The Empty Conversions Report: Auditing a GA4 Key-Events Setup
Objective: Given a synthetic GA4 events list for an ecommerce site, apply the lesson's Step 3 process to identify which events should be marked as key events and which mistake is causing the Conversions report to look empty.
You're a marketing analyst at Peloton Interactive auditing the GA4 property for a new accessories microsite that launched a month ago.
Review the raw events list, decide which should be key events, and diagnose why the Conversions report is nearly empty despite real traffic.
Before you start
What you'll need
Free path (everything below is enough to finish)
Free for any site under 10 million hits per month, no paid tier needed
The process
1 step
Step 01 of 01
GA4 tracks many events automatically, but the Conversions report only shows events explicitly toggled 'Mark as key event.' The lesson calls this the step beginners skip most.
Below are the auto-tracked events and their 30-day counts. Which get marked as key events, and what does the near-zero purchase count actually tell you?
Procedure
- List all events firing in the last 30 days with counts
- Cross off events that are just navigation (page_view, scroll), not business outcomes
- Mark add_to_cart, begin_checkout, and purchase as key events
- Check the purchase count (2) against session count (14,800): a 0.01% rate is not a normal ecommerce conversion rate
- Check Admin > Data streams > Tagging Instructions for a second install
EVENTS, LAST 30 DAYS page_view 61,200 scroll 38,900 add_to_cart 2,140 begin_checkout 1,860 purchase 1,830 <- shown twice in raw event log per transaction purchase (reported) 2 <- what Conversions report actually shows Tagging Instructions: 2 active installs found (WordPress plugin + manual header snippet)
Healthy
purchase, add_to_cart, and begin_checkout all marked as key events, with the reported purchase count roughly matching real order volume.
Unhealthy
Zero key events marked at all, or a reported conversion count that's off from the raw event count by orders of magnitude.
What this means
Two separate problems stack here: the right events were never marked as key events, AND the tracking code is installed twice, which the lesson's own warning callout flags as the most common GA4 setup mistake.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Conversions report is empty despite real sales | Mark add_to_cart, begin_checkout, and purchase as key events in Admin > Events | 5 min |
| Raw purchase event count doesn't match real order volume | Check Tagging Instructions for a duplicate install, remove one | 30 min |
Final deliverable
A list of which events to mark as key events, plus a one-line diagnosis of the duplicate-tracking problem and the fix.
See a reference example
GA4 Audit, YETI Accessories Microsite Mark as key events: add_to_cart, begin_checkout, purchase. Do not mark: page_view, scroll, session_start (navigation, not outcomes). Diagnosis: raw purchase events (1,830) vastly exceed real order volume, and Tagging Instructions shows two active installs (Shopify's native integration plus a manually pasted snippet). Every page view, and every purchase, is being counted twice. Fix: remove the manual snippet, keep only the Shopify integration.
Success criteria
You're done when you can:
- Correctly separates navigation events from business-outcome events
- Identifies duplicate tracking code, not 'no sales happened', as the real cause of the mismatch