{ "version": "https://jsonfeed.org/version/1.1", "title": "SEO Madman", "home_page_url": "https://seomadman.com", "feed_url": "https://seomadman.com/feed.json", "description": "Technical SEO news, original research, studies and free tools for search and AI answer engines, by Adam Hafez.", "language": "en", "authors": [ { "name": "Adam Hafez", "url": "https://seomadman.com/authors/adam-hafez" } ], "items": [ { "id": "https://seomadman.com/news/bing-verify-you-are-human-check", "url": "https://seomadman.com/news/bing-verify-you-are-human-check", "title": "Bing now asks some searchers to verify they are human", "summary": "Search Engine Roundtable reports Bing showing a verify-you-are-human checkbox before results. The trigger is unknown and the check was not reproduced.", "content_text": "Bing is showing some searchers a verify-you-are-human checkbox before any results appear, according to Search Engine Roundtable on 18 September 2026. The screen asks users to solve a challenge to continue. Barry Schwartz could not reproduce it, and the cause is not stated. Google reportedly tightened scraper blocking in the same week.\n\n- Search Engine Roundtable reported on 18 September 2026 that Bing shows some users a verify-you-are-human screen before search results.\n- The screen tells users to take one last step and solve a challenge, which in the reported case was a checkbox.\n- Barry Schwartz said he could not replicate the check, so what triggers it is unknown from the source.\n- The only evidence cited is a screenshot and a post from Sachin Patel, and Bing has not been quoted explaining it.\n- Google is separately reported to be blocking scrapers and third-party tracking tools more effectively since about 13 September.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "news", "technical-seo" ] }, { "id": "https://seomadman.com/news/chatgpt-sponsored-agents-ads-test", "url": "https://seomadman.com/news/chatgpt-sponsored-agents-ads-test", "title": "OpenAI tests Sponsored Agents as a new ChatGPT ad format", "summary": "OpenAI is testing Sponsored Agents: labeled conversations with a business agent that users start after clicking a ChatGPT ad, with select US advertisers.", "content_text": "OpenAI is testing a ChatGPT ad format called Sponsored Agents, reported on September 16, 2026. After clicking an ad, some users may start a clearly labeled conversation with a business-sponsored agent before following a link to the advertiser's website. The test is limited to select advertisers in the United States.\n\n- OpenAI is testing Sponsored Agents, an ad format that lets users start a conversation with a business-sponsored agent after clicking an ad in ChatGPT.\n- The test runs with select advertisers in the United States, and only some users see the option after a relevant ad.\n- Users can explain what matters to them, ask follow-up questions and follow a link to the business's website when ready.\n- OpenAI says the Sponsored Agent conversation is labeled and distinct from ChatGPT's independent answers.\n- OpenAI also announced other ChatGPT Ads changes, including flexible attribution windows, Shopify and HubSpot integrations and Event Quality Score.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "news", "ai-search" ] }, { "id": "https://seomadman.com/news/google-discover-desktop-missing", "url": "https://seomadman.com/news/google-discover-desktop-missing", "title": "Google Discover on desktop has disappeared, and Google is silent", "summary": "Search Engine Roundtable reports Discover no longer appears on desktop Google.com, with no explanation from Google. What is verified and what is not.", "content_text": "Google Discover no longer appears on the desktop version of Google.com, according to Search Engine Roundtable's 17 September 2026 report. The feed was tested in 2023, announced for 2025 and began rolling out, but has been missing for several months. The report cites screenshots showing it removed in every country where it worked, and offers no reason.\n\n- Search Engine Roundtable reported on 17 September 2026 that Discover has disappeared from the desktop interface of Google.com after several months.\n- Google tested Discover on desktop in 2023, announced it for 2025 and began rolling it out, but the feature did not persist.\n- An X post by Damien Andell, cited by the report, says Google shut down desktop Discover in every country where it was working, with June and September 2026 screenshots as evidence.\n- The report gives no explanation for the removal, and it does not cite a Google statement, so the reason and whether the feed returns are both unknown.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "news", "search-console" ] }, { "id": "https://seomadman.com/news/google-search-bar-ai-mode-button-test", "url": "https://seomadman.com/news/google-search-bar-ai-mode-button-test", "title": "Google tests an AI Mode button in the results page search bar", "summary": "Google is testing an AI Mode button in the search bar on results pages, not only the home page. The button is inconsistent and regular search still works.", "content_text": "Google is testing an AI Mode button inside the search bar on the search results page, not only on the home page. Search Engine Roundtable reported the test on September 18, 2026. The button does not always appear, and normal searching still works, though users may need to press Enter.