Discover Report (Google Search Console)

TL;DR — The Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold. in Google Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. shows how your content did in the Google DiscoverGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses. feed — that personalized stream of articles you see in the Google app and on Chrome’s New Tab page. It tracks three numbers: impressions (how often your content was shown), clicks, and CTR. If you don’t see a Discover tab at all, that’s usually not a bug — it only appears once your site gets enough Discover traffic. And it’s a different report from the Google News one.

What this report is

Google Discover is the feed of articles and videos Google surfaces to people based on their interests — no search query involved, it just gets pushed to them. If your content shows up there, the Discover report is where Search ConsoleGoogle's free tool for monitoring crawling, indexing, and search performance. lets you see how it did.

Google’s own name for it is the “Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. (Discover).” “Discover report” is just the common shorthand. You’ll find it in the left menu of Search Console under Performance, as a separate tab next to Search results (that’s regular Google Search) and Google News.

Google describes it plainly: “The Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. for Discover shows important metrics about how your site performs on Discover.”

The three numbers it shows

• Impressions — how many times your content was seen in the Discover feedGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses..
• Clicks — how many times people clicked through to your site.
• Average CTR — clicks divided by impressions.

That’s it. There’s no “average position” here like there is for regular Search, because in Discover there’s nothing to rank for — content just gets served into a scrolling feed, there’s no page 1 or position 3.

Why you might not see a Discover tab

This is the number-one point of confusion, so let me clear it up right away. The Discover tab only appears once your site has received enough Discover traffic. Google says the report is “visible only if your property has reached a minimum number of impressions in Discover.”

So if you don’t see it, it almost certainly means your site hasn’t gotten meaningful Discover traffic yet — not that something is misconfigured. Most small sites and most non-news, non-publisher sites never see this tab, and that’s completely normal.

Discover ≠ Google News

The other big mix-up: people assume the Discover report and the Google News report are the same thing. They’re not. Discover is the personalized feed. Google News is a separate product — news.google.com and the Google News mobile apps — with its own separate report. If you want to understand which content Google decides to push into Discover in the first place, that’s a topic in its own right (Google Discover SEO), and closely related to how news content and content freshnessContent freshness is how recent or up-to-date a page is — by its original publish date, its last substantive revision, or the currency of the facts inside it. It only helps rankings when the query itself benefits from recent results (Query Deserves Freshness), and cosmetic date changes with no real update don't count. get treated. This report is just about reading the numbers.

Want the full picture — the stricter way Discover counts impressions, the four “appearance types,” why the numbers cap at 1,000 rows, and how to tell normal volatility from an actual Google bug? Switch to the Advanced tab.

TL;DR — The Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold. (“Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. (Discover)”) is Search Console’s dedicated view into Google Discover feedGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses. performance — impressions, clicks, and average CTR, sliced by page (canonical URLHow search engines pick one canonical URL among duplicates and consolidate signals onto it.), country, date, and Discover appearance type (AMPAMP (Accelerated Mobile Pages) is an open-source web framework Google launched in 2015 to make mobile pages load near-instantly via restricted HTML/CSS/JS and CDN caching. It was never a ranking factor and, since June 2021, is no longer required for Top Stories. articles, News Showcase, Web Stories, Videos). Read it knowing: the tab only appears once you cross a minimum Discover-impression threshold, and low-volume rows are separately suppressed; there’s no query or device dimension because Discover isn’t query-driven; there’s no average position because nothing “ranks”; impressions are counted more strictly than in Web Search (scroll-into-view, one per session); and it’s a wholly separate report from the Google News report. The same data ladder as Web Search applies — 16 months, a 1,000-row UI cap, a 25,000-row API cap, and the BigQuery bulk export (Discover rows ship in the same tables, tagged by search type). The chart is volatile by design, but Google had two real Discover reporting bugs in 2026, so verify a sharp move against known incidents before you assume you broke something.

Where it lives and the visibility gate

The Discover report sits under Performance in the Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. left nav, as a distinct tab alongside Search results (Web Search) and Google News. Google’s description is direct: “The Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. for Discover shows important metrics about how your site performs on Discover.”

The critical structural fact is the visibility gate. The tab doesn’t exist for every property: “This report is visible only if your property has reached a minimum number of impressions in Discover.” No meaningful Discover traffic, no tab — and that’s expected behavior, not a setup error. This is the same eligibility framing that governs whether you get Discover traffic at all (which lives in the Google Discover SEO side of things); here it just governs whether the report appears.

There’s a second, separate threshold operating at the row level, even once you have the tab: “A data row will be shown only if it has a minimum threshold of impressions in Discover.” So a page (or country, or date) that got only a trickle of Discover impressions can be missing from the table even while the report itself is fully populated. Two thresholds, two different jobs — don’t conflate “I don’t have the report” with “this specific row is missing.”

Google doesn’t publish the actual numbers behind either threshold, and that’s deliberate — treat “why don’t I have a tab/row” as a documented behavior with an undisclosed cutoff, not a bug to troubleshoot. It also means a missing tab or a suppressed row is not proof you got zero Discover exposure: rows sitting below the row-level threshold can still contribute to the report’s aggregate totals even though they don’t appear individually in the table.

The three metrics — and why they’re counted more strictly

The report shows exactly three metrics — impressions, clicks, and average CTR — over the last 16 months. From Google’s Discover documentation: “This report shows impressions, clicks, and CTR for any of your content that has appeared on Discover in the last 16 months, as long as your data reaches a minimum threshold of impressions.”

