How to Read Your Search Console Data in 2026, When Ranking No Longer Means Traffic

If you have opened Search Console lately and found pages sitting near the top of page one that barely get clicked, you are not reading it wrong. You are reading it in a year when ranking and traffic have come apart. A page can hold a strong position, be shown thousands of times, and send almost no one to your site, because the answer was given on the results page before anyone needed to click.

That changes how a publisher should read search data. Position on its own is no longer a report. This guide shows how to read impressions, position and click rate together, what Search Console's newer reports let you see that you could not see a few months ago, and three checks to run this week. It is the measurement half of our guide to how publishers rank in Google Search.

Why position no longer predicts clicks

Two pieces of research reported in September make the shift concrete. Chartbeat's 2026 Publisher Playbook, reported by Search Engine Journal, found Google Search referrals to publishers down 40.2 percent year over year from July 2025 to July 2026, after a 21.9 percent fall the year before. And a University of Pennsylvania and Northeastern University study found that people using AI Mode clicked through to other websites 18.8 percentage points less often than people using regular Google Search.

Neither of those means search is dead, and neither means nothing has changed. Both lazy conclusions miss the real point. Search still decides whether your work is seen, and it still feeds the AI answers that sit on top of results. What it no longer does reliably is turn a good position into a visit. On our own site we see exactly that shape: pages holding positions in the upper half of page one that draw far fewer clicks than those positions used to. If you only watch average position, you would think those pages were fine.

A page at position four with almost no clicks is not a ranking problem. It is telling you the answer was given on the results page.

How to read impressions, position and click rate together

Search Console's Performance report gives you four numbers for every query and page: clicks, impressions, click-through rate and average position. Read them as a set, because each combination tells a different story.

1. High impressions, strong position, very low click rate. The page is being shown and placed well, but searchers are getting what they need without visiting. This is the AI answer pattern. The fix is not a better ranking. It is giving the searcher a reason to click, such as a sharper title that promises something the summary cannot, or depth the summary will not carry. 2. Falling impressions and falling position together. You are losing visibility, which is a ranking problem. Check the dates against known Google updates first, as we walk through in our guide to the September 2026 spam update. 3. Steady position, rising impressions, flat clicks. You are appearing for more searches, but the new ones are not converting into visits. Look at which queries were added. They are often broader questions that an AI answer resolves on the page. 4. Lower position, healthy click rate. Searchers are choosing you even from further down. That is a page worth improving, because the title and topic are already doing their job.

The habit that matters is comparing a page with its own history rather than with an industry average. A click rate that has halved at the same position is a signal. A click rate that is simply lower than someone else's benchmark usually is not.

What Search Console shows now that it did not before

Google has added reporting this year that helps with exactly this problem.

The generative AI performance reports

In June 2026, Google introduced generative AI performance reports in Search Console, which it says give site owners dedicated views of their impressions within generative AI features on Search, such as AI Overviews and AI Mode. For a publisher, that is the first view from Google itself of how often your pages appear inside AI answers.

Read those numbers with one caveat in mind. Search Console treats an entire AI Overview as a single block. As Search Engine Journal reported in September, every link inside an AI Overview is given the same top position in your reports, and Google's John Mueller confirmed that "Search Console still treats entire AI-generated overviews as a single block." So position inside AI features tells you that you were included, not where you were placed. Measure whether you were cited and whether those readers arrived, not a position number.

The multimodal filter

On September 24, 2026, Google added a multimodal filter to the Performance report for searches where someone used an image as part of the search. Google's announcement says the data includes searches with Lens, Circle to Search on Android, image uploads to Google Search, and the Chrome right-click "Search this image" feature, and that it is "designed to give you insights into how your content is surfaced when users search using images." For publishers in visual niches such as food, home, style and travel, this is a slice of search you could not separate out before.

What still moves rankings, briefly

Measurement only helps if you act on it, and the ranking side has not changed in Google's own description. Its guidance says its ranking systems prioritize "helpful, reliable information that's created to benefit people," and it asks whether content provides original information, reporting, research or analysis. We cover the publisher-specific mechanics, from internal links to syndicated copies, in how publishers rank in Google Search.

Where SEO and AI citation overlap

The same work earns both, up to a point. Original, well-attributed, clearly structured content is what ranks in Google and what AI answers choose to cite. Where they part ways is the click: a citation in an AI answer may earn your brand a mention without earning a visit. Our guide to getting cited in Google AI Overviews covers the citation side, so we will not repeat it here.

