When you import Google Analytics (GA4) data into Crazy Egg Web Analytics, you may notice that some numbers don’t exactly match what you see inside the Google Analytics interface.
The differences come down to how data is defined, grouped, and displayed, not missing or incorrect data.
This article explains the most common reasons for these differences and how to accurately compare the two tools.
Crazy Egg Allows You to Import and use Google’s Data
Crazy Egg does not estimate, model, or generate its own analytics data. All Web Analytics metrics that come from Google Analytics are pulled directly from GA via Google’s official APIs.
If numbers differ between interfaces, it’s usually because:
- The same data is being grouped or summarized differently
- Different dimensions or attribution rules are being used
- Data is still processing or syncing in GA
1. Totals vs. Row-Level Breakdowns
One of the most common points of confusion is comparing:
- The total number shown at the top of a report
vs. - The sum of individual rows (channels, pages, sources, etc.)
In Google Analytics, totals are often calculated independently of the rows you see below them. This means:
- The total is not always meant to equal the visible rows
- Some events or sessions may be grouped into categories that aren’t displayed
- GA may apply attribution or filtering rules at the total level that don’t apply evenly to each row
In Crazy Egg Web Analytics:
- Metrics are presented using consistent aggregation
- If you add up the rows in a table, they are designed to match the displayed total for that report view
This difference can make Crazy Egg’s reports feel more predictable and easier to reason about, especially when validating numbers manually.
2. You Must Compare the Same Dimension (This Is Critical)
GA4 contains many similarly named dimensions that sound alike but mean very different things.
For example, GA4 includes 15+ dimensions containing the word “medium.”
Crazy Egg uses the following dimensions when reporting traffic sources:
firstUserManualMedium
The manual medium that originally acquired the user.
Derived from the utm_medium parameter (e.g. cpc, email, social).
This means:
- Crazy Egg answers the question:
“What marketing medium originally brought this user to the site?” - This is stable, user-focused attribution.
By contrast, GA4’s default Medium dimension is often defined as:
The medium is attributed to the key event.
That definition can change depending on:
- Which event are you viewing
- How GA attributes conversions
- Session vs. event-level rules
If you compare Crazy Egg’s numbers to a GA4 report using a different medium definition, the numbers may look inconsistent, even though both are technically correct.
Troubleshooting tip:
If numbers don’t match, always confirm:
- Same date range
- Same dimension
- Same scope (user, session, or event)
3. GA4 Attribution Can Shift Between Rows and Totals
GA4 applies attribution logic that can:
- Reassign credit between channels
- Adjust totals based on conversions or key events
- Display rounded or thresholded values in some reports
As a result:
- Individual rows may add up differently than expected
- Totals may reflect attribution logic not visible in the table
- The same data may look slightly different across GA reports
Crazy Egg avoids re-attributing GA data and instead presents it with a clear, fixed definition per report, helping reduce confusion when interpreting results.
4. Sync Timing and Data Freshness
Another common reason for small differences is timing.
How Crazy Egg syncs GA data
- Initial import:
Pulls the last 6 months of historical data once - Current day:
Synced hourly (intraday data may still change) - Completed days:
Synced once per day after midnight, once GA considers the day final
Important to know
- Google Analytics data is not real-time
- Intraday data can change as GA processes events
- Yesterday’s numbers are usually stable; today’s numbers may still move
If you compare Crazy Egg to GA while a day is still in progress, small differences are expected and typically resolve after GA finalizes the data.
5. Pageviews, Sessions, and “Uniqueness.”
Some metrics, like pageviews, are not inherently unique. Others, like users or sessions, depend heavily on GA’s internal rules.
GA4 may:
- Deduplicate or threshold data in certain views
- Apply sampling or privacy modelling
- Present event-based metrics differently depending on the report
Crazy Egg displays these metrics using the definitions returned by GA, but organizes them in a way that prioritizes:
- Consistent totals
- Clear breakdowns
- Easier comparison across days and sources
Summary
- Crazy Egg uses Google’s data
- Differences usually come from definitions, attribution, or timing
- Web Analytics is designed to make GA data simpler, clearer, and more consistent
- When in doubt, compare like-for-like metrics—not just labels