There is no average position, and this is not an oversight. Discover isn’t a ranked results page; content is proactively served into a feed, so there’s no position to report. That single absence tells you most of what makes Discover structurally different from Web Search.

The impression and click definitions are also narrower than Web Search’s, and this catches people out when they try to reconcile numbers across reports:

• Impression: “The Discover item was scrolled into view, either as a standard Discover item, or embedded in a carousel.” Compare that to Web Search, where a result counts as an impression when it’s seen or potentially seen — a static SERP result can count even if the user never scrolls to it (see the Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. deep dive). In Discover, it genuinely has to scroll into view, and it counts once per result per session. That’s a stricter bar, so don’t expect Discover impressions to behave like Search impressions.
• Click: “The user clicked the Discover item. A click is not counted if the user shares the item, or performs any another action.” Shares, saves, and other interactions don’t inflate your click count — only an actual click-through does.

One more coverage detail worth flagging, because people assume Discover reporting is Google-app-only: “The Discover performance report includes traffic from Chrome, and fully tracks a site’s Discover traffic across all surfaces where users interact with Discover.” That includes the Chrome New Tab surface, not just the Google app feed.

Appearance types — the Discover-only dimension

Discover has one dimension you won’t find in Web Search: appearance type, the format category Google used to surface your content. There are four:

• AMPAMP (Accelerated Mobile Pages) is an open-source web framework Google launched in 2015 to make mobile pages load near-instantly via restricted HTML/CSS/JS and CDN caching. It was never a ranking factor and, since June 2021, is no longer required for Top Stories. articles — content served via AMP.
• News Showcase — items presented through Google’s News Showcase panels.
• Web Stories — the tappable, full-screen Web Story format.
• Videos — video content surfaced in the feed.

These are format categories, not quality or ranking tiers. An item being a “Web Story” versus an “AMP article” tells you how it appeared, not that one is better or higher-priority than another. Use this dimension to understand which formats are actually earning you Discover reach — not as a scorecard.

Dimensions, filters, and what’s deliberately missing

You can group and filter the table by a specific set of dimensions: “You can filter data by date, page, country, or Discover appearance type.” That enumeration is itself the proof of the biggest structural point about this report — query is not on the list, and neither is device.

• No query dimension. There’s no keyword breakdown because Discover isn’t query-driven. As the Google DiscoverGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses. SEO side of this site puts it, “Discover is not a response to a query” — and consequently “there is nothing to rank for in the traditional sense — no keyword, no position, no SERP.” There’s no query table to show because there was never a query. This is the load-bearing difference between the Discover report and every other Performance tab.
• No device dimension. Notably, the Google News report does break down by device — Discover does not. If you’re comparing the two reports, that’s a real difference, not a bug.

A few grouping and filtering mechanics carry straight over from the Web Search report:

• Page grouping is assigned to the canonical URL, not the landing/clicked URL — the same aggregation quirk as Web Search. If you have duplicate or non-canonical versions of a page, the metrics roll up to the canonical, so a non-canonical URL can show zero even if it’s technically the one that got surfaced. When numbers look “missing” per URL, canonicalizationHow search engines pick one canonical URL among duplicates and consolidate signals onto it. is the first thing to check.
• URL filters are case-insensitive except “Exact URL,” which is case-sensitive, and RE2 regex filtering is supported — the same engine as the Web Search filters.
• Preliminary data for the most recent day(s) is drawn with a dotted line on the chart, same convention as Web Search — don’t read a fresh, incomplete day as a drop.
• Default view and granularity (as of this writing): the report defaults to the last three months, the same default the Google News report documents. There’s also a 24-hour view with hourly preliminary data, and you can switch the chart’s time granularity between hourly, daily, weekly, and monthly. Treat these as the current UI state rather than a permanent limit — Search ConsoleGoogle's free tool for monitoring crawling, indexing, and search performance. has changed this surface before and will again.

The data ladder: 16 months, 1,000 rows, API, BigQuery

The Discover report is bound by the same data-access ladder I’ve documented for the Web Search Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data., so I won’t re-derive it — I’ll just note where Discover slots in:

• 16-month retention in the UI, matching Web Search.
• The table is capped at 1,000 rows in the UI. If you have more Discover pages than that, the UI simply won’t show them all.
• The Search Analytics APIA set of REST APIs that let you programmatically read and manage Google Search Console data for properties you've verified — Performance data, URL index status, sitemaps, and properties — authorized with OAuth 2.0. returns up to 25,000 rows per request — the first escape hatch past the UI cap.
• The BigQuery bulk export is the second. And here’s a genuinely useful, settled detail: Discover data ships in the standard bulk export, not a separate one. In the export, the search_type field takes the value discover, and the searchdata_url_impression table carries an is_anonymized_discover field — “Whether the data row is under the Discover anonymization threshold. When under the threshold, some other fields (like URL and country) will be missing to protect user privacy.” So Discover rows sit in the same tables as Web Search rows, distinguished by search_type.

I’ve spent a fair amount of time on how GSC’s Performance data gets filtered and undercounted — my study of GSC’s anonymized “hidden” query terms foundA 302 (\"Found\") is a temporary redirect: it forwards users to a new URL while telling search engines the original URL should stay in the index. It's a weak canonicalization signal, not the zero-equity dead end of SEO folklore. a huge share of query data gets withheld. That specific finding doesn’t transfer to Discover, because Discover has no query dimension to anonymize in the first place. What does transfer is the mental model: the UI is a sampled, capped, privacy-filtered view, and the API and BigQuery export are how you get closer to the complete picture.

