How to do path analysis in Looker Studio using GA4 and BigQuery

by

When I’m trying to understand how users move through my website (what pages they land on, where they go next, and where they drop off) path analysis is one of the most useful techniques I use. It helps me visualise the journey a user takes step-by-step, and spot opportunities to improve user experience and boost conversions.

While Looker Studio doesn’t include a native path analysis feature, I’ve found that combining it with Google Analytics 4 (GA4) and BigQuery gives me everything I need to track and visualise user flows effectively.

What is path analysis?

Path analysis is all about tracking the sequence of user interactions. Essentially, it shows you the order of events as people navigate through your site. It’s useful for answering questions like:

  • What do users typically do after landing on the homepage?
  • How many steps does it take to complete a purchase?
  • Where are users most likely to drop off?

Seeing this journey mapped out helps me make better, more data-informed decisions about what to change on the site.

Looker Studio’s limitations, and the workaround

Looker Studio is great for building visual reports, but it doesn’t offer a built-in path analysis tool like some analytics platforms do. So instead, I integrate it with GA4 and BigQuery. This setup allows me to export raw event data, process it, and then build custom visualisations that show user journeys in detail.

Here’s how I do it:

Step 1: Connect GA4 to BigQuery

First, I link my GA4 property to BigQuery. This integration is essential, as it enables me to export all the raw event data GA4 collects.

In GA4:

  • Go to Admin → BigQuery Linking
  • Follow the prompts to connect your property to your BigQuery project

After this setup, data starts flowing into BigQuery (usually with about a 24-hour delay).

Step 2: Prepare the data in BigQuery

GA4 data is rich, but it’s also nested and not quite ready for analysis straight out of the box. So, I write SQL queries to transform the data into something usable for path analysis.

One of the key techniques I use is the ROW_NUMBER() function. This lets me assign a step number to each event in a session, ordered by timestamp. Here’s a simplified version of the query I use:

SELECT
  user_pseudo_id,
  event_timestamp,
  event_name,
  ROW_NUMBER() OVER (
    PARTITION BY user_pseudo_id 
    ORDER BY event_timestamp
  ) AS step_number
FROM
  `your_project.your_dataset.events_*`
WHERE
  _TABLE_SUFFIX BETWEEN 'start_date' AND 'end_date'

This query sequences each event for a user and gives me the basis for visualising their journey.

Sometimes I’ll also clean up fields (like stripping domain names from page paths) or create calculated fields for better filtering and display. And for visualisations that have issues handling datetime fields, I convert them into strings to avoid formatting problems.

Step 3: Visualise paths in Looker Studio

Once the data is prepared, I connect Looker Studio to the processed BigQuery table or view. For path analysis, I typically use:

  • Sankey diagrams to visualise flow from one page or event to another
  • Community visualisations specifically designed for user journey mapping

When building these charts, I make sure to include:

  • A field for the page or event name (e.g. page_path or event_name)
  • A session ID
  • A step number or timestamp for sequencing

With this in place, I can visualise the paths users take through the site, including the most common routes, loops, and drop-off points.

Best practices I follow

To get meaningful insights from my path analysis, I follow a few key principles:

  • Start with a clear objective: Whether it’s spotting drop-offs in a checkout flow or tracking behaviour after landing on a blog post, I always define what I’m trying to understand first.
  • Segment the audience: New vs returning users, mobile vs desktop, etc… Different groups behave differently. Segmenting helps uncover more specific patterns.
  • Focus on the most important events: I prioritise events that align with key business goals, like view_item, add_to_cart, or purchase.
  • Review regularly: User behaviour changes, especially after site updates or marketing campaigns. I revisit my path analysis regularly to keep insights up to date.

Final thoughts

Path analysis has become a vital part of how I understand user behaviour in Looker Studio. While the platform doesn’t offer this natively, combining GA4 and BigQuery unlocks all the functionality I need to build custom journey visualisations tailored to my site.

It takes a bit of setup, but once in place, this approach provides clear, actionable insights into how users experience your site, and where you can improve it.

If you’re not already exploring path analysis in Looker Studio, I highly recommend giving it a try. It’s one of the most valuable techniques I use.

Leave a Comment