Skip to content

Guide 188 · Conversions · GA4 + Tag Manager

Key events in GA4 Explore reports: build a free-form exploration on one key event

Checked 11 November 2026 · Verification: key-event reference validated structurally (9 checks, all passed) · Steps traced to Analytics Help, read 11 November 2026 · 5-minute read

Explore is where a key event meets a question the standard reports cannot ask, such as how one key event breaks down by landing page and device together. A free-form exploration with Event name as the rows and Key events as the value does that in one table. This guide builds it for purchase, filters it, and works out the share of purchases that came from mobile.

Expected signal

After the steps, the purchase key event appears under the Key events tab in Admin, and the standard reports show it within a day. The worked example below gives the count and the rate to expect.

Download the reference: the purchase key-event reference (JSON) (696 bytes, MD5 e59c84651461aba4e1979f59ccc46f25). It holds the event name, the counting method, the worked-example figures recomputed for this guide, and the checkpoints that prove the key event is recording. The figures in the file are the same figures shown on this page.

What you need

  • Editor access or above at property level in GA4. Marking an event as a key event is a property setting, and the star in the Events table does not appear for a Viewer.
  • At least one key event marked, so the exploration has a value to show. In this guide the marked event is purchase.
  • An exploration that does not already carry a key event filter. A free-form exploration built on all events mixes key events with ordinary ones.

The trace

  1. Open Explore, then start a blank free-form exploration. The free-form template is a table with rows, columns, and values that you choose yourself. Name it so it can be found again, such as Purchase by landing page and device.
  2. Add Event name as the rows dimension. In the Variables panel, add the Event name dimension, then drag it to the Rows area. The table now has one row per event.
  3. Add Key events as the value. In the Variables panel, add the Key events metric, then drag it to the Values area. The table now counts key events per event name.
  4. Filter the exploration to purchase. Add a filter on Event name, set it to exactly match purchase. Without the filter the table counts every event, and the key events metric only fills on the key-event rows.
  5. Add Device category as the columns. Drag Device category to Columns. In the worked example the filtered table shows 104 purchase key events: 62 from mobile, 31 from desktop, and 11 from tablet. The shares are 62 / 104 = 59.62% mobile, 31 / 104 = 29.81% desktop, and 11 / 104 = 10.58% tablet.
  6. Check the total against the standard reports. The exploration total must match the Events report for the same dates: 104 purchase key events. In the worked example it does, because both count the same marked event over the same 28 days.

Dead ends

DEAD END — A FREE-FORM TABLE WITH NO FILTER. An unfiltered exploration lists every event in the property, and the Key events column is empty on the ordinary rows. The table looks broken when the problem is only a missing filter.

DEAD END — USING EVENT COUNT WHERE KEY EVENTS BELONGS. The Event count metric counts every event, marked or not. Dragging it into Values instead of Key events gives 412 for the whole property, not 104 for purchase, and the device split that follows is the split for everything.

DEAD END — LOOKING FOR GOOGLE ADS CONVERSIONS IN EXPLORE. Google states that Google Ads conversions do not appear in standard reports or in Explore. A conversion count must come from Advertising, not from an exploration.

End state

The exploration shows 104 purchase key events split by device: 62 mobile, 31 desktop, and 11 tablet. Mobile holds 59.62% of the purchases, and the total matches the standard reports.

FigureValue
Purchase key events in the exploration104
Mobile purchases62
Mobile share59.62%
Desktop purchases31
Desktop share29.81%
Tablet purchases11
Tablet share10.58%

The shares are 62 / 104 = 59.62%, 31 / 104 = 29.81%, and 11 / 104 = 10.58%, all recomputed in Python 3 and rounded to two places. The device counts add to 104. The figures are a worked example, not a real property.

Sources

Checked 11 November 2026. The click paths above are the ones in Google's help pages, read 11 November 2026. The free-form exploration with dimensions, metrics, and filters, and Google Ads conversions not appearing in Explore, are as Google documents them. No step in this guide was run inside a live Analytics property, because Explore needs a Google sign-in and this site does not sign in as anyone. The figures on this page are a worked example recomputed in Python 3 with decimal arithmetic, not numbers from a real property. Google moves menu items without warning; if a path on this page no longer matches, the linked help page is the current map.