What changed about Google's num=100 parameter?

Google removed num=100 from search results. The maximum is now 10 results per page instead of allowing 100 on one page.

Why does num=100 matter for LLMs?

Many LLM systems, including OpenAI and Perplexity, rely on Google's index directly or indirectly. Cutting deep result pages removed much of the long tail—about 90% of that access in the post's framing.

Who was hit hardest in the analysis cited?

The post cites 88% of websites seeing impression drops and Reddit—often ranking 11–100—seeing sharp declines in LLM citations, with Reddit's stock falling about 15%.

Google's num=100 change rewrote web distribution

November 4, 2025 · 2 min read

Originally posted on LinkedIn, November 4, 2025.

Google capped search pages at 10 results by removing num=100, shrinking long-tail access that many LLM pipelines relied on.

Google removed the num=100 parameter, so search results max out at 10 per page instead of 100. Most LLM systems, including OpenAI and Perplexity, depend directly or indirectly on Google's index, so long-tail access was cut by about 90%. One analysis reported 88% of websites saw impression drops; Reddit, which often ranked in positions 11–100, saw sharp declines in LLM citations and its stock fell about 15%. Discovery and SEO/AEO shifted overnight, especially for startups.

Last month, Google quietly removed the num=100 parameter from its search results. You can no longer view 100 results on one page. The maximum is now 10.

That sounds minor, especially since most people stop at the first few results anyway. But it affects the entire industry. Most LLM systems, including OpenAI and Perplexity, depend directly or indirectly on Google's index in addition to their own crawlers. Overnight, their access to the long tail of the internet was cut by 90%.

The effects were immediate:

According to one analysis, 88% of websites saw a drop in impressions.
Reddit, which often ranks between positions 11-100, saw a sharp decline in citations from LLMs. Its stock fell by 15% as a result.

For startups, this is devastating. Getting discovered became much harder, and the entire SEO and AEO business changed.

One small change rewrote the rules. Building a great product is no longer enough. You have to solve distribution first, because people cannot appreciate a product they never encounter.

Most founders ignore this, but a mediocre product with excellent distribution will always beat an excellent product with weak distribution. We see it constantly. Companies that know how to market succeed at a massive scale, even with fairly ordinary products whose innovation or value is questionable.

As Peter Thiel said:

Most businesses get zero distribution channels to work: poor distribution, not product, is the number one cause of failure. If you can get just one distribution channel to work, you have a great business. If you try for several but don't nail one, you're finished.

Superior sales and distribution by itself can create a monopoly, even with no product differentiation. The converse is not true. No matter how strong your product, even if it easily fits into already established habits and anybody who tries it likes it immediately, you must still support it with a strong distribution plan.

Distribution > product.

Two companies control most online distribution: Google controls search, and Meta controls discovery. Every tiny change either company makes can reshape an entire industry in front of us.

Illustration for “Google's num=100 change rewrote web distribution”

FAQ

What changed about Google's num=100 parameter?: Google removed num=100 from search results. The maximum is now 10 results per page instead of allowing 100 on one page.
Why does num=100 matter for LLMs?: Many LLM systems, including OpenAI and Perplexity, rely on Google's index directly or indirectly. Cutting deep result pages removed much of the long tail—about 90% of that access in the post's framing.
Who was hit hardest in the analysis cited?: The post cites 88% of websites seeing impression drops and Reddit—often ranking 11–100—seeing sharp declines in LLM citations, with Reddit's stock falling about 15%.