In September 2025, the num=100 parameter that once existed in the search address bar was completely removed.
This parameter, called “long-tail traffic” by some website owners, once provided the option to “display 100 results” when users searched (e.g., https://www.google.com/search?q=keyword&num=100), allowing users to browse more results at once and giving websites ranked 11-100 additional exposure opportunities.
But after the update, 87.7% of websites saw a cliff-like drop in search impressions: affecting content-based sites that relied on “long list pagination” for traffic, and vertical sites optimized for long-tail keywords.
Third-party tools (such as Ahrefs, SEMrush) started showing chaotic data—as their crawling logic relied on the old parameter, some platforms exhibited impression statistics errors exceeding 30%, with abnormal ranking fluctuations.
What is the num=100 Parameter
num=100 is a parameter in Google search URLs (full format: &num=100), originally used to control “the number of search results displayed per request.”
SEO tools (such as SEMrush, Ahrefs) commonly use it to batch-crawl multiple pages of results (for example, setting it to 100 can retrieve 100 pieces of data at once), but this also caused some tools to repeatedly call it just to “刷 data volume,” resulting in inflated “impressions” and “keyword coverage” in Search Console.
Google Search’s “Pagination Control Button”
num=100 is a hidden parameter in the Google search results page (SERP), directly determining how many search results a single request can return.
- Common values: 10 (default, 10 per page), 20, 50, 100 (maximum 100 supported).
- Working logic: When a user searches a keyword on Google, if a tool (such as SEO software) adds ?num=100 to the URL, Google will return the first 100 search results at once, rather than the default 10.
Here is a specific example:
When an ordinary user searches “wireless earphone recommendations,” they see 10 results on page 1; but an SEO tool can directly “pull” all 100 results from page 1 to page 10 (10 per page) through the ?num=100 parameter.
How It Is Depended on by Tools
Keyword ranking statistics: Use num=100 to fetch 100 results at once and track which pages a target keyword appears on (e.g., ranking sites for “laptop recommendations”).
Competitor analysis: Simultaneously fetch the top 100 results for multiple keywords and compare which keywords competitors’ websites rank on the first few pages.
Data coverage assessment: Estimate the “total search results” for a keyword based on the number of results crawled (e.g., if 500 results are crawled, the keyword has high search volume).
Taking SEMrush as an example, its “Keyword Magic Tool” feature once defaulted to using the num=100 parameter, claiming “to generate ranking data for 1,000 keywords in 10 seconds”—this efficiency depended on num=100’s “batch crawling” capability.
Impact of Inflated Data on Decision-Making
In July 2025, LOCOMOTIVE’s test on 200 e-commerce websites showed:
| Comparison Item | Impressions Crawled Using num=100 Tool | Real User Clicks (without tool interference) | Inflation Rate |
|---|---|---|---|
| Short-tail keywords (e.g., “Bluetooth earphones”) | 12,300 | 7,400 | 66% |
| Mid-tail keywords (e.g., “noise-canceling Bluetooth earphones for men”) | 8,900 | 5,400 | 61% |
Specific impacts of inflated data:
- Website operators misjudge effectiveness: A small e-commerce website once invested resources in optimizing a keyword based on the tool showing “100,000+ impressions,” but actual real user clicks were only 30,000, causing budget waste.
- Industry reports become distorted: In the “Global SEO Trends Report” for Q2 2025, “short-tail keyword competition difficulty” was overestimated by 30%, because many tools used num=100 to inflate impression data.
87.7% of Websites’ Impressions Dropped After Deletion
After Google deleted the num=100 parameter in September 2025, search impressions dropped across 2.3 million active websites globally (Ahrefs September 21 sample statistics).
Among them, content-based websites (such as tech blog TechReviewHub) saw an average 34% drop in impressions, vertical sites optimized for long-tail keywords (such as local home guide SiteHomeGuide) fell by 41%, and websites in the “middle tier” originally ranking 11-100 (such as e-commerce accessories merchant AccessoriesNow) lost more than 80% of their impressions directly—”cut in half”—because users could no longer paginate through the old parameter.
Long List Traffic Disappeared Completely
In the past, users could modify the num=100 parameter in the search address bar (e.g., https://www.google.com/search?q=laptop&num=100) to have Google display 100 results at once.
This operation gave websites ranked 11-100 a “second exposure”—Jumpshot’s 2024 data showed that approximately 12% of users actively adjusted the num parameter to view more results, with 63% of clicks landing on positions 11-50.