\n\n- Google is testing an AI Mode button in the search bar on the search results page, not just on the home page.\n- Dwayne Cubbins first flagged the test on X earlier in the week, and Barry Schwartz confirmed it while testing.\n- The button does not always appear, and it does not show for every search suggestion even on queries that can trigger it.\n- Regular search still works while the button is displayed, though users may need to press Enter to continue with a normal search.\n- Multiple test variations may exist, so what one searcher sees may differ from another.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "news", "ai-search" ] }, { "id": "https://seomadman.com/news/google-search-profiles-update", "url": "https://seomadman.com/news/google-search-profiles-update", "title": "Google Search profiles get new article cards and a lower bar", "summary": "Google cut the follower requirement for Search profiles to 10,000, added richer article views, and lets one account manage up to 10 profiles.", "content_text": "Google updated Search profiles on 16 September 2026. Publishers now see enhanced article views with optimized thumbnails and longer headlines, the follower requirement dropped to 10,000 on YouTube, Instagram, X or TikTok, and one Google account can manage up to 10 separate profiles for different brands. Google also added website badges that let visitors follow you.\n\n- Search profiles now show enhanced article views with optimized image thumbnails and longer headlines, which Google says help users dive deeper into your content.\n- The follower requirement started at 100,000, was cut to 35,000, and is now 10,000 on YouTube, Instagram, X or TikTok.\n- One Google account can now manage up to 10 separate Search profiles, aimed at media organizations that run many publications.\n- Google has also added an editable handle, plus website badges and buttons that let visitors follow you from your own site.\n- Badge rules forbid stretching, distorting, rotating or recoloring the Super G icon and the Search profile button.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "news", "technical-seo", "content-strategy" ] }, { "id": "https://seomadman.com/reports/canonical-tag-adoption-web-almanac-2025", "url": "https://seomadman.com/reports/canonical-tag-adoption-web-almanac-2025", "title": "Canonical tag adoption, 2025: 68% of desktop pages set one", "summary": "The 2025 Web Almanac SEO chapter puts rel=canonical on 68% of desktop and 67% of mobile pages, up from 65%, with mismatches under 1%.", "content_text": "HTTP Archive's 2025 Web Almanac found canonical tags on 68% of desktop and 67% of mobile pages in its July 2025 crawl, up from 65% a year earlier. Almost all sit in the HTML head: HTTP header canonicals appear on only 0.58% of desktop sites, and canonical mismatches affect just 0.72% of them.\n\n- Canonical tag usage rose to 68% of desktop pages and 67% of mobile pages in 2025, up from 65% on both in 2024.\n- Pages that other pages treat as the canonical version, called canonicalized, rose to 7% on desktop and 9% on mobile.\n- HTTP header canonicals remain rare, found on 0.58% of desktop sites and 0.50% of mobile sites, against 64.40% and 64.27% for raw HTML canonicals.\n- Canonical mismatches were detected on 0.72% of desktop sites and 0.71% of mobile sites, a small drop from 2024.\n- Rendered-change canonicals, where JavaScript alters the tag after load, affect 2.71% of desktop sites and 3.02% of mobile sites.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "reports", "technical-seo", "indexing" ] }, { "id": "https://seomadman.com/reports/cloudflare-ai-training-crawler-share-2026", "url": "https://seomadman.com/reports/cloudflare-ai-training-crawler-share-2026", "title": "AI training now drives 52% of crawler requests, Cloudflare says", "summary": "Cloudflare reports AI training made up 52% of identified crawler requests in June 2026, up from 22% in Spring 2025, with search crawling shrinking.", "content_text": "Cloudflare reports that 52% of crawler requests it identifies by purpose were for AI training as of June 2026, up from 22% in Spring 2025. Mixed-use crawlers add over 36% more. Pure search crawling, the kind that can send readers back to a site, is now a small and declining share of crawler activity.\n\n- Cloudflare says 52% of crawler requests were for AI training as of June 2026, up from 22% in Spring 2025.\n- Mixed-use crawlers, which blend search, agent use and training, account for over 36% of crawler activity.\n- Cloudflare describes pure search crawling as a small and declining share, despite remaining critical for publisher visibility.\n- More than one-third of crawler activity still comes from mixed-use bots, so purpose labels cannot fully separate training from search.\n- Cloudflare defines its crawl-to-refer ratio as HTML requests from a platform's crawlers divided by HTML requests referred from that platform, normalized to one referral.