Discover vs. Google News vs. the News tab in Search

This is the single most valuable thing to get right, because most write-ups blur it. There are three genuinely different “news-adjacent” reporting surfaces, and they are not interchangeable:

Google is explicit that the Google News report is its own thing: “The Performance report for Google News shows data from news.google.com, and from the Google News app on Android and iOS.” And critically, “It does not include the ‘News’ tab in Google Search, which is covered in the Performance report for Search, filtered to the News search type.”

A couple of things distinguish the Google News report from the Discover one, beyond the surfaces they cover:

• The Google News report defaults to the last 3 months in the date picker — the same default the Discover report itself now documents (see the data-ladder section above), so that’s no longer a distinguishing detail between the two.
• It has a device dimension and a “Google News appearance” grouping — Discover has neither.

So: three surfaces, three reports, and the News tab you see inside a regular Google search is a fourth thing that’s actually part of the Web Search report. If your content is a mix of evergreen and news, you may legitimately need to read all three. None of the three give you a query or position/ranking concept for Discover specifically — that comparison only exists inside the Web Search report and the News-tab filter of it, since Discover and Google News are both proactively-served feeds with no query behind the impression.

Reading the chart: normal volatility vs. an actual bug

Discover traffic swings around a lot more than Search traffic, and that’s by design, not a symptom of a problem. Google says so directly: “Discover traffic is somewhat less predictable than Search traffic, in part because it is proactively served by Google rather than dependent on user queries. Therefore you might see more variation in your Discover traffic than in your Search traffic.”

That volatility is exactly why Discover reps have long warned it can appear and disappear unpredictably — a spiky chart alone is not evidence that you broke something. Before you go hunting for a content or eligibility problem, the first question is whether the move lines up with a known Google-side reporting incident. In 2026 there were two real ones, and they’re worth knowing so you can tell them apart from your own changes:

• May 7–8, 2026 — a two-day Discover logging bug. Per Search Engine Land, Google described it as “a logging error caused a decrease in reported clicks and impressions for the Discover report during the two-day window” — and stressed it was a data-logging issue only, with “your positioning in Google Discover was not impacted.” TechWyse corroborated the account and suggested the sensible remediation: annotate the reporting period and exclude those dates from short-term Discover comparisons where you can.
• June 24, 2026 — a separate Gen-AI-in-Discover logging bug. Per Search Engine Roundtable, Google’s notice read: “A logging error caused a decrease in clicks and impressions on the Discover performance report for data on June 24, 2026,” and “For properties with access to Generative AI features in Discover, this error also caused a decrease in reported impressions.” This one hit both the standard Discover report and the new Gen-AI-in-Discover report at once, with the AI report showing an impressions drop only (it doesn’t report clicks).

The takeaway isn’t “don’t trust the report” — it’s “before you panic at a sharp move, check Search Central’s status/blog channels and the SEO news for an announced incident.” Normal Discover volatility and a genuine logging bug can look identical on the chart; only the incident record tells you which one you’re looking at.

What’s new: Gen-AI performance reporting in Discover (2026)

As of mid-2026, Google began extending Search Console’s Generative AI performance reporting to Discover, so some properties now have a separate Generative AI performance report (Discover) alongside the standard one — that’s the surface the June 24 bug above touched. As documented, it’s narrower than the standard report:

• It’s a subset-of-properties rollout, with its own separate access and impression thresholds — having the standard Discover report doesn’t mean you have this one.
• It reports organic impressions only, grouped by canonical page, country, and date — there’s no clicks, CTR, query, or position data in it at all.
• The same 1,000-row and time-period limits that apply to the standard Discover report apply here too.
• It excludes Search Labs experiments from its numbers.

This area is actively evolving, and I’d re-verify the exact shape of it against Google’s own current documentation before leaning on any specific detail beyond what’s above — Google was still changing this surface month to month in 2026. Treat it as an emerging report to watch, not settled ground, and treat any specific incident chronology (like the June 24 bug above) as reported through industry coverage rather than confirmed against a primary Google status page.

How this connects to Discover SEO

Everything above is about reading the report. What actually gets your content into Discover in the first place — the eligibility signals, image requirements, content policies, and why traffic spikes and vanishes — is a separate topic (Google Discover SEO on this site owns it, and it’s tightly bound up with content freshnessContent freshness is how recent or up-to-date a page is — by its original publish date, its last substantive revision, or the currency of the facts inside it. It only helps rankings when the query itself benefits from recent results (Query Deserves Freshness), and cosmetic date changes with no real update don't count. and, for publishers, news SEO). If your question is “why did my Discover numbers move,” start there for the why; this report is the what.

One boundary worth stating plainly here, since it’s easy to over-read the report as a scorecard for your optimization work: indexed content that meets Discover’s content policies is automatically eligible without any special tag or markup, and Google’s image guidance (large, compelling images — at minimum 1200px wide, with max-image-preview:large or AMP) is a real recommendation. But eligibility, correct image sizing, and policy compliance are not a guarantee — they don’t guarantee your content actually appears in the feed, gets stable traffic, or performs any particular way in the numbers above. Discover distribution stays proactive and discretionary on Google’s side.

Bing note

There’s no direct Bing equivalent. Microsoft doesn’t run a Discover-style feed inside Bing Webmaster ToolsMicrosoft's free portal for monitoring and improving how a site appears in Bing search — the peer to Google Search Console, plus IndexNow instant indexing, richer backlink data, and keyword volumes. Because Bing's index also feeds Microsoft Copilot, it doubles as a window into AI-search visibility., and there’s no comparable “feed performance” reporting tab. The closest analog on the content/publisher side is Microsoft Start (the MSN and Edge New Tab feed), which has its own separate Microsoft Start Partner Hub / publisher dashboard rather than anything inside Bing Webmaster Tools. Don’t go looking for a Discover-report parity feature in Bing — it isn’t there.

