Recently, I worked on a project where I used Google Analytics (GA4) data, accessed via BigQuery, to create a packed bubbles visualization. This visualization showcased key event counts per landing page on a website and included an interactive table that provided deeper insights. Here’s how I approached the project, the insights we derived, and the impact this had on the client’s conversion tracking strategy.
The client’s website had numerous events and conversions set up, but they lacked clarity on where the most important events were happening and how their tracking setup could be optimized. The main goals were:
To identify which landing pages were driving the most key events (such as clicks, purchases, form submissions, etc.).
To audit the client’s conversion tracking setup and identify gaps or inconsistencies.
To visualize the data in an intuitive and interactive way to make insights actionable.
Solution
I created a packed bubbles visualization in Tableau using GA4 data queried through BigQuery. This visualization was filtered to display data for the most recent three months, and it allowed users to click on a bubble (representing a landing page) to filter a table below.
This made it easy to see where the most conversion events were occurring on the site and drill down on which events were being tracked for each page.
Step-by-Step Process
1. Creating the Packed Bubbles Visualization
Using Tableau, I created a packed bubbles chart where:
Each bubble represented a landing page.
The size of the bubble corresponded to the total count of key events on that page.
The chart was date-filtered to show data from the most recent three months.
This visualization made it easy to quickly identify which landing pages had been driving the highest event activity.
2. Adding an Interactive Event Table
The next step was to make the visualization more actionable. I added a filterable table below the packed bubbles chart, which displayed:
Event names that occurred on the selected landing page.
Counts of each event.
When users clicked on a bubble (representing a landing page), the table dynamically filtered to show:
The specific events that occurred on that landing page.
How many times each event occurred.
This interactivity provided a more granular view of the event data and allowed stakeholders to drill down into what was being tracked on specific pages.
3. Auditing and Structuring Conversion Tracking
The packed bubbles visualization and interactive table provided a clear picture of how events were distributed across landing pages. By analyzing this data, we identified:
Overlapping events: Some landing pages had multiple similar events, indicating redundancy in the tracking setup.
Missing key events: Certain high-traffic landing pages lacked critical conversion events, such as form submissions.
Underutilized pages: Some pages with significant traffic were not generating meaningful event activity, suggesting potential UX or content issues.
Using these insights, we were able to:
Optimize the event taxonomy by grouping redundant events and defining new, consistent naming conventions.
Refine conversion goals to ensure they aligned with the client’s business objectives.
Propose tracking enhancements on underutilized landing pages to capture more actionable data.
Takeaways
Visualization Brings Clarity: The packed bubbles view was an effective way to visualize key event distribution across landing pages, highlighting patterns and anomalies at a glance.
Interactivity Facilitates Exploration: The clickable bubbles and dynamic table allowed the client to easily explore event data without needing technical expertise.
Data Informs Tracking Improvements: By auditing the event data, we uncovered inconsistencies in the tracking setup and proposed a more structured approach, enabling better storytelling with conversion data.
Focusing on High-Value Pages: The visualization helped prioritize pages for optimization efforts, particularly those with high traffic but low event activity.
Comments