Web Analysis: Through the Use of GA4 and Chat GPT

•

MaraJay
In this blog post, I will guide you through evaluating the performance of my digital marketing efforts using Google Analytics 4 (GA4). This powerful tool allowed the analyzation of data across various metrics, offering valuable insights into the effectiveness of my digital marketing strategies. To complete this I focused on three areas: user acquisition, page views, and demographic details.

User Acquisition:

Understanding how new users find your website is crucial to evaluating your digital marketing strategies. In the first section, we’ll focus on User Acquisition, which tracks the number of first-time visitors coming to your site through LinkedIn, email campaigns, and other channels.

In order to start, I selected the data range “last 90 days”  to reflect the data from when I first launched the website three months ago. I aslo made sure that the table used was “first user source/ medium”. Once that’s set, I dove into three key metrics:

  1. New Users – The number of first-time visitors, giving insight into how effective your outreach efforts are at attracting fresh traffic.
  2. Engaged Sessions Per Active User – A metric that shows how often users are returning and interacting with your content.
  3. Average Engagement Time – The average duration users spend actively engaged with your site, reflecting how compelling your content is.

The information was then tranfered into an Excel Spreadsheet, for example:

 

Next, in order for ChatGPT to help analyze the information, I used the following prompt:

“When I sorted the (insert metric) in descending order in the First user source/medium classification, the following result came out. Please explain.”

The result for each metric are as follows:

New Users:

When sorting New Users by First User Source/Medium, the results indicate the following:

  1. (direct) / (none) – 152 new users:
    The highest number of new users come from direct traffic, which means visitors who either typed the website’s URL directly into their browser or came from bookmarks. This suggests that the website is gaining a steady stream of new visitors who are familiar with it, or possibly from word-of-mouth or offline promotions. This is a positive sign of brand recognition and interest.
  2. google / cpc – 13 new users:
    The number of new users coming from Google Ads (CPC) is relatively low, with only 13 new users. This indicates that the paid ads are not yet driving a substantial amount of new traffic to the site. There could be a need for better ad targeting, optimization of the ad copy, or more appealing landing pages to convert clicks into new users.
  3. linkedin.com / referral – 2 new users:
    Traffic from LinkedIn referrals is also quite low, with only 2 new users. This suggests that the content shared on LinkedIn is not effectively attracting new visitors to the site. It may be worth re-evaluating the messaging or targeting on LinkedIn to increase visibility and conversion.
  4. google / organic – 1 new user:
    Despite organic search being a major traffic source, only 1 new user has come from Google organic search. This points to a need for better SEO strategies or more targeted content that can rank higher in search results, thus attracting more new users. There may also be an opportunity to focus on improving keyword optimization and content relevance.
  5. url-opener.com / referral – 1 new user:
    Url-opener.com referrals bring in 1 new user, which is minimal. While this source is generating some traffic, it is not a major contributor to new user acquisition. It might be helpful to analyze the nature of this referral and determine if it is worth continuing to target.

Key Insights:

  • Direct traffic is by far the most significant source of new users, indicating that brand awareness or direct access is strong. This could mean users are coming back, or the brand is recognized through offline or organic channels.
  • Google Ads (CPC) and LinkedIn referrals are bringing in a small number of new users, suggesting the need for optimization in both paid campaigns and LinkedIn outreach efforts.
  • The low count of new users from Google organic search signals an opportunity to invest more in SEO efforts to increase visibility and attract new visitors through search engines.
  • Referral traffic from url-opener.com is minimal, suggesting that this source might not be as valuable in driving new users compared to other channels.

