Why Doesn’t My Data Match?

There is nothing more frustrating than looking at your Meta Ads dashboard, seeing $500 in spend, and then running a Metric Might query that says $485.

When this happens, the first instinct is to assume the connector is broken. However, Metric Might does not calculate, alter, or estimate your data. We are a direct conduit. We query the official ad platform APIs and print exactly what they return into your Google Sheet.

If your numbers don’t match, it is almost always due to one of the common causes below. Here is how to diagnose and fix them.

1. Timezone Mismatches (The Most Common Cause)

If your clicks, spend, or impressions look slightly off for a specific day, you likely have a timezone mismatch between your Ad Account and your Google Sheet.

For example, if your Google Ads account is set to EST, but your Google Sheet (or Metric Might query) defaults to UTC, the API will slice your data at different midnight cut-offs.

How to fix it:

1. Check the timezone of your specific Ad Account in the platform’s billing/settings page.
2. In Google Sheets, go to File > Settings and ensure the Spreadsheet timezone matches your Ad Account.
3. If the data source in Metric Might asks for a timezone during query building, ensure it matches the platform exactly.

2. Attribution Window Differences (The Meta/TikTok Special)

If your Spend and Clicks match perfectly, but your Purchases, Leads, or ROAS do not, you are looking at an attribution window mismatch.

Ad platform UIs often default to a specific attribution setting (e.g., Meta defaults to 7-day click and 1-day view). If your Metric Might query is requesting the API default (which might be 28-day click, or click-only), your conversion numbers will be drastically different.

How to fix it:

• Open your Ads Manager UI and check the exact attribution window applied to the columns you are looking at.
• Edit your Metric Might query and ensure you have explicitly selected that same attribution window in the query settings.

3. Data Freshness and API Delays

Ad platforms process millions of events a second. To keep their web dashboards feeling fast, platforms like Meta, Google Ads, and especially GA4 use “estimated” or “real-time” processing for the UI, while the API only serves fully processed, settled data.

• Google Analytics 4: Can take up to 24–48 hours to fully process event data. GA4’s API consistently lags behind its real-time UI by 24–48 hours.
• Meta & TikTok Ads: Spend data is usually accurate within 15–30 minutes, but conversion data can lag by several hours.

How to fix it: If you are comparing “Today’s” data in the UI to “Today’s” data in Google Sheets, they will rarely match perfectly. To test if Metric Might is pulling correctly, run a query for yesterday or last week. If historical data matches but today’s data doesn’t, you are simply seeing an API processing delay.

4. GA4 Thresholding & Sampling

If you are pulling data from Google Analytics 4 and notice that rows with very low user counts are missing, GA4 has applied Data Thresholding.

Google does this to prevent you from identifying individual users. If a specific campaign only drove 3 clicks, Google might hide that row entirely from the API to protect user privacy.

How to fix it: This is a Google-enforced limit. To reduce thresholding, try pulling broader date ranges or removing highly specific dimensions (like “City” or “Device”) from your query.

Still stuck? Do the “Raw Export” Test

If you’ve checked timezones, attribution, and processing delays, and the data still looks wrong, do this test:

1. Go to your ad platform dashboard (e.g., Facebook Ads Manager).
2. Set the exact date range, columns, and attribution window you used in Metric Might.
3. Click Export as CSV directly from the ad platform.
4. Compare that raw CSV to your Metric Might output.

99% of the time, the raw CSV export from the platform will perfectly match what Metric Might put in your Sheet. If the platform’s own CSV doesn’t match its own UI, it’s a known bug on the ad network’s side!