AI summary

A condensed take on the Advanced version:

• What it is: Google Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results.’s view into Google DiscoverGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses. feed performance — Google’s own label is “Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. (Discover).” It lives under Performance, as a tab next to Search results and Google News.
• Three metrics only: impressions, clicks, average CTR — over the last 16 months. No average position (nothing ranks in a feed).
• The tab only appears above a minimum Discover-impression threshold, and Google doesn’t publish the number. No meaningful Discover traffic → no tab, and that’s normal. A separate, also-undisclosed row-level threshold can hide individual low-volume rows — but a suppressed row can still count toward the report’s totals, so a missing row isn’t proof of zero exposure.
• Default view is the last 3 months, with a 24-hour view showing hourly preliminary data, and chart granularity switchable between hourly/daily/weekly/ monthly — the same default the Google News report uses.
• Stricter counting than Web Search: an impression requires scroll-into-view (once per session); a share is not a click. Coverage includes Chrome New Tab traffic, not just the Google app.
• Four appearance types: AMPAMP (Accelerated Mobile Pages) is an open-source web framework Google launched in 2015 to make mobile pages load near-instantly via restricted HTML/CSS/JS and CDN caching. It was never a ranking factor and, since June 2021, is no longer required for Top Stories. articles, News Showcase, Web Stories, Videos — format categories, not quality tiers.
• No query dimension and no device dimension — Discover isn’t query-driven, so there’s genuinely nothing else to slice by (unlike Google News, which does have device). Page metrics roll up to the canonical URLHow search engines pick one canonical URL among duplicates and consolidate signals onto it., not the clicked URL.
• Data ladder mirrors Web Search: 16 months, 1,000-row UI cap, 25,000-row API cap, and the BigQuery bulk export (Discover ships in the same tables, tagged search_type: discover, with an is_anonymized_discover field).
• Discover ≠ Google News ≠ the News tabGoogle News SEO is the practice of getting eligible for and ranking well in Google's news-specific surfaces — the News tab of Search and the Top Stories carousel. As of the 2024–2025 Publisher Center transition there's no application to file: content that complies with Google's news content policies is automatically eligible, and ranking within that pool is driven by relevance, prominence, authoritativeness, freshness, usability, and location/language. in Search — three separate surfaces. Google News covers news.google.com + the News apps only; the News tab in Search is part of the Web Search report (filtered to the News search type).
• Chart volatility is expected (Discover is proactively served) — but Google had two real Discover reporting bugs in 2026 (May 7–8 logging error; a separate June 24 Gen-AI-Discover bug), so check for a known incident before assuming you broke something.
• New in 2026: a separate Generative AI performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. (Discover) is rolling out to a subset of properties, with its own access/impression thresholds. It’s impressions-only — no clicks, CTR, query, or position — grouped by canonical page, country, and date, still under the same 1,000-row/time-period caps, and it excludes Search Labs experiments. It’s evolving, so re-verify specifics against Google’s docs, and treat reported incident chronology (like the June 24 bug) as industry coverage, not a primary Google confirmation.
• Eligibility and image guidance are recommendations, not guarantees. IndexedStoring a crawled page in the search index so it can appear in results. Crawled is not the same as indexed — Google selects what to keep, and indexing isn't guaranteed., policy-compliant content with Google’s image guidance (1200px+ width, max-image-preview:large or AMPAMP (Accelerated Mobile Pages) is an open-source web framework Google launched in 2015 to make mobile pages load near-instantly via restricted HTML/CSS/JS and CDN caching. It was never a ranking factor and, since June 2021, is no longer required for Top Stories.) is automatically eligible — but none of that guarantees actual appearance, stable traffic, or any specific number in this report.

Official documentation

Primary-source documentation from Google.

Google — the reports themselves

• Performance report (Discover) — the canonical Help doc: metrics, impression/click definitions, appearance types, the visibility and row thresholds, filters, and the volatility note.
• Performance report (Google News) — the separate Google News report; the disambiguation source (what it covers, the device dimension, the 3-month default).
• Discover and your website — the 16-month retention line and the Chrome-traffic-inclusion line; also the source Google’s own Discover eligibility guidance is built on.
• Search Console bulk data export — confirms Discover rows are included in the standard BigQuery export via the search_type: discover value and the is_anonymized_discover field.

Google — background and announcements

• Search Console reporting for your site’s Discover performance data (April 2019) — the original launch announcement for the Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold..
• Search Console Discover report now includes Chrome data (Feb 2021) — the dedicated announcement for Chrome New Tab traffic being folded into the report.
• Introducing Search Generative AI performance reports in Search Console (June 2026) — the rollout post for Gen-AI performance reporting, which extends to Discover for a subset of properties.

Quotes from the source

On-the-record statements. Each Google Help/docs link is a deep link that jumps to the quoted passage on the source page.