 

Engaged Sessions per Active User:

When sorting Engaged Sessions per Active User in descending order based on First User Source/Medium, the results show the following:

  1. (direct) / (none) – 0.36 engaged sessions per active user:
    Users who came directly to the website (without a referral or paid ad) have an average of 0.36 engaged sessions per active user. This means that visitors who are familiar with the site (possibly through direct URL entry, bookmarks, or offline marketing) are returning to the site at least once, which is a positive indicator of engagement. However, the engagement rate is still relatively low, suggesting that while these users are coming back, they might not be interacting with the site as frequently as they could be. This may indicate a need for more compelling content or incentives to increase engagement.
  2. google / cpc – 0 engaged sessions per active user:
    Users who arrived via Google Ads (CPC) show 0 engaged sessions per active user, which suggests that visitors from paid search campaigns are not returning to the site after their first visit. This could indicate that the landing pages or the ad content might not be engaging enough, causing users to leave the site without returning. It may be worthwhile to optimize the landing pages and ad targeting to better match user expectations and encourage repeat visits.
  3. linkedin.com / referral – 0 engaged sessions per active user:
    Similar to the CPC traffic, users arriving from LinkedIn referrals show 0 engaged sessions per active user. This could mean that the content shared on LinkedIn is not compelling enough to keep visitors coming back. The low engagement suggests that the referral traffic from LinkedIn is not driving users to interact with the site or explore additional content.
  4. google / organic – 1 engaged session per active user:
    Visitors coming from Google organic search show 1 engaged session per active user, which is the highest engagement rate in this classification. This suggests that organic search traffic is more likely to return to the site after their first visit, indicating that the content is more relevant to the search queries and better aligned with user interests. This is a positive indicator for SEO efforts, but there may still be room for improvement in encouraging further engagement.
  5. url-opener.com / referral – 1 engaged session per active user:
    Traffic from url-opener.com referrals also has 1 engaged session per active user, which is comparable to the Google organic traffic. While the overall volume of traffic from this source is low, users who do visit are engaging with the site and returning for additional sessions. This could indicate that visitors from this referral source find the content valuable enough to come back.

Key Insights:

  • Direct traffic users show some engagement with a moderate number of repeat sessions, suggesting that there is a base of returning visitors who are somewhat engaged with the content.
  • Paid traffic (CPC) and LinkedIn referrals have no engaged sessions per active user, indicating a need for better optimization in both paid campaigns and social media referral efforts to drive user retention and engagement.
  • Organic search users are the most engaged, with a higher likelihood of returning to the site after their first visit. This signals that SEO strategies are bringing in users who are more interested in the content and likely to engage further.
  • Referral traffic from url-opener.com also shows a decent level of engagement despite low traffic volume, indicating that the users coming from this source are finding value in the site.

Average Engagement Time:

When sorting Average Engagement Time in descending order based on the First User Source/Medium, the results show the following:

  1. (direct) / (none) – 31s:
    Users who came to the website directly (without a referral or campaign) have the highest average engagement time. This indicates that visitors who type in the URL or have it bookmarked are more likely to engage with the content for a longer period. This could be a sign that they are familiar with the site and find it valuable enough to spend time on.
  2. google / cpc – 0s:
    Users who arrived via Google Ads (CPC) show 0 seconds of average engagement time. This could imply that the landing page may not be appealing or relevant to the ad content, causing users to leave quickly. It may also indicate a need to improve the landing page experience or the targeting of the ads.
  3. linkedin.com / referral – 0s:
    Visitors who came through LinkedIn referrals also show 0 seconds of engagement. This suggests that the content shared on LinkedIn may not be compelling enough, or the link might not lead to a page that meets the visitors’ expectations, causing them to leave immediately.
  4. google / organic – 13s:
    Users coming from organic search (Google) have an average engagement time of 13 seconds. While this is a bit higher than those arriving from ads or LinkedIn, it still suggests that there may be room for improvement in terms of keeping organic visitors engaged. Enhancing content relevance and optimizing landing pages could increase this engagement time.
  5. url-opener.com / referral – 21s:
    Visitors who arrive through url-opener.com referrals spend 21 seconds on average, which is higher than visitors from Google organic or LinkedIn. This could indicate that the referring source is driving more engaged traffic, though it still falls short of the direct traffic group.

Key Insights:

  • Direct visitors (those who access the site without intermediaries) tend to engage the longest, which might suggest familiarity or loyalty.
  • Traffic from paid ads (Google CPC) and LinkedIn referrals is not converting well in terms of engagement, with 0 seconds of interaction. This could point to a need for better-targeted campaigns, more compelling landing pages, or more relevant content to keep users on the site.
  • Traffic from organic search and url-opener.com referrals shows moderate engagement, suggesting room for improvement in keeping visitors interested and engaged.

