The Google num=100 parameter stopped working in September 2025. Overnight, rank trackers lost most of their depth and Search Console impressions fell for most sites. Here is what the parameter did, what changed, what it means for your reports and five steps to take now.
The &num=100 Change In Numbers
- Sep 2025: Google stopped honouring &num=100; the SEO community noticed on September 12 (Source: Brodie Clark Consulting (opens in a new tab))
- Up to 10×: More requests (and cost) for a rank tracker to see the top 100 results, now served about 10 per page (Source: Brodie Clark Consulting (opens in a new tab))
- 87.7%: Of 319 Search Console properties saw impressions fall (Aug 25 to Sep 1 vs Sep 8 to 15, 2025) (Source: Search Engine Land, Tyler Gargula data (opens in a new tab))
- 77.6%: Of those properties lost unique ranking queries over the same period (Source: Search Engine Land, Tyler Gargula data (opens in a new tab))
What was &num=100 anyway?
In a Google search URL, the num parameter asked for more results on one page. A URL such as https://www.google.com/search?q=example&num=100 returned up to 100 organic results instead of the usual 10.
Google never documented it as a supported feature, but it worked for years, so rank trackers, SERP scrapers and SEOs doing quick manual checks all relied on it to see deep into the results with a single request.
What changed: Google drops support for num=100
Around September 11 to 12, 2025, Google began ignoring &num=100: requests that include it now return the normal page of about 10 results. Asked about the change, Google told Search Engine Land: “The use of this URL parameter is not something that we formally support.” (opens in a new tab) Google has not said why it made the change.
| What you look at | Before September 2025 | After September 2025 |
|---|---|---|
| Results per request | Up to 100 | About 10, so deeper results need one request per page |
| Rank trackers | One request covered positions 1 to 100 | Around 10 paginated requests for the same depth, or tracking fewer positions |
| Search Console impressions | Widely believed to include loads of positions 11 to 100 by trackers and scrapers | Lower for most sites, closer to what real people saw |
| Average position | Pulled down by those deep impressions | Higher, even when real rankings did not move |
| Unique queries reported | More long-tail queries ranking on page 3 and beyond | Fewer, with more of the remaining queries in the top 3 and on page 1 |
Why this matters (and why Google might have done it)
For a casual searcher nothing changed: few people scroll past page one or two. For SEOs, analysts and tool vendors the impact landed in four places.
Rank tracking and SERP tools
Tools that fetched 100 results in one go now need many more requests for the same depth.
- Cost: more requests, latency and load per keyword tracked.
- Depth: some tools now track the top 20 or 50 instead of 100.
- Gaps: reports built on 100-deep data may show “lost” keywords.
Search Console data
Many sites saw impressions drop, mostly on desktop, and average position jump overnight.
- Likely cause: trackers loading positions 11 to 100 were counted as impressions. This is a community hypothesis; Google has not confirmed it.
- Effect: fewer low-ranking impressions, so average position looks better.
- Upside: data that sits closer to genuine human visibility.
Scraping and AI crawlers
Fewer results per request raises the effort of scraping Google results at scale.
- Speculation only: SEOs suggest the move discourages bulk scraping, including by AI systems.
- Terms of service: automated scraping of Google results was already against them.
Reporting and strategy
Historical comparisons now mix two different measurement methods.
- Before vs after: year-on-year charts can look worse without any real loss.
- Focus: top-20 visibility, clicks and conversions now say more than raw ranking depth.
What you (as an SEO, marketer or tool user) should do now
If the change touches your tools or reports, work through these five steps:
| Step | What to do | Why |
|---|---|---|
| 1. Don’t panic | Check clicks and conversions before reacting to impressions or position. | The change targets how results are fetched, not your rankings; human traffic behaves the same. |
| 2. Re-evaluate the depth you track | Track positions 1 to 20 or 1 to 50 for most keywords; treat 50 to 100 as a bonus. | For most businesses the top 10 to 20 results hold nearly all the value. |
| 3. Update tools and APIs | Use vendors that paginate or cache, and change your own scripts to fetch page by page with start=0, start=10, start=20. | One large num=100 call no longer works; budget for more requests. |
| 4. Annotate Search Console | Mark September 2025 in your reporting timeline and compare clicks, not impressions, across that date. | Prevents a reporting artefact from being read as a traffic loss later. |
| 5. Tell stakeholders | Explain that impressions fell and average position rose across the industry, not just on your site. | Keeps the conversation on traffic, conversions and keyword intent. |
Final thoughts & implications
Dropping &num=100 changed how search data is collected more than how people search. Toolmakers have recalibrated, analysts read dashboards differently, and the likely long-term result is cleaner Search Console data with less bot noise.
Cleaner data shows who visits. Can they use the page?
With less bot noise, Search Console tells you more honestly which pages real people find. Whether they can actually use those pages is a separate check, and it often fails: the WebAIM Million 2026 (opens in a new tab) found automatically detectable WCAG failures on 95.9% of the top one million home pages. Our guide to automated vs manual accessibility testing explains what a scanner can catch and what still needs a person, and accessibility audit cost sets out what a proper audit costs.
Choose Your Next Step
Start with a free scan, learn what automated tools miss, or talk it through with us. Prefer a form? Try the free accessibility needs assessment. Ready for an expert check of your key pages? The $500 quick accessibility audit covers up to 5 templates with automated scans plus manual keyboard and screen reader spot checks.
Free One-Page Scan
Best for: a quick first look at one public page
AccessibilityScore.in is our free WCAG checker powered by axe-core. It flags automated issues on one page in seconds: a starting point that complements a manual audit, not a replacement for one.
Run a free scan (opens in a new tab)
Automated vs Manual Testing (Guide)
Best for: understanding what scan scores miss
What axe, WAVE and Lighthouse catch well, the barriers only keyboard and screen reader testing find, and how to combine both.
Free 30-Minute Call
Best for: first conversations, scoping and quotes
A first discussion with founder Shadab Saifi (CPACC, WAS, CPWA) to understand where your site stands and which audit, if any, fits. Need a hands-on technical session instead? Book the $100 one-hour consultation.
Book a free 30-minute call (opens in a new tab)