Google — what the report is and the visibility gate

• “The Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. for Discover shows important metrics about how your site performs on Discover.” — Google Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. Help. Jump to quote
• “This report is visible only if your property has reached a minimum number of impressions in Discover.” Jump to quote
• “A data row will be shown only if it has a minimum threshold of impressions in Discover.” Jump to quote

Google — retention and Chrome coverage

• “This report shows impressions, clicks, and CTR for any of your content that has appeared on Discover in the last 16 months, as long as your data reaches a minimum threshold of impressions.” — Google Search Central, Discover and your website. Jump to quote
• “The Discover performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. includes traffic from Chrome, and fully tracks a site’s Discover traffic across all surfaces where users interact with Discover.” Jump to quote

Google — the stricter metric definitions

• “The Discover item was scrolled into view, either as a standard Discover item, or embedded in a carousel.” (the impression definition) Jump to quote
• “The user clicked the Discover item. A click is not counted if the user shares the item, or performs any another action.” (the click definition) Jump to quote

Google — filters and volatility

• “You can filter data by date, page, country, or Discover appearance type.” Jump to quote
• “Discover traffic is somewhat less predictable than Search traffic, in part because it is proactively served by Google rather than dependent on user queries. Therefore you might see more variation in your Discover traffic than in your Search traffic.” Jump to quote

Google — the separate Google News report (do not conflate)

• “The Performance report for Google News shows data from news.google.com, and from the Google News app on Android and iOS.” — Google Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. Help. Jump to quote
• “It does not include the ‘News’ tab in Google Search, which is covered in the Performance report for Search, filtered to the News search type.” Jump to quote
• “The default date period is the last 3 months.” (the Google News report’s default range) Jump to quote

Google — Discover in the BigQuery bulk export

• “Whether the data row is under the Discover anonymization threshold. When under the threshold, some other fields (like URL and country) will be missing to protect user privacy.” (the is_anonymized_discover field) — Google Search ConsoleGoogle's free tool for monitoring crawling, indexing, and search performance. Help, Search Console bulk data exportGSC BigQuery export (bulk data export) is a Google Search Console feature that schedules a daily, unsampled export of a property's Performance data into a Google Cloud BigQuery project — bypassing the UI's ~1,000-row export cap and its rolling ~16-month retention window. It doesn't backfill and still excludes anonymized queries at the query level.. Jump to quote

Industry coverage — the 2026 reporting bugs (relayed)

• On the May 7–8, 2026 bug, Google said “a logging error caused a decrease in reported clicks and impressions for the Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold. during the two-day window,” and that “your positioning in Google DiscoverGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses. was not impacted.” Search Engine Land
• On the June 24, 2026 bug, Google’s notice read “A logging error caused a decrease in clicks and impressions on the Discover performance report for data on June 24, 2026,” and “For properties with access to Generative AI features in Discover, this error also caused a decrease in reported impressions.” Search Engine Roundtable

Which path should I take?

“I don’t see a Discover tab in Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. — where is it?”

“My Discover clicks/impressions dropped suddenly — what do I do?”

“Am I looking at the right report — Discover, Google News, or the News tabGoogle News SEO is the practice of getting eligible for and ranking well in Google's news-specific surfaces — the News tab of Search and the Top Stories carousel. As of the 2024–2025 Publisher Center transition there's no application to file: content that complies with Google's news content policies is automatically eligible, and ranking within that pool is driven by relevance, prominence, authoritativeness, freshness, usability, and location/language. in Search?”

Common mistakes with the Discover report

“Every site has a Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold. — mine is missing, so something’s broken.” Why it’s wrong: the tab is “visible only if your property has reached a minimum number of impressions in Discover.” Most low-traffic and non-publisher sites never cross that threshold and never see the tab. Do instead: treat a missing tab as “no meaningful Discover traffic yet,” not a configuration error — and if you want the traffic, work on Discover eligibility, not Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. settings.

“The Discover report and the Google News report are the same thing.” Why it’s wrong: they’re genuinely separate reports for separate Google surfaces. Discover is the personalized feed; the Google News report shows “data from news.google.com, and from the Google News app on Android and iOS” only. Do instead: pick the report by the surface the traffic came from — and remember the News tabGoogle News SEO is the practice of getting eligible for and ranking well in Google's news-specific surfaces — the News tab of Search and the Top Stories carousel. As of the 2024–2025 Publisher Center transition there's no application to file: content that complies with Google's news content policies is automatically eligible, and ranking within that pool is driven by relevance, prominence, authoritativeness, freshness, usability, and location/language. you see inside a regular Google search is a third thing, tracked in the Web Search report filtered to the News search type.

“A volatile Discover chart means I broke something.” Why it’s wrong: Google says “Discover traffic is somewhat less predictable than Search traffic” because it’s proactively served, so “you might see more variation.” Variance is the default state of this chart. Do instead: only escalate to “check for a Google-side bug” when a sharp move matches an announced incident (like the May 7–8 or June 24, 2026 bugs); otherwise watch the trend, not the daily wiggle.

“Discover impressions count the same way Search impressions do.” Why it’s wrong: Discover requires the item to be “scrolled into view” and counts once per session, while Web Search counts results that were seen or potentially seen — a looser bar. Do instead: don’t try to reconcile Discover and Search impression counts one-to-one; they’re measuring different things.

“There should be a keyword or query breakdown in here somewhere.” Why it’s wrong: there’s no query dimension because Discover isn’t query-driven at all — “Discover is not a response to a query.” The filter list (date, page, country, appearance type) has no query and no device on it. Do instead: analyze Discover by page and appearance type, not by keyword — the keyword simply doesn’t exist for this surface.

“A share counts as a click, so my CTR should be higher.” Why it’s wrong: Google is explicit that “a click is not counted if the user shares the item, or performs any another action.” Only an actual click-through counts. Do instead: read Discover CTR as click-throughs only, and don’t expect engagement actions like shares or saves to show up in it.

