In mid‑September 2025 many SEO dashboards went red for many B2B SaaS companies who found their impressions fall by half overnight. Imagine months of SEO efforts going down the drain.

What had happened was that Google had quietly removed a hidden URL parameter, &num=100, that let SEO tools fetch 100 organic results per page. Without it, impressions collapsed because the extra pages had never mattered to real users.

This episode was not just a technical fix –  it exposed how fragile our metrics are and signalled a broader shift towards AI‑powered search and zero‑click journeys.

Why Legacy SEO Metrics Mislead Us

For years, traditional SEO has been built around rankings, impressions and clicks across multiple pages. This model assumes that people scroll through long result lists.  In reality though, they rarely go beyond the first page and almost 60 % plus of queries end without a click.

When Google switched off num=100, 87.7 % of sites saw impressions decline and it wasn’t a penalty –  the missing impressions were bots scraping deep results. (Source)

Meanwhile, AI‑generated overviews now sit atop many results pages. These overviews answer questions directly and takes away clicks from the links.

So, the question then arises that if a hidden parameter can slash your visibility overnight, what else might you be missing about how people and machines find information?

A Shrinking SERP and an Expanding AI Layer

The num=100 story underlines a larger trend –  search visibility is no longer about ranking somewhere down a long list.

For years, adding &num=100 to a Google URL allowed rank‑tracking tools to pull 100 results with one request. Dr. Ernesto Lee, CEO @ LVNG AI notes that this trick let bots discover 100,000 URLs with 1000 queries instead of 10000.

When Google turned it off, SEO tools suddenly had to send ten times more requests just to get the same data  and that’s when impression counts fell off a cliff.

It turned out the so-called “long tail” of search was mostly an illusion  –  results that were buried beyond page one that real users rarely saw and AI could no longer reach without extra effort. At the same time, generative AI tools like ChatGPT, Gemini, and Perplexity are reshaping discovery itself compressing search into direct, conversational answers.

As per a Gartner report, half of all organic search traffic could come through AI assistants by 2028  – a shift that favors clarity and authority over sheer keyword volume.

Zero‑Click Queries and AI Overviews

Zero-click behavior is not something new.

Studies have long shown that nearly 60% of Google searches end without a single click –  but AI overviews are amplifying that trend. With generative summaries now answering questions directly on the results page, users often get what they need without ever visiting a website.

That shift has sparked mixed reactions.

Some see Google’s num=100 change as a way to curb chatbots from scraping full result sets and skipping ads while others view it as a sign of a deeper transformation –  one where visibility is no longer distributed across dozens of pages  but concentrated at the very top.

Whatever the motive, the outcome is clear – ranking on page three is like speaking to an empty room. In today’s search landscape, attention lives on page one  and to earn it, your content must be not just be discoverable but credible enough for AI to quote.

From Ranking to Retrievability

Traditional SEO has always treated search like a ladder – climb higher and you will get traffic. But AI search doesn’t work that way anymore. It behaves more like a recommendation engine surfacing answers it finds most relevant, trustworthy, and contextually rich.

Large language models pull insights from multiple sources, weighing factors such as topical authority, entity relevance, and citation quality. Even Google’s own definition of an impression –  counted only when a listing actually appears on the current results page –  shows how easily old metrics could be distorted by bots.

When num=100 was removed, many sites suddenly saw their average positions improve not because they ranked better, but because those low-visibility impressions on page five and beyond vanished. And that’s the point –  chasing distant rankings no longer matters.

For B2B SaaS, where buying committees research over weeks or months, what really count now is being visible in AI-driven overviews and trusted sources.

Think of visibility less like climbing a ladder and more like brand recall in a recommendation system where you are not trying to be the seventh blue link, but trying to be the model answer.

Navigating the New Search Ecosystem

The num=100 change was not just a technical tweak –  it was a signal that SEO dynamics are changing. B2B SaaS marketers now need to think beyond rankings and focus on how their brand shows up in a world where AI curates the answers.

• Accept cleaner data  – Focus on metrics that matter – clicks, conversions, and qualified pipeline and not just vanity visibility metrics such as impressions. If you still need deeper SERP data, you can use the start parameter (like &start=20), but remember that it comes at a higher crawl cost.

• Dominate page one – With fewer results visible, page one is the new battleground – the long tail keywords are shrinking. Try to narrow your keyword list, consolidate overlapping pages, and focus on queries where you can realistically win. For everything else, diversify. Moreover, build awareness through social content, newsletters, and communities where buyers now start their journey long before they ever search.

• Optimize for generative engines – AI assistants like ChatGPT, Gemini, and Perplexity don’t just rank results – they recommend them. To be recommended, your brand needs authority, clarity, and structure. Use schema markup, publish well-sourced guides, and earn citations from reputable sites. Track where your brand is mentioned in AI responses, and monitor branded search trends which are often early signs that you are being discovered via chatbots.

• Strengthen real-world signals – AI models don’t just read your website –  they read the entire web to learn about you. Customer reviews, analyst mentions, podcasts, and case studies all strengthen your brand’s authority footprint. Participate in industry discussions, host webinars, and publish authentic thought leadership to establish your authority and improve your chances to get picked up by AI crawlers.

• Think like a data scientist – Generative AI adoption is rising fast –  around 70% of organizations globally are already experimenting with it which means your buyers are too. The best approach is to treat AI like another channel – experiment with prompts your customers might use (“best customer engagement tools for SaaS”), and see whether your brand surfaces. Then optimize your content to match those queries.

Conclusion – Opportunities Amid Fewer Pages

Google’s removal of num=100 was more than a technical cleanup – it was a reality check.

For years, marketers chased inflated impression numbers and page-five rankings that no buyer ever saw. The update took away that illusion forcing teams to confront what visibility really means in a world where algorithms and AI models decide what gets seen.

As search compresses into shorter, AI-generated answers, the focus shifts from where you rank to why your brand deserves to be cited. Visibility is no longer earned through volume but through clarity, authority, and usefulness. The brands that stand out will be the ones whose content answers questions better than anyone else – factually, contextually, and conversationally.

In this new ecosystem, search is not a leaderboard –  it’s a dialogue between your expertise and an AI’s understanding. The winners in this new era of search will be the companies that build depth and not just data.

Frequently Asked Questions

1. What exactly was Google’s num=100 parameter?

The num=100 parameter allowed SEO tools and rank trackers to fetch up to 100 search results per query instead of the default 10. It made bulk data collection faster but also inflated impressions and visibility metrics with results that real users rarely saw.

2. Why did Google remove the num=100 parameter?

Google likely removed it to improve data accuracy and reduce automated scraping by bots. It also aligns with the broader shift toward AI-powered summaries, where results beyond page one matter less.

3. How did this change affect SEO data?

Many sites saw impressions drop sharply after the change  but this didn’t mean they lost visibility. It simply removed bot-generated and deep-page impressions, leaving behind cleaner, more realistic data.

4. Does the num=100 change impact actual rankings?

No, your rankings didn’t change only how they are measured did. Average positions often improved because low-visibility impressions on page five or beyond stopped counting.

5. How does this relate to the rise of zero-click searches?

The removal of num=100 highlights how few results actually matter. With AI overviews answering questions directly, most clicks now happen on the first page  or not at all. Studies show nearly 60% of searches end without a click, a trend that’s growing as AI overviews expand.

6. What should B2B SaaS marketers do now?

Treat this shift as an opportunity to build retrievability –  not just rankings. Focus on structured, authoritative content that AI can quote confidently, and track how your brand shows up across tools like ChatGPT, Gemini, and Perplexity