to collect.

In addition to defining events, there are a few GA4 settings we should be aware of. The native events are either web, app, or both. You can find a complete list of these events at Analytics Help.

One of the best things about GA4 is that it unlocks more powerful segmentation abilities, giving us better context around our organic performance. Segments are a subset of users, sessions, or events.

You can apply segments to explorations to focus on specific data sets. We can create segments based on users, sessions, and events, allowing us to ask essential questions and assess our SEO performance more effectively.

GA4 makes for better analytics

Google Analytics 4 is revolutionizing how businesses measure their online presence, and we’re learning about it together. With its advanced features and improved measurement capabilities, GA4 will help us better track and optimize our digital content to improve our SEO performance. So let’s keep learning and growing together!

Please remember, At the most basic level, you should extend the retention of user-level data to a maximum of 14 months as soon as we set up our GA4 property. Google is sunsetting Universal Analytics (UA) on July 1.

But don’t worry; many tools are out there to capture and store your UA data so you keep your historical UA data. Some of these tools host the event and will display it for you, and many more host it as google sheets.

In addition, you can use any number of visualization tools to access all of your historical performance data, giving you access to GA4 and UA metrics simultaneously for the same reports. If you need any help having your data retrieved or setup, please don’t hesitate to contact us.

SEO AND AI – Today

While writing another piece about the future of AI and Large Language models for our blog, it hit me: I need to address today’s realities more. So here is a piece on what we have on the relationship between SEO and AI, today.

Even though AI models like ChatGPT are exceptional at generating statistically likely text, there may be better choices for creative storytelling. However, AI-generated content serves a valuable purpose in functional content creation, which can benefit industries such as e-commerce, reference materials, and more.

To effectively utilize AI-generated content, consider the following action plan:

1. Identify functional content requirements: Understand the applicable content AI can efficiently generate, such as data-driven reports, event summaries, FAQ sections, how-to guides, news aggregation, job listings, legal documents, technical documentation, and directory listings. There are certain content types that AI excels at, and we recommend that you limit your functional content to the following:
  1. - Data-driven reports
    - Event summaries
    - FAQ sections
    - How-to guides and tutorials
    - Compiling and summarizing news
    - Job listings
    - Legal and policy documents
    - Technical documentation
    - Directory listings
2. Kickstart large-scale content projects: Leverage AI-generated content to quickly create a critical mass of content for projects like recipe collections or travel guides. Building on AI content can help you reach your content goals more rapidly while maintaining quality through proper SEO and content management practices.
3. Optimize keywords with AI: Use AI like ChatGPT to explore alternative keywords, questions, or concerns related to your content and rewrite sections accordingly. Using AI to optimize your long-chain keywords will help ensure your content is more engaging and relevant to your target audience.

For example, we worked with a startup founder to successfully use ChatGPT to generate his pitch outline to present his platform, resulting in a significant increase in interest and, finally, his crucial customer’s adoption of his platform.

In summary, to create a cohesive thought and action plan, you should focus on:

- Identifying the correct type of content for AI-generated content creation
- Utilizing AI-generated content to kickstart large-scale projects
- Employing AI for keyword optimization and audience engagement

By following this action plan, you can make the most of AI-generated content to enhance your content strategy and achieve your goals.

The Future of SEO and Virtual Agents

This article is the fourth installment in our series discussing search and the increasing influence of artificial intelligence (AI) on the future of SEO (Search Engine Optimization). This installment delves into the evolution of the search function, the Intelligent Agents, and the interaction of goal-oriented search, Virtual Assistants (VA), end-users, and businesses.

The future of SEO means shifting to L2O.

Artificial intelligence is changing how we interact with the digital world. Soon, AI Assistants will collect and act on massive amounts of data on our behalf. These assistants will force drastic changes in the search and search display definition. In many instances, making the act of searching nearly undetectable. However, we have not reached the utopian society depicted in 24th-century Star Trek. So, the creators of these VAs will devise ways to keep their systems updated with the latest information. The updated content will be retrieved from public content. The fresh content will be selected with a new type of SEO, L2O.

Virtual Assistant and the Future of Search

With the rapid advance of AI, bots can quickly process, analyze, and synthesize data and take action. Bots are transforming from our knowledge engines to our Virtual Assistants (VA). VAs predict our needs, desires, and preferences with unprecedented accuracy and will affect desired results. So, instead of googling and using phones or browsers, we interact with our VAs, state our goals, and watch it complete the task.

The shift towards proactive, predictive and personalized search will have profound implications for how we experience the internet. Traditional search engines will be replaced by AI Agents(1). The AI will operate seamlessly in the background, and take appropriate action for its user.

The Future of SEO craft

The need for a new approach to search engine optimization is apparent as the web is increasingly driven by AI. L2O, or LLM Optimization, ensures that AI-powered systems can easily access, understand, and use the latest information.

L2O helps optimize the content and improve the compatibility between human-generated content and the AI systems that will increasingly dominate our online experience. The future of SEO involves creating content easily digestible by AI systems and developing strategies to ensure that the most relevant and up-to-date information is readily accessible to these platforms.

If you own a Moroccan restaurant, you want to provide detailed dynamic content about the ambiance, experience, quality, availability, and the menu. Then when a VA tries to find the best North African Cuisine in Los Angeles, you are available. Finally, combined with your reviews and social information, the VA makes an informed yet biased decision and makes a reservation at your restaurant.

The future of SEO is about biasing the VAs rather than actively increasing the traffic to a website. But here is the funny thing, L2O involves most of the components of today’s SEO.

One key component of L2O involves creating structured data that AI systems can quickly process and interpret. L2O will include using metadata, semantic markup, and voice search optimization that help AI systems understand the context, the flavor, and the meaning of content more effectively. L2O also emphasizes the importance of maintaining a robust online presence through regular updates, high-quality content, and strategic partnerships with other reputable sources of information.

Challenges and Opportunities

Several challenges exist as we move toward a future where Virtual assistants dominate(2) the web. For one, the increased reliance on AI may exacerbate existing inequalities and social biases, as AI systems may be more likely to favor the freely available content overload from established or well-funded sources.

VAs may prove far more gullible because of the upcoming data privacy limitations. Social bias, knowledge overload, and data privacy concerns could limit the diversity of information available to the VAs and easily allow for Black Hat techniques.

However, the evolution of SEO to L2O presents numerous opportunities for businesses, content creators, and individuals. By embracing L2O strategies, organizations can position themselves at the forefront of this new digital landscape, ensuring that their content remains visible and relevant in the age of AI-driven search.

The rise of AI-powered Assistant Artifacts signals a fundamental shift in how we interact with the digital world. As search becomes increasingly predictive and personalized, the importance of L2O as a means of optimizing content for AI systems will only continue to grow.

By understanding and embracing this new paradigm, we can ensure that the internet remains a rich and diverse source of information, even as we navigate the uncharted territory of an AI-driven future.

1- CMS Wire, What to Expect From the Future of Virtual Agents, Andie Burjek, 02/24/23

2- Data Driven Investor, AI Agent Companies, Harsha Angeri, 04/06/23