 

 

Page Views:

Page Views & Unique Page Views: they are metrics used to identify the most popular blog posts, providing insight into the topics and content types that are most appealing to employees of the target companies.

The 3 metrics analyzed are:

Average Engagement Time: This metric measures the duration users spend on the content. A higher average engagement time indicates captivating and relevant content, which is essential for leaving a positive impression.

Views: This metric reflects the reach of the content. A high view count indicates successful discovery and consumption, highlighting the effectiveness of promotion strategies, such as LinkedIn and email marketing, while also boosting visibility.

Users: This metric tracks the number of unique individuals accessing the content, providing insights into audience diversity and reach.

The information was then transfered to an Excel Spreadsheet, for example:

Next, ChatGPT was used to perform the analysis:

The prompt given was: “When I sorted the (insert metric) in descending order in the page title and screen class classification, the following result came out. Please explain.” Here are the results for each metric!

Average Engagement Time:

The data provided shows the Average Engagement Time for various pages when sorted in descending order under the “Page Title and Screen Class” classification. Here’s an explanation of the results:

  1. Metrolina Greenhouses Inc: AI Generated Marketing Content – MaraJay Creative has the highest average engagement time of 1 minute and 8 seconds, indicating that this content is the most captivating or relevant to users. This suggests that the page holds users’ attention effectively, likely due to its topic or structure.
  2. Metrolina Greenhouses Inc: Customer Persona – MaraJay Creative comes next with an average engagement time of 18 seconds, showing a moderate level of interest among users.
  3. Blogs – MaraJay Creative follows with an engagement time of 12 seconds, indicating lower but still noticeable user interaction.
  4. Hello my name is Mara! – MaraJay Creative has an average engagement time of 10 seconds, reflecting less user engagement compared to the previous pages.
  5. MaraJay Creative – Created for the purpose of BADM 445 has an engagement time of 11 seconds, suggesting average attention levels from users.
  6. Contact – MaraJay Creative has the lowest engagement time of 3 seconds, suggesting that users likely skim or quickly leave this page, possibly because it provides straightforward contact information with little other content to explore.

The descending order highlights the pages that are most effective at engaging users, offering insights into which content resonates more and which areas may need improvement to better capture audience attention.

Views:

The data sorted by Views in descending order under the “Page Title and Screen Class” classification provides insights into the reach and visibility of various pages. Here’s an explanation and key observations:

Analysis of the Sorted Data:

  1. MaraJay Creative – Created for the purpose of BADM 445 has the highest number of views (128 views).
    • Insight: This page is the most visited, suggesting it either has strong visibility (e.g., effective promotion or prominent placement) or is serving a purpose that draws significant interest from users.
    • Actionable Tip: Evaluate why this page is receiving high traffic and determine whether the engagement (time spent) aligns with its goals. Consider replicating the promotional strategies for this page across others.
  2. Hello my name is Mara! – MaraJay Creative has 33 views, making it the second most visited page.
    • Insight: This level of traffic might reflect user interest in learning more about the creator or background information.
  3. Contact – MaraJay Creative has 30 views, slightly behind the “About” page.
    • Insight: A reasonable number of visits to the contact page suggests users are seeking additional engagement or connection. This aligns with its purpose but could benefit from better integration with other pages to direct traffic back to content.
  4. Metrolina Greenhouses Inc: Customer Persona – MaraJay Creative (26 views) and Blogs – MaraJay Creative (25 views) show moderate performance.
    • Insight: These pages might have a niche audience or less effective promotion. However, their traffic levels indicate potential growth opportunities with better marketing efforts.
  5. Metrolina Greenhouses Inc: AI Generated Marketing Content – MaraJay Creative has the lowest views (13) despite having the highest engagement time in the previous analysis.
    • Insight: This low view count suggests a lack of visibility, even though the content is highly engaging.
    • Actionable Tip: Increase promotion of this page (e.g., via LinkedIn, email campaigns, or SEO optimization) to improve reach. The combination of high engagement and increased views could maximize its impact.