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "reports", "ai-search", "technical-seo" ] }, { "id": "https://seomadman.com/reports/desktop-search-click-distribution-datos-q4-2025", "url": "https://seomadman.com/reports/desktop-search-click-distribution-datos-q4-2025", "title": "Where desktop Google searches end: Datos Q4 2025 data", "summary": "Datos and SparkToro's Q4 2025 clickstream shows 24.5% of US desktop Google searches ended with no click in December, versus 25.1% a year earlier.", "content_text": "Datos and SparkToro's Q4 2025 report found that 24.5% of US Google desktop searches ended with no click in December 2025, against 25.1% in December 2024. Organic clicks reached 42.0% and searches staying inside Google-owned properties 16.6%. The report describes zero-click behavior as stabilizing on desktop rather than accelerating.\n\n- Datos found that 24.5% of US Google desktop searches ended with no click in December 2025, down slightly from 25.1% in December 2024.\n- In the same December 2025 data, 42.0% of US searches ended in an organic click on a non-Google result and 2.1% in a paid click.\n- Another 16.6% of US searches ended on a Google-owned property such as maps, weather or hotels, up from 14.5% in December 2024.\n- The report states that zero-click behavior fluctuated but did not accelerate in Q4 2025, and that its data covers desktop only.\n- Fewer Google searches per desktop searcher is the report's larger shift, which SparkToro's Rand Fishkin puts at a nearly 20% decline in the US from 2024 to 2025.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "reports", "ai-search" ] }, { "id": "https://seomadman.com/reports/inp-good-scores-by-site-rank-2025", "url": "https://seomadman.com/reports/inp-good-scores-by-site-rank-2025", "title": "INP in 2025: 77% of mobile sites pass, but the top 1,000 lag at 63%", "summary": "The 2025 Web Almanac's INP data: 77% of mobile sites have good INP, popular sites trail at 63%, and mobile home pages beat inner pages 80% to 69%.", "content_text": "About 77% of mobile websites had good Interaction to Next Paint in the 2025 Web Almanac, up from 74% in 2024. Popular sites trail: the top 1,000 reached 63%, up from 53%. Desktop stood at 97%. On mobile, home pages hit 80% good INP while secondary pages fell to 69%, an 11 point gap.\n\n- Mobile INP improved to 77% of websites with good scores in 2025, up from 74% in 2024, while desktop held at 97%.\n- The top 1,000 websites gained the most, rising from 53% to 63% good INP, yet they still sit 14 points below the overall mobile average.\n- Mobile home pages reached 80% good INP, a 7 point gain, while secondary pages slipped to 69%, opening an 11 point gap that did not exist in 2024.\n- Lab Total Blocking Time worsened by 58% even as field INP improved, so lab scores alone no longer predict what real users experience.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "reports", "technical-seo" ] }, { "id": "https://seomadman.com/reports/llms-txt-fetch-counts-by-ai-crawler", "url": "https://seomadman.com/reports/llms-txt-fetch-counts-by-ai-crawler", "title": "llms.txt fetch counts: what AI crawlers requested on 83 sites", "summary": "EZY Research's 12-week server-log panel shows OpenAI and Anthropic crawlers fetching llms.txt in single digits while pulling robots.txt thousands of times.", "content_text": "Server logs from 83 websites over 12 weeks show major AI crawlers barely touch llms.txt: OpenAI's crawler fetched it 7 times against 3,990 robots.txt fetches, Anthropic's 9 against 3,120, and PerplexityBot never. Meta-ExternalAgent is the exception at 193 llms.txt fetches, more than its 172 robots.txt fetches.\n\n- OpenAI's crawler fetched llms.txt 7 times across 83 sites in 12 weeks, against 3,990 robots.txt fetches from the same crawler.\n- Anthropic's ClaudeBot fetched llms.txt 9 times versus 3,120 robots.txt fetches, and PerplexityBot fetched it zero times versus 775.\n- Googlebot requested llms.txt 67 times, roughly a hundredth of the 5,125 times it requested robots.txt.\n- Meta-ExternalAgent is the one crawler that fetched llms.txt more than robots.txt, with 193 llms.txt requests against 172.\n- The panel skews toward small-business sites whose owners sought an AI-visibility tool, so the counts describe that population and not the whole web.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "reports", "ai-search" ] }, { "id": "https://seomadman.com/research/ai-brand-visibility-factors-ahrefs-75k-brands", "url": "https://seomadman.com/research/ai-brand-visibility-factors-ahrefs-75k-brands", "title": "What correlates with AI brand mentions? Ahrefs' 75,000-brand study", "summary": "Ahrefs correlated 12 metrics with brand mentions in ChatGPT, AI Mode and AI Overviews: web mentions beat backlinks, and it warns of no causation.", "content_text": "Ahrefs' December 2025 study of 75,000 brands found branded web mentions correlate with AI mentions at 0.656 to 0.709, far above backlinks at roughly 0.20 to 0.28. YouTube mentions correlated highest, near 0.74. Ahrefs itself warns correlation is not causation, so these figures do not show that raising mentions causes visibility.\n\n- Ahrefs found branded web mentions correlated with AI mentions at 0.664 in ChatGPT, 0.709 in AI Mode and 0.656 in AI Overviews.