Analytics & Tracking

GA4 Data Thresholds: Why Rows Disappear and How to Get Complete Numbers

GA4 sometimes hides rows to protect user privacy. Learn when data thresholds apply, how to spot them and how to get complete numbers for eCommerce reporting.

TL;DR
  • GA4 applies data thresholds to hide rows that could reveal individual users, usually when user counts are small and Google signals or demographic data are involved.
  • Thresholds are shown with an orange warning icon. Totals in a report may not match the sum of the visible rows.
  • To reduce them, use longer date ranges, check your reporting identity, avoid demographic breakdowns for small segments and use the BigQuery export for detailed analysis.

What are GA4 data thresholds?

Data thresholds are a privacy feature in Google Analytics 4. When a report or exploration contains small numbers of users, GA4 may withhold some rows so that the data cannot be used to identify individuals.

For eCommerce teams, this shows up as:

  • Missing rows for small campaigns, landing pages or products.
  • Totals that do not match the sum of the visible rows.
  • Reports that look complete for a month but thin out when you look at a single day.

Thresholds are different from sampling (which estimates results from a subset of data in some explorations) and from the “(other)” row (which groups data when a report has too many unique values).

When does GA4 apply thresholds?

Thresholds are most likely when several conditions combine:

Condition Why it matters
Small number of users in the report or segment Fewer users means a higher risk of identifying individuals
Short date range Fewer users per row
Google signals enabled with a reporting identity that uses it GA4 applies extra protection to data linked to signed-in Google users
Demographic or interest dimensions Age, gender and interests are sensitive
Search terms or very granular dimensions Each row may represent very few users

Google does not publish the exact thresholds, and the system can change. Treat thresholding as something to check for rather than something you can calculate.

How do you know if a report is thresholded?

Look at the top of the report or exploration for the data quality icon:

  • A green check means the report includes all available data.
  • An orange warning means thresholds (or another limitation) were applied. Click it to see the reason.

If totals look higher than the sum of visible rows, or if a familiar row is missing, check the icon before drawing conclusions.

How can you reduce or avoid thresholds?

Use longer date ranges

More users per row means fewer rows hidden. For small campaigns, look at weeks or months rather than single days.

Check your reporting identity

In the property’s reporting identity settings, GA4 offers options such as blended, observed and device-based. Identity options that rely on Google signals are more likely to trigger thresholds. Many eCommerce teams switch to device-based identity for reporting when thresholds get in the way. The setting changes how reports are displayed, not the data collected, so you can switch back.

Avoid sensitive dimensions on small segments

Demographic and interest breakdowns are most affected. Use them on large audiences only, or remove them when you need complete numbers.

Use the BigQuery export

The GA4 BigQuery export gives you event-level data without the reporting thresholds that apply in the GA4 interface. It is the best option for detailed analysis of products, campaigns and customer journeys. It requires some SQL knowledge or a connector.

Report revenue from the store

For orders and revenue, your store platform remains the source of truth. Use GA4 for behavior and channel analysis, not as the definitive revenue number.

Worked example: a “missing” campaign in GA4

(Illustrative numbers.) A store runs a small influencer campaign with UTM tags. The influencer’s dashboard shows 1,100 clicks. In GA4’s Traffic acquisition report for the campaign week, the influencer’s row is missing, even though total sessions look normal.

Investigation:

  1. The orange icon shows thresholds were applied.
  2. The property uses a reporting identity that includes Google signals.
  3. The campaign row represents only a few hundred users on a daily view, with an age dimension added in the exploration.

Fixes:

  • Remove the age dimension and widen the range to the full month: the row appears.
  • Switch reporting identity to device-based for the analysis: more small rows appear.
  • For future creator campaigns, the team also checks results in the BigQuery export and in store orders by discount code.

What should eCommerce teams set up?

  • Decide on a reporting identity for the property and document it.
  • Add the data quality icon to your team’s checklist when reviewing reports.
  • Set up the BigQuery export early, since it does not backfill old data.
  • Build key dashboards on longer date ranges or aggregated views.
  • Keep store data as the reference for revenue and orders.

Checklist: handling GA4 data thresholds

  • [ ] Team knows how to read the data quality icon
  • [ ] Reporting identity chosen and documented
  • [ ] Device-based identity tested when thresholds block analysis
  • [ ] Demographic dimensions used only on large segments
  • [ ] Small campaigns analyzed over weeks or months, not days
  • [ ] BigQuery export enabled for detailed analysis
  • [ ] Revenue and orders reported from store data
  • [ ] Thresholds considered before concluding a campaign “did nothing”

FAQ

Do GA4 data thresholds delete my data?

No. Thresholds only hide rows in reports and explorations. The underlying data is still collected, and it is available in the BigQuery export.

Does turning off Google signals remove thresholds?

It often reduces them, because thresholds are more likely when data linked to Google signals is used. However, turning off Google signals also removes some features, such as certain demographic reports and remarketing uses. Changing the reporting identity is a less disruptive first step.

Is GA4 sampling the same as thresholding?

No. Sampling estimates results from a subset of data in some explorations. Thresholding hides rows to protect privacy. Both can make reports incomplete, and the data quality icon shows which one applies.

Turn this into a plan for your store

The Free Profit Audit scans your store, sizes your biggest profit leaks and builds a 90-day plan.

Free Profit Audit