Reading the Discover report — checklist

• Open Performance in Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. and confirm whether a Discover tab exists at all (no tab usually means below the visibility threshold — that’s normal).
• Confirm you’re on the Discover tab, not Search results or Google News — three different surfaces.
• Read the three metrics for what they are: impressions, clicks, average CTR — and don’t go looking for an average position (there isn’t one).
• Remember impressions require scroll-into-view and clicks exclude shares — don’t try to reconcile them 1:1 with Web Search numbers.
• Slice by the available dimensions — page, country, date, appearance type — and don’t expect a query or device breakdown.
• When a page looks like it’s missing metrics, check whether the traffic rolled up to a canonical URLHow search engines pick one canonical URL among duplicates and consolidate signals onto it. rather than the URL you were looking at.
• Treat appearance types (AMPAMP (Accelerated Mobile Pages) is an open-source web framework Google launched in 2015 to make mobile pages load near-instantly via restricted HTML/CSS/JS and CDN caching. It was never a ranking factor and, since June 2021, is no longer required for Top Stories. articles, News Showcase, Web Stories, Videos) as format categories, not quality tiers.
• For a sharp move, cross-check against known incidents (e.g. May 7–8 and June 24, 2026) before assuming a content or eligibility problem.
• Hitting the 1,000-row UI cap? Pull more via the Search Analytics APIA set of REST APIs that let you programmatically read and manage Google Search Console data for properties you've verified — Performance data, URL index status, sitemaps, and properties — authorized with OAuth 2.0. (25,000/request) or the BigQuery bulk export (search_type: discover).
• For the why behind the numbers — eligibility, image specs, volatility — go to Discover SEO; this report is the what.

Discover report — cheat sheet

The essentials

Dimensions

Appearance types — AMPAMP (Accelerated Mobile Pages) is an open-source web framework Google launched in 2015 to make mobile pages load near-instantly via restricted HTML/CSS/JS and CDN caching. It was never a ranking factor and, since June 2021, is no longer required for Top Stories. articles · News Showcase · Web Stories · Videos (format categories, not quality tiers)

How Discover counts differently

• Impression = item scrolled into view, once per session (Web Search: seen or potentially seen).
• A share is not a click — only a click-through counts.
• Coverage includes Chrome New Tab, not just the Google app.

Discover vs. Google News vs. News-in-Search

Data ladder — 16 months · 1,000-row UI cap · 25,000-row API · BigQuery export (search_type: discover, is_anonymized_discover)

Volatility — expected by design. Real 2026 bugs: May 7–8 (Discover logging), June 24 (Gen-AI-in-Discover logging). Check incidents before panicking.

Scripts & snippets

Pulling Discover data past the UI’s 1,000-row cap, and cross-checking it against your analytics. Treat these as starting points — swap in your property and credentials.

Search Analytics APIA set of REST APIs that let you programmatically read and manage Google Search Console data for properties you've verified — Performance data, URL index status, sitemaps, and properties — authorized with OAuth 2.0. — pull Discover rows (Python)

Discover data comes back when you set the type to discover on a Search Analytics query. Note: valid dimensions are date, page, and country only — there’s no query or device for Discover, so don’t request them or the call will error.

# Requires: google-api-python-client, google-auth, and OAuth/service-account creds
from googleapiclient.discovery import build

service = build("searchconsole", "v1", credentials=creds)

request = {
    "startDate": "2026-04-01",
    "endDate": "2026-06-30",
    "type": "discover",              # <- the key line: Discover, not "web"
    "dimensions": ["page"],          # also valid: "date", "country" (NOT query/device)
    "rowLimit": 25000,               # the API cap, vs. 1,000 in the UI
}

resp = service.searchanalytics().query(siteUrl="https://example.com/", body=request).execute()
for row in resp.get("rows", []):
    print(row["keys"][0], row["clicks"], row["impressions"], round(row["ctr"] * 100, 2))

BigQuery — isolate Discover rows from the bulk export (SQL)

If you’ve enabled the Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. bulk export, Discover rows live in the same tables as Web Search, tagged by search_type. This pulls Discover pages and flags the anonymized ones:

SELECT
  url,
  SUM(clicks)       AS clicks,
  SUM(impressions)  AS impressions,
  SAFE_DIVIDE(SUM(clicks), SUM(impressions)) AS ctr,
  MAX(is_anonymized_discover) AS any_anonymized
FROM `your_project.searchconsole.searchdata_url_impression`
WHERE search_type = 'discover'          -- the Discover-only filter
  AND data_date BETWEEN '2026-04-01' AND '2026-06-30'
GROUP BY url
ORDER BY clicks DESC;

Curl — verify a Google incident source with a browser user-agentA user agent is the HTTP request header a client (browser, crawler, or bot) sends to identify itself. For crawlers, a short user-agent token — a substring of that string — is what robots.txt rules actually target.

Some SEO-news hosts (e.g. Search Engine Roundtable) return HTTP 403An HTTP 403 client-error status code meaning the server understood the request but refuses to fulfill it — access is denied, whether or not credentials are involved. For SEO, a 403 served to Googlebot on a page meant to be public and indexable usually points to a misconfigured rule; a persistent 403 keeps a page out of the index either way. to bare fetch tools but respond fine to a normal browser user-agent. Handy when you’re confirming whether a Discover dip lines up with a reported bug:

curl -sL -A "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 \
(KHTML, like Gecko) Chrome/125.0 Safari/537.36" \
  "https://www.seroundtable.com/google-generative-ai-discover-report-bug-41601.html" \
  | grep -i "logging error"

GA4 — sanity-check Discover referral traffic (regex)

Discover click-throughs typically arrive with a Google referrer. To spot-check that your Discover clicks in GSC roughly track your analytics, filter your referrer/source report with a pattern like:

# RE2-style regex for Discover-adjacent Google referrers
^https?://([a-z0-9-]+\.)?google\.[a-z.]+/  # google.com + ccTLDs
# In practice, isolate the Google-app / googleapis referral rows;
# GSC's Discover report is the authoritative count, GA4 is the corroboration.