\n- YouTube mentions had the strongest correlation of any metric, at roughly 0.737 in ChatGPT, 0.740 in AI Mode and 0.712 in AI Overviews.\n- Backlink counts correlated at only about 0.20 to 0.28, and Domain Rating at 0.266 to 0.326, so link metrics were weak predictors here.\n- Ahrefs states that correlation is not causation and that improving these metrics will not automatically boost AI visibility.\n- Every brand in the sample had a Domain Rating above 40, so the results say little about small or new sites.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "research", "ai-search" ] }, { "id": "https://seomadman.com/research/bing-sitemap-lastmod-accuracy-study", "url": "https://seomadman.com/research/bing-sitemap-lastmod-accuracy-study", "title": "How accurate are sitemap lastmod values? Bing's study", "summary": "Bing's own study of sitemaps found 18% of lastmod values set incorrectly, usually to the generation date, and Google says it only trusts accurate ones.", "content_text": "Bing's study found that among hosts with a sitemap, 84% of those sitemaps set lastmod, but 18% set it incorrectly, most often as identical dates equal to the sitemap's generation time. Google states it uses lastmod only when it is consistently and verifiably accurate, so a wrong date is worse than none.\n\n- Bing reported that 58% of hosts with at least one indexed URL had a known XML sitemap, and 84% of those sitemaps set a lastmod attribute.\n- Bing reported that 79% of lastmod values were correct and 18% were not correctly set, with identical dates across URLs the most common fault.\n- After talking to webmasters, Bing concluded most wrong dates were the sitemap generation date, not the date the content was modified.\n- Bing said it was rebuilding its crawl scheduling to use lastmod, so accurate dates reduce crawling of unchanged pages and favor recently updated ones.\n- Google documents that it uses lastmod only if it is consistently and verifiably accurate, which makes an untrustworthy date a liability.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "research", "technical-seo", "indexing" ] }, { "id": "https://seomadman.com/research/content-freshness-ai-citations-ahrefs-study", "url": "https://seomadman.com/research/content-freshness-ai-citations-ahrefs-study", "title": "Do AI assistants cite fresher content than Google? Ahrefs data", "summary": "Ahrefs compared the age of 16.975 million URLs cited by AI assistants against organic Google results. Here is the method, the numbers and the limits.", "content_text": "Ahrefs analyzed 16.975 million URLs cited by AI assistants and Google, and found AI-cited URLs averaged 1,064 days old versus 1,432 for organic results, 25.7 percent fresher. Cited content still averaged 2.9 years old, and Ahrefs warns that freshness is one factor among many.\n\n- Ahrefs extracted 16.975 million cited URLs from Brand Radar across ChatGPT, Perplexity, Gemini, Copilot, AI Overviews and organic Google results.\n- URLs cited by AI assistants averaged 1,064 days old, against 1,432 days for URLs in organic Google results, which Ahrefs reports as 25.7 percent fresher.\n- AI-cited content still averaged about 2.9 years old, so assistants also lean on long-lived pages rather than only recent ones.\n- Ahrefs notes Google has warned against changing publish dates without changing page content, so date edits alone are not a tactic.\n- The study measures the age of cited URLs, not the effect of updating a page, so it shows association and not cause.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "research", "ai-search" ] }, { "id": "https://seomadman.com/research/reddit-top-3-share-se-ranking-core-update-study", "url": "https://seomadman.com/research/reddit-top-3-share-se-ranking-core-update-study", "title": "Reddit's top-3 share after the May 2026 core update", "summary": "SE Ranking tracked 100,000 keywords in 20 niches and measured Reddit's share of Google's top three results. This piece reviews its method and limits.", "content_text": "SE Ranking tracked 100,000 keywords across 20 niches from one New York location and found Reddit's share of Google's top three organic positions rose from 8.56% to 10.24% after the May 2026 core update. Reddit held the number one result for 13,872 keywords, up from 8,993, and grew in all 20 niches.\n\n- SE Ranking measured Reddit's share of all top-three organic positions at 10.24% after the May 2026 core update, up from 8.56% after the March update.\n- The number of tracked keywords where Reddit held the number one result rose from 8,993 to 13,872, a 54% increase.\n- Reddit's top-three share grew in all 20 niches in the sample, so the gain was not confined to one topic area.\n- YouTube's share of top-three positions moved the other way, falling from 2.50% to 2.14% over the same two snapshots.\n- The data covers organic blue links only, for one US location and before-and-after snapshots, so it cannot speak to SERP features, other regions or AI answers.