Key Insights and Recommendations:

  1. High Traffic vs. Engagement:
    • The page with the most views (BADM 445) does not necessarily have the highest engagement time.
    • Recommendation: Investigate whether users are finding the content valuable or simply skimming it. Adjust content quality and layout if needed to keep users engaged.
  2. Balancing Visibility and Engagement:
    • Pages like AI Generated Marketing Content are highly engaging but not widely viewed. Conversely, high-view pages like BADM 445 may lack engagement.
    • Recommendation: Focus on cross-promoting high-engagement pages to increase their reach while optimizing high-view pages for better user retention.

 

Users:

he data sorted by Users under the “Page Title and Screen Class” classification highlights the unique individuals visiting each page. Here’s a breakdown of insights and observations:

Analysis of the Sorted Data:

  1. MaraJay Creative – Created for the purpose of BADM 445 leads with 94 unique users.
    • Insight: This page has the broadest reach, suggesting it has been effectively promoted or is highly relevant to a wide audience.
    • Actionable Tip: Since this page attracts a high number of users, assess its content to ensure it provides value and aligns with its goals. Consider linking this page to other less-viewed but high-quality pages to encourage further exploration.
  2. Hello my name is Mara! – MaraJay Creative has 31 unique users, making it the second most visited page by unique individuals.
    • Insight: This likely reflects interest in learning about the brand or individual behind the site.
    • Actionable Tip: Enhance this page with engaging elements, such as a call-to-action (CTA) that directs users to other key pages.
  3. Contact – MaraJay Creative has 27 unique users, showing a reasonable level of interest in connecting or finding contact information.
    • Insight: This is typical for a contact page, as users generally visit briefly for specific information.
  4. Metrolina Greenhouses Inc: Customer Persona – MaraJay Creative has 15 unique users, and Blogs – MaraJay Creative follows closely with 14 unique users.
    • Insight: These pages are attracting a smaller, more focused audience. The niche nature of their topics may explain the lower numbers.
    • Actionable Tip: Use promotional strategies such as email campaigns or targeted social media posts to increase visibility for these pages.
  5. Metrolina Greenhouses Inc: AI Generated Marketing Content – MaraJay Creative has the lowest unique user count (8 users) despite its previously identified high engagement time.
    • Insight: This highlights an opportunity to increase its exposure since the page performs exceptionally well in holding attention but is not reaching enough users.
    • Actionable Tip: Promote this page more actively via SEO, LinkedIn posts, or email newsletters to attract more visitors.

Key Insights and Recommendations:

  1. High Traffic Pages:
    • Pages like BADM 445 attract the most unique users, but their engagement metrics should be monitored to ensure that the content meets user expectations.
    • Consider optimizing CTAs on this page to direct traffic to lower-performing pages.
  2. Low Traffic Pages:
    • Pages like AI Generated Marketing Content show a disconnect between the quality of content (high engagement time) and visibility (low unique users).
    • Prioritize improving the reach of such pages by increasing promotion through targeted ads, social sharing, or inclusion in email campaigns.

By analyzing user counts alongside engagement and view metrics, it becomes clear which pages resonate with the audience and which need enhanced visibility or optimization to achieve their potential.

 

Demographic Details:

This section examines the geographic locations of website visitors, providing insights into how effectively specific demographic areas and companies in those regions are being targeted. The analysis uses the “city and country” dropdown feature in GA4 to assess performance across various metrics, including:

  • Active Users (Previously Users): The total number of unique visitors from each location.
  • Engagement Rate: The percentage of users from each location actively engaging with the content.
  • Engaged Sessions Per Active User (Previously Engaged Sessions Per User): The frequency with which users from each location return to the site.
  • Average Engagement Time Per Active User (Previously Average Engagement Time): The amount of time users from different locations spend actively interacting with the content.

 

 

The information for the five cities: Boydton, Des Moines, Washington, Spearfish, and Moses Lakes, was then tranfered to an Excel Spreadsheet.

