Guide 64 · Reports · GA4 + Tag Manager
Build a path exploration: what users actually do before and after a page or event
A path exploration draws the branching routes users take from a starting point, or backwards from an ending point. Start backwards from purchase to see what converters really did.
After step 5, the tree shows the most common next pages after your starting point, or the most common pages before your ending point, with user counts on each branch.
Download the report reference: the Path exploration reference (JSON) (1469 bytes, MD5 b76125082c3e9768760464322a660876). It holds the report or exploration definition and the worked example with the recomputed figures shown on this page.
What you need
- Viewer access or above on the property to build.
- The entry point: Explore in the left navigation, then the path exploration template.
- One starting point or ending point, named in advance — for example, the /pricing page, or the purchase event. A path exploration without a fixed point shows everything and explains nothing.
The trace
- Open Explore and pick the path template. Click Explore in the left navigation, then create a blank exploration and change the technique to Path exploration, or pick the path template. The canvas shows a branching tree.
- Pick the direction before the point. Choose forward from a starting point, or backward from an ending point. Backward from purchase is the more useful direction for a store: it shows what converters actually did before they bought, instead of what you hoped they did.
- Set the starting or ending point. In Tab Settings, set the point — for example, the purchase event as the ending point, or the /pricing page as the starting point. The tree builds outward from that point, one node per page or event.
- Read the branches by user count, not by shape. Each branch carries the users who took that route. The thickest branch is the common route; a thin branch is a rare one. A route that looks interesting but carries twelve users is not a pattern.
- Follow one branch one level deeper, then stop. Click a node to expand the next step along that route. Two levels deep is usually enough to name the pattern; five levels deep is a maze. Save the exploration once the pattern is named.
Dead ends
DEAD END — STARTING WITH NO FIXED POINT. A path tree with no starting or ending point is every route at once. Fix one point first, then read the tree.
DEAD END — READING A THIN BRANCH AS A PATTERN. A branch with a handful of users is noise, however interesting its shape. Patterns need counts.
DEAD END — CHASING THE TREE FIVE LEVELS DEEP. Each level splits the users further, so deep branches hold almost nobody. Name the pattern at two levels and stop.
End state
A path exploration shows the routes users actually take. The worked example below recomputes the conversion from a starting page to purchase along three common routes, so the tree's thickest branch can be checked against its numbers.
Worked example — stated inputs, recomputed results. The routes below are stated inputs for the example. Every rate in the table was recomputed from them.
| Row | Sessions | Engaged sessions | Engagement rate | Key events | Key event rate | Revenue |
|---|---|---|---|---|---|---|
| /pricing then purchase | 640 | 520 | 81.3% | 96 | 15.0% | $7,420.50 |
| /features then purchase | 410 | 246 | 60.0% | 41 | 10.0% | $3,180.25 |
| / then purchase | 380 | 190 | 50.0% | 22 | 5.8% | $1,690.00 |
| Total | 1,430 | 956 | 66.9% | 159 | 11.1% | $12,290.75 |
Revenue per session is $12,290.75 ÷ 1,430 sessions = $8.59. Engagement rate is 956 ÷ 1,430 = 66.9%.
Sources
- Analytics Help — Get started with Explorations — path exploration visualises the paths users take as they interact with the site and app; explorations are built in the Explore section
- Analytics Help — Confirm that you are collecting data — many reports and explorations can take 24 to 48 hours to process data
Checked 11 November 2026. The click paths above are the ones in Google's help pages, opened or read 11 November 2026 (the Explorations and detail-report pages were opened in full; the report, library, navigation and thresholds pages were read from Google's help results the same day, after the browser hit a rate limit). No step in this guide was run inside a live Analytics property, because that needs a Google sign-in and this site does not sign in as anyone. The worked-example figures on this page were recomputed in Python with decimal arithmetic, and the downloadable JSON file carries the same recomputed figures. Google moves menu items without warning; if a path on this page no longer matches, the linked help page is the current map.