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "research", "core-updates" ] }, { "id": "https://seomadman.com/research/sistrix-ctr-serp-layout-study", "url": "https://seomadman.com/research/sistrix-ctr-serp-layout-study", "title": "Sistrix CTR study: how SERP layout changes position-1 clicks", "summary": "Sistrix analyzed 80 million keywords and found position 1 averages 28.5% CTR, ranging from 13.7% to 46.9% depending on the SERP layout.", "content_text": "Sistrix's analysis of more than 80 million keywords found Google position 1 earns a 28.5% average click-through rate, but that average hides a wide range: 46.9% with sitelinks, 34.2% for pure organic results and 13.7% when Google Shopping appears. SERP layout, not ranking position alone, decides how many clicks a result can earn.\n\n- Sistrix's study of more than 80 million keywords reports an average click-through rate of 28.5% at position 1, 15.7% at position 2 and 11% at position 3.\n- Position 10 averages only 2.5% click-through rate in the Sistrix data, so position 1 earns more than ten times as many clicks.\n- Position 1 click-through rate ranges from 46.9% with sitelinks to 13.7% when Google Shopping results appear on the page.\n- Sistrix concludes that SERP layout determines how many potential clicks an organic result can get for a keyword.\n- A single average CTR curve is a poor forecasting input, because the same position earns very different click shares under different layouts.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "research", "technical-seo" ] }, { "id": "https://seomadman.com/studies/google-merchant-center-feed-optimization-uproer", "url": "https://seomadman.com/studies/google-merchant-center-feed-optimization-uproer", "title": "Merchant Center feed fixes and three click gains, per Uproer", "summary": "Uproer's published case study reports feed-level Google Merchant Center changes lifted clicks 130%, 116% and 3,500 for three unnamed ecommerce brands.", "content_text": "Agency Uproer's published case study reports that Google Merchant Center feed changes lifted clicks for three ecommerce clients: 130% for a supplements brand in three months, 116% for a fashion brand in 28 days, and 3,500 incremental clicks for a bakeware brand. The clients are unnamed and the figures are the agency's own.\n\n- Uproer reports a 130% rise in free Merchant Center listing clicks for a supplements brand within three months of rewriting descriptions and titles.\n- The fashion brand moved from Shopify's built-in Google feed app to Simprosys and, per the agency, saw clicks up 116% and impressions up 263% after 28 days.\n- The bakeware brand added sizes to titles and used the product_detail attribute, earning 3,500 incremental organic listing clicks year over year in three months.\n- All three clients are anonymous and the numbers are self-reported by the agency, so they are documented claims, not independently audited results.\n- The source states percentage lifts and one absolute click gain, but no baseline click counts, so the before values here are set to zero.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "studies", "technical-seo" ] }, { "id": "https://seomadman.com/studies/hcu-recovery-junkcarsus-seoprofy-case", "url": "https://seomadman.com/studies/hcu-recovery-junkcarsus-seoprofy-case", "title": "JunkCarsUs: HCU recovery from 30-50 to 400-500 clicks a day", "summary": "SeoProfy's case study on JunkCarsUs: after the September 2023 HCU, pruning pages and rewriting templates preceded a rise to 400-500 daily clicks.", "content_text": "SeoProfy reports that JunkCarsUs, a US car removal site hit by the September 2023 Helpful Content Update, went from 30-50 to 400-500 organic clicks per day. The agency cut pages from about 4.5K to 2.7K, removed a Spanish version and rewrote templates. First gains came with the August 2024 core update.\n\n- SeoProfy says JunkCarsUs fell to roughly 30-50 clicks per day after the September 2023 HCU and the March 2024 core and spam updates.\n- The first ordinary fixes, covering technical cleanup, review blocks and disavows, did not stop the decline, according to the case study.\n- The turn came after a more aggressive round that cut pages from about 4.5K to 2.7K and switched off the Spanish version.\n- SeoProfy reports daily clicks of 400-500 at publication, with first movement coinciding with the August 2024 core update.\n- The figures are the agency's own client data, and the case study does not isolate which single change drove the recovery.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "studies", "core-updates", "content-strategy" ] }, { "id": "https://seomadman.com/studies/jetoctopus-log-analysis-parameter-crawl-waste-case-study", "url": "https://seomadman.com/studies/jetoctopus-log-analysis-parameter-crawl-waste-case-study", "title": "Log analysis cut a jewelry site from 16M to 730K crawlable pages", "summary": "A JetOctopus case study: log files showed Googlebot spending crawl on filter, sort and pagination URLs, and closing them shrank the site to 730K pages.", "content_text": "JetOctopus's case study describes an unnamed jewelry shop whose category, filter, pagination and sort combinations produced about 16 million crawlable pages while relevant pages stayed unindexed for months. After log analysis, the agency recommended closing those combinations, and the site shrank to 730,000 pages, giving Googlebot far fewer URLs to spend its crawl on.\n\n- JetOctopus's case study reports a jewelry online shop with over 16 million pages but only 342,000 monthly visits, an extreme mismatch.