In order for ChatGPT to analyze this data set, I gave the following prompt:

Let’s begin the analysis assuming my cities of interest are Boydton, Des Moines, Washington, Spearfish, and Moses Lake

When I sorted Active users in descending order in the Country and city classification, the following result came out. Please explain. Boydton 20 Des Moines 15 Washington 14 Spearfish 13 Moses Lake 11

When I sorted Engagement rate in descending order in the Country and city classification, the following result came out. Please explain. Boydton 15% Des Moines 6.67% Washington 14.29% Spearfish 72.72% Moses Lake 36.36%

When I sorted Average engagement time per active user in descending order in the Country and city classification, the following result came out. Please explain. Boydton 0s Des Moines 0s Washington 6s Spearfish 4m21s Moses Lake 0s

When I sorted Engaged sessions per active user in descending order in the Country and city classification, the following result came out. Please explain. Boydton .15 Des Moines .07 Washington .14 Spearfish 1.23 Moses Lake .36

Finally, please provide a conclusion based on the answers to these four questions.

 

  • The result:

  • Analysis by Metrics and Locations

    1. Active Users (Descending Order)

    • Boydton: 20 users
    • Des Moines: 15 users
    • Washington: 14 users
    • Spearfish: 13 users
    • Moses Lake: 11 users

    Explanation:
    Boydton has the highest number of active users, indicating it is the most successful location in attracting visitors. Des Moines and Washington follow closely, showing similar levels of reach. Spearfish and Moses Lake have fewer users, but they are still part of the target audience, suggesting potential for growth with targeted campaigns.

    2. Engagement Rate (Descending Order)

    • Spearfish: 72.72%
    • Moses Lake: 36.36%
    • Washington: 14.29%
    • Boydton: 15%
    • Des Moines: 6.67%

    Explanation:
    Spearfish shows an exceptional engagement rate, with nearly three-quarters of its users actively interacting with content, suggesting strong resonance with the content or a highly focused audience. Moses Lake also performs well, while Washington, Boydton, and Des Moines exhibit significantly lower engagement rates, indicating a need for content optimization or better targeting in these locations.

    3. Average Engagement Time Per Active User (Descending Order)

    • Spearfish: 4 minutes 21 seconds
    • Washington: 6 seconds
    • Boydton, Des Moines, Moses Lake: 0 seconds

    Explanation:
    Spearfish significantly outperforms all other cities in average engagement time, indicating that users from this location are deeply engaged with the content. Washington shows a minimal engagement time, while Boydton, Des Moines, and Moses Lake register no measurable engagement, suggesting that visitors from these locations may leave the site quickly or are not finding the content relevant.

    4. Engaged Sessions Per Active User (Descending Order)

    • Spearfish: 1.23
    • Moses Lake: 0.36
    • Boydton: 0.15
    • Washington: 0.14
    • Des Moines: 0.07

    Explanation:
    Spearfish again leads with the highest number of engaged sessions per user, indicating repeat visits and strong content retention. Moses Lake shows moderate repeat engagement, while Boydton, Washington, and Des Moines exhibit minimal repeat interactions, suggesting either low interest or insufficient motivation for users to return.

    Conclusion

    The analysis reveals significant variations in performance across the cities of interest:

    1. Spearfish emerges as the most engaged location, with high engagement rates, long average engagement times, and frequent repeat visits. This suggests that the content resonates well with users in this city. Efforts should be focused on understanding what makes Spearfish successful and applying similar strategies to other locations.
    2. Moses Lake performs moderately in engagement metrics, indicating potential for further optimization. Enhanced promotion and content tailored to this audience could boost performance.
    3. Boydton, Des Moines, and Washington attract the highest numbers of active users but show poor engagement rates, little or no average engagement time, and minimal repeat sessions. This indicates that while the site is reaching users, it struggles to keep their attention or prompt meaningful interaction. Focused strategies are needed to improve content relevance, user experience, and calls-to-action in these areas.

    By analyzing these metrics, the key takeaway is to replicate the success seen in Spearfish across other locations while addressing the challenges in cities like Boydton and Des Moines to maximize overall website performance.

    Conclusion:

    After analyzing the geographic performance of my website, it’s clear that some cities are engaging more effectively than others. Spearfish stands out with the highest engagement rates, long average engagement times, and frequent return visits, showing that the content is resonating well with this audience. On the other hand, while cities like Boydton, Des Moines, and Washington attract the most visitors, their engagement levels are lower, with users spending minimal time on the site and returning less often. This suggests a need to better connect with these users, refine the content, and boost engagement through targeted strategies.

•

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *