Common Issues
Here are the most common issues you might run into when working with the Google Analytics 4 source in Coupler.io, along with how to fix them.
Don't confuse GA4 with Universal Analytics — GA4 uses an event-based model, not the session-based model from Universal Analytics. Metrics like "conversions" are now called "key events" and are counted differently.
Connection issues
OAuth token expired
Google access tokens can expire or be revoked. If your imports start failing, go to the Sources page in Coupler.io and reconnect your Google account.
Wrong Google account
If you signed in with a personal Google account that doesn't have access to your GA4 property, disconnect and reconnect using the correct account.
Permissions not granted
Make sure you granted the analytics.readonly permission during the OAuth flow. If you skipped it, disconnect and reconnect your Google account.
Internal Server Error when loading metrics or dimensions
If you see "Something went wrong; Internal Server Error" when selecting metrics and dimensions, this is usually a temporary issue on Google's API side. Wait a few minutes and try again. If it persists, check whether the GA4 property itself is accessible in the GA4 interface and that the selected metrics/dimensions are compatible.
Missing data or properties
GA4 property not showing up
You need at least Viewer access to the GA4 property. If you were recently granted access, try disconnecting and reconnecting your Google credentials. Also make sure you're looking for a GA4 property, not a Universal Analytics property.
No data returned
Check that the selected property has been collecting data for the date range you specified. A new property or one with a misconfigured tracking tag may have no data.
Missing recent data
GA4 data processing can take a few hours. Some metrics (especially those tied to Google Ads or Search Console) may take up to 24–48 hours to appear. Avoid using today's date as your end date.
Rows missing due to thresholding
If Google Signals is enabled on your property, GA4 may apply data thresholds and withhold rows that represent too few users. Adding high-cardinality dimensions like Landing page + query string makes thresholding more aggressive. Check your GA4 property settings or disable Google Signals if this is causing issues.
Thresholding is one of the most overlooked causes of "missing" data. If you see gaps, check this first.
Missing dates in export
If your export has gaps in dates, check two things: (1) your account's row limit — on the free plan, only the first 1,000 rows are exported, and (2) your date range — a very long range with multiple dimensions can exceed row limits. Try narrowing the date range or reducing the number of dimensions.
Historical data not available
GA4 properties have a configurable data retention period (2 or 14 months). Coupler.io can only pull data within this retention window. If you need older data, export it to BigQuery or another destination before it expires.
Permission errors
"Insufficient permissions" error
Your Google account must have at least Viewer access to the GA4 property. Check your access in Google Analytics > Admin > Property Access Management.
Property access revoked
If someone removed your access to a GA4 property, imports for that property will fail. Contact the property owner to restore access.
Data discrepancies
Numbers don't match the GA4 interface
The GA4 web interface may apply sampling or filters that differ from the API. Coupler.io pulls unsampled data via the GA4 Data API, which can produce slightly different numbers. To compare accurately:
Make sure the date range, dimensions, and metrics match exactly between GA4 and your data flow
Verify the GA4 property timezone matches your API request settings
Check that no filters are applied in the GA4 interface that aren't in the data flow
Remove any dimensions you're not using — extra dimensions change how rows are grouped
Users total doesn't add up across dimensions
When you break down Total users by a dimension like Default channel group, the sum of users across all groups will be higher than the overall total. This is because a single user can appear in multiple channel groups across different sessions. This is standard GA4 behavior — users are deduplicated at the total level but counted per-dimension when broken down.
This is the same behavior you see in the GA4 interface. If you sum users in a downloaded CSV, the total will exceed the header total. Do not use SUM for user counts across dimensions — use the total row provided by the report instead.
Metrics show "(not set)"
This happens when GA4 doesn't have data for a specific dimension value. For example, if UTM parameters aren't configured on some traffic, the campaign dimension will show "(not set)" for those sessions.
"(not set)" is standard GA4 behavior, not a Coupler.io issue.
Key events count differs
In GA4, "key events" (formerly conversions) are counted per event, not per session. If you're comparing to a tool that counts per session, the numbers will differ.
Incompatible dimensions and metrics
"The request's dimensions & metrics are incompatible" error
Not all GA4 metrics and dimensions can be used together. For example, combining Referrer with Ads clicks, or Item name with Advertiser ad clicks, will trigger a 400 error. This is a GA4 API limitation.
To fix this, remove one of the conflicting metrics or dimensions, then try again. Use the GA4 Dimensions & Metrics Explorer to check compatibility before configuring your flow. If you need both incompatible fields, create separate data flows and combine them using a data join.
This is the most common GA4 error in Coupler.io. Always verify your metric–dimension combination before running the data flow.
"Metric names must only contain letters, numbers, or _" error
This error occurs when you select a key event whose name contains special characters (spaces, hyphens, brackets, or non-Latin characters). GA4 sends the metric key as-is from the API. The only fix is to rename the key event in your GA4 property settings so it only uses letters (A–Z), numbers (0–9), and underscores.
Deprecated metrics
"Conversions" metrics stopped returning data
Google deprecated all "Conversions"-related metrics in GA4 and replaced them with "Key events" equivalents. If your data flow still uses the old metric names, update them:
Conversions
Key events
Cost per conversion
Cost per key event
Session conversion rate
Session key event rate
User conversion rate
User key event rate
First-time purchaser conversion rate
First-time purchaser rate
Purchaser conversion rate
Purchaser rate
After updating metrics, also update any Transformations (joins, unions, formulas, filters) that reference the old column names. If you use BigQuery with Append mode, you may need to create a new table — BigQuery does not allow schema changes on auto-detected tables.
Rate limits
Last updated
Was this helpful?