\n- Profitable pages on that site were not indexed for months while Googlebot spent its resources on canonical-tagged filter and pagination URLs.\n- The recommendation was to close category, filter, pagination and sort combinations, which cut the site to 730,000 pages.\n- The same article shows a real estate site where 23.5 million of 38 million monthly Googlebot requests went to AJAX content.\n- The client is unnamed and JetOctopus sells the log analyzer used, so treat this as the vendor's own account, not an audited result.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "studies", "technical-seo", "indexing" ] }, { "id": "https://seomadman.com/studies/nuvemshop-lcp-image-priority-case-study", "url": "https://seomadman.com/studies/nuvemshop-lcp-image-priority-case-study", "title": "Nuvemshop lifted good LCP from 57% to 96% by fixing image loading", "summary": "Nuvemshop's web.dev case study: removing lazy loading on top images and adding fetchpriority took good LCP from 57% to 96% across its stores.", "content_text": "Nuvemshop, an ecommerce platform, raised the share of its stores with good Largest Contentful Paint from 57% to 96% in one year. Its web.dev case study credits three fixes: removing lazy loading from top images, adding fetchpriority high, and removing CSS transitions. Mobile Google organic conversion rose 8.9% alongside.\n\n- Nuvemshop reports that the share of its stores with good Largest Contentful Paint rose from 57% to 96% in one year.\n- The share of its stores passing all Core Web Vitals thresholds rose from 48% to 72% over the same period, per the same case study.\n- Nuvemshop names three root causes: delayed CSS transitions, lazy loading on top-of-viewport images, and missing priority signals on the most important images.\n- Mobile visitors from Google organic search converted 8.9% better from session to paid order, and cart engagement rose 8.4%, according to Nuvemshop.\n- The case study is a first-party account with a percentage-only result, so read it as a directional data point, not a rate other sites should expect.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "studies", "technical-seo" ] }, { "id": "https://seomadman.com/studies/programmatic-seo-saas-100-pages-suso-digital", "url": "https://seomadman.com/studies/programmatic-seo-saas-100-pages-suso-digital", "title": "Programmatic SEO case: 100 pages, 398% traffic, per Suso Digital", "summary": "Suso Digital reports a software supply chain client grew organic traffic 398.49% in 18 months from about 100 template pages. What the case shows.", "content_text": "Suso Digital says about 100 template-built pages lifted a software supply chain client's organic traffic 398.49%, from 1,920 to 9,571 users per month, over eighteen months. Top-10 keywords rose from 2,021 to 3,381. The client is unnamed and the agency published the case itself, so the figures are self-reported.\n\n- Suso Digital reports monthly organic users for an unnamed software supply chain company rose 398.49%, from 1,920 to 9,571, in about 18 months.\n- The agency says it created just over 100 pages through programmatic SEO for this client, a small page count compared with the scale usually associated with the tactic.\n- Keywords ranking in Google's top 10 rose from 2,021 to 3,381, an increase of about 67%, which matches the +67% figure on the case page.\n- The client is unnamed and the figures are the agency's own, so the case is documented but not independently verifiable.\n- The write-up describes other workstreams too, so the traffic gain cannot be credited to the templated pages alone.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "studies", "content-strategy" ] }, { "id": "https://seomadman.com/tools/hreflang-sitemap-xml-generator", "url": "https://seomadman.com/tools/hreflang-sitemap-xml-generator", "title": "Hreflang XML sitemap generator", "summary": "Build a sitemap with xhtml:link hreflang annotations for localized pages, with every alternate listed on every URL as Google's documentation requires.", "content_text": "This generator turns a list of language codes and URLs into a sitemap with hreflang annotations, the sitemap alternative to link tags in the page head. Each page URL lists every alternate version of itself, including itself, as an xhtml:link element, and an optional x-default gives a fallback. It checks codes and URLs before you publish.\n\n- Google supports hreflang annotations in an XML sitemap as an alternative to link tags in the page head or HTTP headers.\n- Each url element must list every alternate version of the page as an xhtml:link, including the page itself.\n- The sitemap needs the xhtml namespace declared on its urlset element for the xhtml:link elements to be valid.\n- The x-default value is optional and points at a fallback page for visitors whose language matches none of the listed versions.\n- Language codes use ISO 639-1 and an optional ISO 3166-1 Alpha-2 region, so en and en-GB are valid but eng is not.