Why SEOKit numbers can differ from the Search Console interface
Both read the same Search Console data, so a gap almost always comes from a setting that is not identical: the dates, a filter, the search type, how the data is grouped, or what Google leaves out. Six checks find it.
✓ Checked against Google Search Console Help and the SEOKit code · October 2026
SEOKit and the Search Console interface read the same data, so a gap almost always comes from a setting that is not identical: the dates, a filter, the search type, how the data is grouped, or something Google leaves out. Make the period, filters and search type the same in both places, compare the same kind of total, and what remains is usually documented by Google itself (troubleshooting data discrepancies).
Start with the dates
Search Console opens on the past three months, while SEOKit windows of 7, 28, 90 or 180 days end at the newest available date. A 28-day window set against a three-month default is not a like-for-like comparison, so set the same start and end date on both sides (Performance report overview). Daily data is labelled in Pacific Time, so a day can shift against Analytics or another system.
Freshness matters too. The newest data can be preliminary, and collected data is usually available after two to three days. SEOKit shows a delay label next to the period; if yesterday’s number moved overnight, it was preliminary and has been updated. Reopen the report later and compare complete days.
Then the filters and the search type
SEOKit’s default filters are applied to your reports automatically, including a branded or non-branded split if you set one, and the report does not show a banner for them. Open Default Filters in the account menu and compare them with the filters on the Search Console report. Do the same for the search type: Search Console can show web (text-based or multimodal), image, video or news, and both sides must use the same one.
The totals Google itself does not make equal
Even with identical settings, three documented effects make numbers disagree. Adding a page or Search appearance filter changes how data is aggregated, so chart totals can rise or, rarely, fall. Anonymized, rare queries are left out of the query table but counted in the chart totals unless you filter by query, so a sum of the visible rows is lower than the total. And the table shows at most 1,000 rows. Compare chart with chart and table with table, and do not expect “matches” and “does not match” totals to add up to the unfiltered one.
That is why a SEOKit total can be higher than the sum of its query rows, and why Search Console can list fewer queries: its table is capped at 1,000 rows, while SEOKit loads more through the API, up to what the API returns (getting all your data).
When a gap remains
SEOKit reads the Search Console API. If the two still disagree after these checks, treat the Search Console figure as the reference and use the gap to find which setting differs. If the dates, filters, search type and totals are identical and a gap is left, send the property, the exact dates, the filters and a screenshot of both reports to the contact address in the privacy policy. The documentation on dates, comparisons and data delay and on default filters and exports describes the SEOKit side, and the Search Console and Analytics article explains why a different tool disagrees.
Go further
Docs: dates, comparisons and data delay: Periods, comparison modes and freshness labels. Docs: default filters and exports: What default filters do and what an export contains. FAQ: Search Console vs Google Analytics: Why a different tool disagrees.