Three checks to run this week

1. Find your high-position, low-click pages. In Search Console, open Performance, then Search results, turn on all four metrics, open the Pages tab and sort by impressions. Look for pages with a strong average position and a click rate far below their own figure from a year ago, using the date comparison. Those are your AI answer pages. Rewrite their titles and openings to promise what a summary cannot give. 2. Open the generative AI performance report. See which of your pages appear inside AI Overviews and AI Mode, and set that list next to the pages from check one. Overlap tells you where AI answers are absorbing your traffic. 3. Switch the search type to multimodal. In the Performance report's search type filter, compare the multimodal view with standard web search. If image-led searches are a meaningful share for you, make sure your key pages carry strong, relevant images with descriptive alt text.

Then make it a routine rather than a one-off. Pull Search Console data by page and query every week through the Search Analytics API, whether you wire it in n8n with an HTTP node on a schedule, build a Make scenario, or run a small scheduled script, and flag pages whose click rate falls well below their own baseline at a steady position. The tool is a preference. The habit is what turns a confusing chart into a list of pages to fix.

Where to go from here

Being found in 2026 means ranking in Google and being cited in AI answers, the discipline called Generative Engine Optimization, or GEO, often referred to as AI search optimization. Our GEO Authority System at $499 runs an LLM Visibility Evaluation on your site and gives you the playbook and distribution flow to build authority that both recognize. It does not promise rankings or traffic, because nobody honest can.

Frequently asked questions

Why does my page rank on page one but get almost no clicks?

Often because the searcher's question was answered on the results page, by an AI Overview, AI Mode or another feature, before they needed to click. A University of Pennsylvania and Northeastern University study found AI Mode users clicked through to other websites 18.8 percentage points less often than regular Search users.

How should I read Search Console data in 2026?

Read clicks, impressions, click-through rate and average position together, and compare each page with its own history. Strong position with a collapsing click rate points to AI answers absorbing the visit. Falling impressions and position together points to a ranking problem.

Does Search Console show where I appear in AI Overviews?

Google introduced generative AI performance reports in June 2026 with dedicated views of impressions in AI features such as AI Overviews and AI Mode. Search Console treats an entire AI Overview as a single block, so every link inside it gets the same top position, which means position there shows inclusion rather than placement.

What is the multimodal filter in Search Console?

Added on September 24, 2026, it separates searches where someone used an image as part of the search, including Google Lens, Circle to Search on Android, image uploads to Google Search, and Chrome's right-click "Search this image."

Is SEO still worth it for publishers in 2026?

Yes, with a different scorecard. Ranking still decides whether you are seen and feeds the AI answers on top of results. What has changed is that a good position no longer guarantees a visit, so publishers should measure click rate and citations alongside position.

Key takeaways

• In 2026 a page can rank well and still not be clicked, so average position on its own is no longer a report.
• Read clicks, impressions, click-through rate and position together, and compare each page with its own history.
• Google Search referrals to publishers fell 40.2 percent year over year to July 2026, and AI Mode users clicked out 18.8 points less than regular Search users.
• Search Console's generative AI performance reports show impressions in AI features, but an entire AI Overview counts as one position.
• The September multimodal filter separates image-led searches such as Lens and Circle to Search.
• Find high-position, low-click pages, check them against the AI report, and rewrite titles and openings to earn the click.

Sources

• Search Engine Journal, SEO Pulse, September 26, 2026 (Chartbeat figures and the AI Mode click-through study): https://www.searchenginejournal.com/seo-pulse-google-spam-update-image-search-data/590882/
• Search Engine Journal, Google Search Decline Steepened To 40% YoY For Publishers (Chartbeat 2026 Publisher Playbook), September 24, 2026: https://www.searchenginejournal.com/google-search-decline-publishers-chartbeat/590681/
• Google Search Central Blog, Introducing Search Generative AI performance reports in Search Console, June 2026: https://developers.google.com/search/blog/2026/06/gen-ai-performance-reports
• Search Engine Journal, Search Console uses block flattening for AI Overviews, September 24, 2026: https://www.searchenginejournal.com/search-console-uses-block-flattening-for-aios-forget-position-focus-on-outcomes/589582/
• Google Search Central Blog, Announcing web multimodal Search performance reporting in Search Console, September 2026: https://developers.google.com/search/blog/2026/09/web-multimodal-in-sc
• Search Engine Journal, Google Search Console Adds Multimodal Filter For Image-Based Searches: https://www.searchenginejournal.com/google-search-console-multimodal-filter/590781/
• Google Search Central, Creating helpful, reliable, people-first content: https://developers.google.com/search/docs/fundamentals/creating-helpful-content