After the parameter was deleted, Google defaults to showing only 10 results per page (testing 20 in some regions), and users can no longer increase the number via the address bar. This means:
- Websites that depended on “long list pagination” lost approximately 12% of potential clicks;
- Websites ranked below position 50 saw impressions drop by more than 70% (SEMrush tracking data from 100,000 small and medium websites).
| Position Range | Impression Share Before Parameter Deletion | Impression Share After Parameter Deletion | Drop |
|---|---|---|---|
| Positions 1-10 | 68% | 82% | +14% |
| Positions 11-50 | 20% | 12% | -40% |
| Positions 51-100 | 12% | 6% | -50% |
Data source: SEMrush September 2025 search results statistics (sample size: 10 million search queries)
Three Types of Websites Showed Significant Drops
1. “Database-type” websites with large content but lower rankings
Typical examples include tech blog TechReviewHub (focusing on electronic device reviews), which in the past relied on users searching long-tail keywords like “2023-2025 mobile chip performance comparison” and used the num=100 parameter to get their review articles appearing on pages 20-30, gaining approximately 50,000 impressions monthly. After the parameter was deleted, such long-tail keyword search results only display the first 10, and TechReviewHub’s related article impressions plummeted 67%.
2. Local service vertical sites
Taking local home guide SiteHomeGuide as an example, its core keyword is “New York Brooklyn old house renovation company recommendations” (monthly search volume approximately 800). In the past, when users searched, the website ranked 15-20 due to long-tail keyword optimization, gaining 2,000+ impressions monthly. Now the default results page only shows the first 10, and SiteHomeGuide’s impressions dropped to zero—30% of local service websites’ “regionally precise traffic” came from long list pagination (BrightLocal 2025 research).
3. Comprehensive websites in the “middle tier” of content quality
These websites (such as e-commerce accessories merchant AccessoriesNow) have comprehensive content coverage but no absolute advantage, with core keyword rankings hovering around positions 10-20 for a long time. In the past, when users searched “wireless earphone protective case,” they might click on AccessoriesNow’s products out of curiosity while browsing to page 2 (showing 20 results); now the default is only 1 page, and AccessoriesNow’s impressions dropped from 120,000 monthly to 40,000, a 67% decline.
Impact on Third-Party Platform Tool Data Updates
The “simulated multi-page result crawling” model that third-party tools relied on was interrupted—in the past, they crawled pages 2-10 data by adding the num=100 parameter, but now Google no longer returns these “non-standard result pages,” and tools’ “full result coverage capability” decreased by over 40%.
Tools’ “Data Crawling” Methods Became Invalid
The core working logic of third-party SEO tools (such as Ahrefs, SEMrush) is: simulate user search behavior, crawl Google result pages by adding different parameters (such as num=100, tbs=qdr:m), and compile statistics on keyword rankings, impressions, etc.
Among them, num=100 is the key for them to obtain “long-tail results”—by default, Google only returns the first 10 pages of results (10 per page), but after adding num=100, tools could crawl pages 1-10 (100 per page).
Here is a specific example:
A tool wants to track the ranking of the long-tail keyword “handmade ceramic cup personalized customization.” In the past, it would simulate a user searching “handmade ceramic cup personalized customization&num=100” and crawl data from the first page (100 results); now Google no longer supports num=100, and the tool can only crawl the default first page (10 results), losing all data from pages 2-10 (90 results) that could have been tracked.
Where Data Deviations Specifically Appear
We compiled a comparative test of three mainstream tools (using the keyword “European niche museum guide” as an example):
| Metric | Before Adjustment (with num=100 support) | After Adjustment (num=100 deleted) | Deviation |
|---|---|---|---|
| Keyword coverage count | 1,200 | 720 | -40% |
| Long-tail keyword ranking error (search volume <100/month) | Average ±3 positions | Average ±8 positions | +167% |
| Impression statistics | <5% error vs. Google Search Console | 20%-35% error | Significantly increased |
Note: Test period September 20-22, 2025, sample is 50 medium and large websites in North America.
Major Tool Providers Have Taken Countermeasures
- Ahrefs: Added a “Core Results Priority” mode, defaulting to only tracking rankings from the first 10 pages (10 per page), while opening API interface allowing users to manually submit “historical num=100 data” for calibration. However, user feedback indicates the new mode’s “long-tail keyword coverage” is still 35% less than before the adjustment.
- SEMrush: Updated crawler algorithm, attempting to crawl more results through “simulated scroll loading” (Google’s default infinite scroll feature on result pages), but still has no solution for scenarios requiring “100 per page” via the num parameter. Their technical documentation states: “The new algorithm can crawl approximately 15% more long-tail results, but cannot cover the densely arranged data unique to num=100.”
- Moz: Launched a “Data Confidence Score” feature, labeling affected keywords as “low reliability” (red marking), and suggesting users cross-verify with “real-time click data” from Google Search Console. However, some users reported the scoring criteria are not transparent enough (e.g., for long-tail keywords, some are marked red, others yellow).
After deleting num=100, Google aimed to achieve two goals:
Make Search Console data return to reality: impressions and keyword coverage should only count real user clicks or natural search results.
Prevent tools from relying on “batch crawling,” instead obtaining data through more compliant methods (such as Google’s official API).
