Analytics

GA4 Setup | Common Pitfalls That Skew Your Data

Short data retention, unfiltered internal traffic, self-referrals and missing key events quietly distort GA4 reports. The settings to check and how to fix them.

BenchMarketing editorial team Updated October 2, 2026 3 min read

Google Analytics 4 works out of the box, but several default settings and common mistakes make the numbers less useful than they should be. Most take minutes to fix.

1. Data retention left at two months

GA4 keeps user-level and event-level data for explorations for two months by default. Standard reports are not affected, but any custom exploration over a longer period will come up short. In Admin, under data retention, set it to 14 months.

2. Internal traffic counted as customers

Your team visits the site constantly. Define internal traffic by IP address in the data stream settings, then activate the internal traffic filter in data filters. Filters are created in testing mode and do nothing until activated.

3. Payment providers and your own domains showing as referrals

When customers return from a payment page or move between your domains, GA4 can start a new session credited to that referral, stealing credit from the real source. Add payment providers to the unwanted referrals list and configure cross-domain measurement for your own domains.

4. Key events not set up, or set up twice

Only events marked as key events (previously called conversions) count as conversions in reports. Mark the ones that matter. Then make sure each is triggered once: a purchase tracked by both a tag and an automatic event doubles revenue.

5. Inconsistent UTM tagging

"facebook", "Facebook" and "fb" become three different sources. Agree a naming convention and use a UTM builder so campaigns group correctly.

6. Ecommerce events without values

Purchase events need value, currency and transaction ID. Without the transaction ID, refreshes of the confirmation page can count as extra purchases.

7. Comparing GA4 with ad platforms and expecting a match

GA4, Google Ads and Meta use different attribution and count different things, so their numbers will never match exactly. Decide which tool answers which question, and look for large unexplained gaps rather than small differences.

A quick audit

Data retention set to 14 months.
Internal traffic filter active.
Unwanted referrals and cross-domain configured.
Key events marked, each firing once.
Purchases include value, currency and transaction ID.

See server-side tracking explained and bounce rate benchmarks.

About the figures

Benchmark figures in this article come from the BenchMarketing dataset and update when the benchmark pages do. Each benchmark page lists its sources and period; see our methodology.