A caveat on all of these: the GSC Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold. is the authoritative count for Discover performance. The API and BigQuery pulls are how you get more rows of the same data; GA4 is a corroborating sanity check, not a substitute — attribution differs, so the numbers won’t match exactly.

Tools for working with Discover data

• Google Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. — Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold. — Performance → Discover; the authoritative, over-time view of your Discover feedGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses. performance (this article).
• Search ConsoleGoogle's free tool for monitoring crawling, indexing, and search performance. — Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. (Search results) — the Web Search sibling; useful for contrast and for the shared data-ladder (16 months, 1,000-row cap, API, BigQuery).
• Search Console — Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. (Google News) — the separate Google News report; keep it distinct from Discover when your content spans both.
• Search Analytics APIA set of REST APIs that let you programmatically read and manage Google Search Console data for properties you've verified — Performance data, URL index status, sitemaps, and properties — authorized with OAuth 2.0. — the escape hatch past the 1,000-row UI cap (25,000 rows/request); set type: discover, and note there’s no query or device dimension to request.
• BigQuery bulk data exportGSC BigQuery export (bulk data export) is a Google Search Console feature that schedules a daily, unsampled export of a property's Performance data into a Google Cloud BigQuery project — bypassing the UI's ~1,000-row export cap and its rolling ~16-month retention window. It doesn't backfill and still excludes anonymized queries at the query level. — the fullest Discover dataset; Discover rows sit in the standard export tables, filtered by search_type = 'discover'.
• GA4 / your analytics — cross-check Discover referral traffic against GSC’s counts (corroboration, not a substitute — attribution differs).
• Bing Webmaster ToolsMicrosoft's free portal for monitoring and improving how a site appears in Bing search — the peer to Google Search Console, plus IndexNow instant indexing, richer backlink data, and keyword volumes. Because Bing's index also feeds Microsoft Copilot, it doubles as a window into AI-search visibility. / Microsoft Start Partner Hub — there’s no Discover equivalent in Bing; Microsoft Start’s feed reporting lives in its own separate publisher dashboard, not Bing Webmaster ToolsThe free, first-party consoles search engines give site owners — Google Search Console and Bing Webmaster Tools — to see how their pages are crawled, indexed, and ranked, and to fix problems. The legacy name is 'webmaster tools.'.

Resources worth your time

My related writing

• I Studied Google Search Console’s Anonymized “Hidden” Query Terms — my study of how much GSCA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. Performance data gets withheld through anonymization. It’s about the query dimension, which Discover doesn’t have — so the finding doesn’t transfer, but the mental model does: the UI is a sampled, capped, privacy-filtered view, and the API and BigQuery export are how you see more.
• The Beginner’s Guide to Technical SEO — where Search ConsoleGoogle's free tool for monitoring crawling, indexing, and search performance. and its reports fit into the wider technical picture.

My speaking

• How Search Works (SlideShare) — my walkthrough of crawlingCrawling is how search engines use automated bots (like Googlebot and Bingbot) to discover URLs and download pages. A page has to be crawlable to be indexed, but crawling on its own isn't a ranking factor., renderingTurning HTML, CSS, and JavaScript into the final visual page and DOM., indexingStoring a crawled page in the search index so it can appear in results. Crawled is not the same as indexed — Google selects what to keep, and indexing isn't guaranteed., and serving. (My standing disclaimer applies: “This is my understanding of systems… not going to be 100% complete or accurate.”)

From around the industry

• Performance report (Discover) (Google) — the primary Help doc; keep it open while you read your own report.
• Performance report (Google News) (Google) — the disambiguation source for the report Discover is most often confused with.
• Discover and your website (Google) — the 16-month-retention and Chrome-inclusion lines, plus Discover eligibility context.
• Google Discover performance reporting bug in Search Console (Search Engine Land, May 2026) — coverage of the May 7–8, 2026 Discover logging bug, with Google’s direct statement.
• Google Discover Reporting Bug Hits Search Console (TechWyse, May 2026) — corroborating coverage of the same bug, plus the annotate-and-exclude remediation.
• Google Search Console Has First Generative AI Reporting Bug On June 24 (Search Engine Roundtable, Barry Schwartz, June 2026) — the separate June 24 Gen-AI-in-Discover bug, with Google’s notice quoted.
• What is Google Discover? A complete guide for publishers & SEOs (Search Engine Land) — a broader Discover primer for the eligibility side this report doesn’t cover.

A framework for reading Discover without chasing noise

1. Confirm the surface. Start by proving the traffic is actually Discover. The Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold., Google News report, and the News search-type filter cover different surfaces, so combining them creates a false trend before analysis even begins.

2. Separate visibility from completeness. A missing Discover tab means the property has not crossed Google’s report-level impression threshold. Missing rows inside a visible report are a separate low-volume suppression issue. Neither means Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. is misconfigured.

3. Read the right dimensions. Discover has no query or device dimension. Build the analysis around page, country, date, and Discover appearance type instead of trying to force a keyword-reporting workflow onto a queryless feed.