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "tools", "technical-seo", "indexing" ] }, { "id": "https://seomadman.com/tools/http-status-code-crawl-handling-checker", "url": "https://seomadman.com/tools/http-status-code-crawl-handling-checker", "title": "HTTP status code crawl handling checker", "summary": "Paste status codes and see how Google's crawlers document handling each one, plus a redirect hop count against Google's 10-hop default.", "content_text": "This checker takes a list of HTTP status codes and shows what Google documents doing with each: 301 and 308 as strong redirect signals, 302 and 307 as weak ones, 404 and 410 as not found, and 429 plus 5xx as server errors that slow crawling. It counts redirect hops against the default limit of 10 that Google's crawlers follow.\n\n- Google treats a 301 as a strong signal that the redirect target should be processed, and a 302 as a weak one.\n- A 307 is handled like a 302 and a 308 like a 301, even though the codes mean different things to other clients.\n- Google's crawlers follow up to 10 redirect hops by default, so a longer chain risks never reaching its destination.\n- A 429 is treated as a server error that signals overload, the same as 5xx codes, and makes Google's crawlers slow down.\n- A 2xx status only lets Google consider the content for processing; it does not guarantee the page gets indexed.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "tools", "technical-seo", "indexing" ] }, { "id": "https://seomadman.com/tools/link-rel-attribute-auditor", "url": "https://seomadman.com/tools/link-rel-attribute-auditor", "title": "Link rel attribute auditor: sponsored, ugc, nofollow", "summary": "Audit the links in pasted HTML against Google's rel qualifiers: sponsored for paid links, ugc for user content and nofollow when neither applies.", "content_text": "This auditor reads the links in pasted HTML and checks each rel attribute against the three qualifiers Google documents: sponsored for advertisements and paid placements, ugc for user-generated content, and nofollow when neither fits. Choose the kind of content the links sit in, and it flags external links missing the qualifier that content calls for.\n\n- Google documents three link qualifiers, sponsored, ugc and nofollow, and links carrying them will generally not be followed.\n- Google asks for rel=\"sponsored\" on advertisements and paid placements, and no longer recommends nofollow alone for paid links.\n- Links in comments, forum posts and other user-generated content should carry rel=\"ugc\" according to Google.\n- Several rel values can share one attribute, separated by spaces or commas, so ugc nofollow and ugc,nofollow are both accepted.\n- Values such as noopener and noreferrer are valid HTML but are not Google link qualifiers, so they do not mark a link as paid or user-generated.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "tools", "technical-seo", "content-strategy" ] }, { "id": "https://seomadman.com/tools/robots-directive-validator", "url": "https://seomadman.com/tools/robots-directive-validator", "title": "Robots meta tag and X-Robots-Tag validator", "summary": "Validate robots meta tags and X-Robots-Tag headers against the rules Google documents, and see the combined result for Googlebot.", "content_text": "This validator reads pasted robots meta tags and X-Robots-Tag header lines and checks each rule against the ones Google documents: noindex, nofollow, none, nosnippet, max-snippet, max-image-preview, max-video-preview, unavailable_after and more. It flags bad values, retired rules and conflicts, then shows the combined result Googlebot ends up with.\n\n- Google documents the robots meta tag names robots, googlebot and googlebot-news, and treats the header, agent name and values as case insensitive.\n- When robots rules conflict, Google applies the more restrictive one, so nosnippet wins over max-snippet.\n- The none rule is equivalent to noindex combined with nofollow, so listing both alongside it is redundant.\n- The max-image-preview rule accepts only none, standard or large, and max-snippet and max-video-preview take a whole number of at least minus one.\n- The noarchive rule no longer controls anything in Google Search, because the cached link feature it governed has been removed.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "tools", "technical-seo", "indexing" ] }, { "id": "https://seomadman.com/tools/url-normalizer-duplicate-variant-finder", "url": "https://seomadman.com/tools/url-normalizer-duplicate-variant-finder", "title": "URL normalizer and duplicate variant finder", "summary": "Normalize a list of URLs by RFC 3986 rules and your own policy, then see which URLs collapse into the same page and need one canonical.", "content_text": "This tool normalizes a list of URLs with the RFC 3986 rules for case, percent-encoding, dot segments and default ports, then applies policy options you choose: https, www, trailing slash, tracking parameters, parameter order and fragments. URLs that collapse to the same form are grouped, so you can see which variants need a single canonical.\n\n- RFC 3986 defines syntax-based normalization: lowercase the scheme and host, uppercase percent-encoding hex, decode unreserved characters and remove dot segments.\n- RFC 3986 also allows scheme-based normalization, such as dropping a default port and treating an empty path as a single slash.