4. Compare stable cohorts. Hold the page set, country filter, appearance type, and date-window length constant. Discover is volatile, so a like-for-like cohort is more useful than comparing whichever pages happened to surface in two different weeks.

5. Escalate only with corroboration. Treat a one-day move as feed volatility. Escalate when a sustained change lines up with lost eligibility, a publishing or technical change, analytics referral data, or a confirmed Google reporting incident.

The standing KPIs for Discover reporting

Eligible-page exposure

• Metric — Discover impressions by page and appearance type.
• What it tells you — Which content Google is choosing to expose in the feed and whether that exposure is broadening or concentrating.
• How to pull it — Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. → Performance → Discover; use the API or BigQuery bulk export when the 1,000-row UI limit hides the long tail.
• Benchmark / realistic range — Establish a property-specific baseline. Discover demand is volatile and eligibility differs by publisher, so there is no honest universal impression target.
• Cadence — Weekly for direction; monthly for a more stable page-level review.

Discover click-through rate

• Metric — Clicks divided by impressions, segmented by page and appearance type.
• What it tells you — Whether exposed items earn a visit; it does not measure shares, saves, or other in-feed actions.
• How to pull it — Average CTR in the Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold., with the same filters and date-window length on both sides of a comparison.
• Benchmark / realistic range — Compare each cohort with its own historical baseline. Mixing countries, page types, or appearance types makes a single target misleading.
• Cadence — Monthly, with a weekly watch during major publishing pushes.

Discover traffic concentration

• Metric — Share of Discover clicks coming from the leading pages.
• What it tells you — Whether performance depends on a small number of volatile winners or is spread across a healthier set of content.
• How to pull it — Export clicks by page, sort descending, and divide the chosen leading-page cohort by total Discover clicks for the same period.
• Benchmark / realistic range — Use your own rolling baseline; the useful signal is a material shift in concentration, not an invented ideal percentage.
• Cadence — Monthly and after a sharp sitewide Discover move.

Test yourself: the Discover report

Five quick questions on reading Google Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results.’s Discover reportThe Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold.. Pick an answer for each, then check.

Discover Report

The Discover report — Google's own label is 'Performance report (Discover)' — is a separate tab in Google Search Console's Performance section showing how your content performed in the Google Discover feed: impressions, clicks, and average CTR by page, country, date, and appearance type. It only appears once a property has crossed a minimum Discover-impression threshold.

Related: Google Discover, Performance report, Google Search Console

Discover Report

The Discover report (Google’s UI/help-doc label is “Performance reportThe Google Search Console report that shows how your site actually performed in Google Search, built from real impressions and clicks. It reports four metrics — clicks, impressions, average CTR, and average position — and keeps the most recent 16 months of data. (Discover)”) is one of the separate performance tabs inside Google Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results., alongside “Search results” (Web Search) and “Google News.” It reports how your site performed specifically in the Google DiscoverGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses. feed — the personalized content stream in the Google app, on the Chrome New Tab page, and on google.com.

It shows three metrics — impressions, clicks, and average CTR — for content that appeared in Discover over the last 16 months, broken out by page (assigned to the canonical URLHow search engines pick one canonical URL among duplicates and consolidate signals onto it.), country, date, and Discover appearance type (AMPAMP (Accelerated Mobile Pages) is an open-source web framework Google launched in 2015 to make mobile pages load near-instantly via restricted HTML/CSS/JS and CDN caching. It was never a ranking factor and, since June 2021, is no longer required for Top Stories. articles, News Showcase, Web Stories, Videos). There is deliberately no query dimension and no device dimension, because Discover isn’t triggered by a search query at all.

Two things trip people up:

• It only shows up once you have meaningful Discover traffic. The tab is “visible only if your property has reached a minimum number of impressions in Discover,” and individual low-volume rows can also be suppressed by a separate row-level threshold. Google doesn’t publish either number, and a suppressed row can still count toward the report’s totals — so a missing tab or row isn’t proof of zero exposure. No traffic, no tab — that’s normal, not a misconfiguration.
• It is not the Google News report. The Discover report covers the Discover feedGoogle Discover is a personalized, mobile-first content feed built into the Google app, Chrome's mobile New Tab page, and google.com that surfaces articles and videos based on a user's interests and activity — not a response to a search query. There's nothing to 'rank' for in the traditional sense; eligibility is governed by Discover's content policies plus the same helpful-content, image, and page-experience signals Google Search already uses.; the separate Google News report covers news.google.com and the Google News mobile apps only; and the News tabGoogle News SEO is the practice of getting eligible for and ranking well in Google's news-specific surfaces — the News tab of Search and the Top Stories carousel. As of the 2024–2025 Publisher Center transition there's no application to file: content that complies with Google's news content policies is automatically eligible, and ranking within that pool is driven by relevance, prominence, authoritativeness, freshness, usability, and location/language. inside regular Search is tracked in the Web Search report filtered to the News search type. Three different surfaces, three different reports.

The same data-access ladder as the Web Search report applies: a 1,000-row cap in the UI, up to 25,000 rows per request via the Search Analytics APIA set of REST APIs that let you programmatically read and manage Google Search Console data for properties you've verified — Performance data, URL index status, sitemaps, and properties — authorized with OAuth 2.0., and the BigQuery bulk export (where Discover rows ship in the same tables, tagged by search type). Discover traffic is more volatile than Search traffic by design, so a jumpy chart usually isn’t a bug — though Google has had genuine Discover reporting glitches, so a sharp move is worth cross-checking against known incidents.

Related: Google Discover, Performance report, Google Search Console

Changelog