\n- Choices such as dropping www or a trailing slash are site policy rather than RFC rules, so the tool labels them as options.\n- Google names redirects and rel=\"canonical\" as strong signals for the canonical URL, and a sitemap as a weak one.\n- Google prefers the HTTPS page over an equivalent HTTP page as canonical, except when there are issues or conflicting signals.", "date_published": "2026-09-18T00:00:00.000Z", "date_modified": "2026-09-18T00:00:00.000Z", "tags": [ "tools", "technical-seo", "indexing" ] }, { "id": "https://seomadman.com/news/ga4-custom-dashboards-launch", "url": "https://seomadman.com/news/ga4-custom-dashboards-launch", "title": "Google Analytics rolls out drag-and-drop custom dashboards", "summary": "GA4 Dashboards let you drag dimensions and metrics onto a grid canvas, combining six chart types into one KPI page instead of separate reports.", "content_text": "Google Analytics is rolling out Dashboards, a feature that lets you build a custom, single-page KPI view from GA4 data instead of navigating separate reports. It uses a grid canvas with drag-and-drop chart creation, manual placement and alignment, and a card ceiling of 15 on standard properties and 30 on GA4 360 properties. Access is still spreading account by account.\n\n- Google Analytics is rolling out Dashboards, a custom grid canvas that consolidates GA4 metrics into one page.\n- You build charts by dragging dimensions and metrics onto the canvas, then place and align them manually.\n- The feature supports six chart types: score cards, table charts, line charts, bar charts, donut charts and funnel charts.\n- A standard GA4 property is capped at 15 cards per dashboard, while a GA4 360 property is capped at 30.\n- A saved dashboard can publish directly to the left navigation menu, skipping the library step.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "news", "technical-seo" ] }, { "id": "https://seomadman.com/news/search-central-live-india-bengaluru", "url": "https://seomadman.com/news/search-central-live-india-bengaluru", "title": "Search Central Live returns to India for a Bengaluru date in October", "summary": "Google's Search Central team holds a free, one-day Search Central Live event in Bengaluru on October 30, 2026, its first India event since 2023.", "content_text": "Google is holding a free, one-day Search Central Live conference in Bengaluru, India on October 30, 2026, from 10:00 to 17:30 local time at the Radisson Blu Bengaluru Outer Ring Road. It is Google's first India event in the series since 2023. Applications to attend close October 15, 2026, and seats are assigned by selection, not first-come-first-served.\n\n- Search Central Live comes to Bengaluru, India on October 30, 2026, running 10:00 to 17:30 local time.\n- This is Google's first Search Central Live event in India since 2023.\n- The event targets site owners, developers and SEO practitioners, with Google's own Search engineers speaking.\n- Applications close October 15, 2026, and seats go through a selection process rather than first-come-first-served registration.\n- Nothing from the event is streamed or recorded, so the material exists only for attendees in the room.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "news", "technical-seo" ] }, { "id": "https://seomadman.com/news/search-console-ai-report-logging-bug", "url": "https://seomadman.com/news/search-console-ai-report-logging-bug", "title": "Logging bug undercounted impressions in Search Console's AI report", "summary": "A logging error cut reported impressions in Search Console's Generative AI report from August 13, 2026, Google says, with no real drop in visibility.", "content_text": "A logging error made Search Console's Generative AI performance report show fewer impressions starting August 13, 2026. Google confirmed the drop is a data logging issue only, not a real change in Search visibility, and said it would add an annotation inside Search Console to flag the affected dates.\n\n- A logging error cut reported impressions in the Generative AI performance report for data starting August 13, 2026.\n- Google confirmed on its data anomalies page that the issue affects data logging only, not actual Search visibility.\n- John Mueller said on Bluesky that Google was working on the fix and would add an annotation inside Search Console to flag the affected dates.\n- Search Engine Land reported the data was fixed a few days after the issue was first reported.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "news", "search-console", "ai-search" ] }, { "id": "https://seomadman.com/news/september-15-ranking-volatility-spike", "url": "https://seomadman.com/news/september-15-ranking-volatility-spike", "title": "Google ranking volatility spikes sharply, no update confirmed", "summary": "SEO chatter and rank-tracking tool activity spiked hard on 15 September 2026. Barry Schwartz suspects an update is starting; Google has confirmed nothing.", "content_text": "Search Engine Roundtable reported a sharp spike in SEO chatter and third-party rank-tracking tool activity starting 15 September 2026. Barry Schwartz suspected it could be the start of a new Google update, though he doubted Google would confirm it soon. Google had made no announcement as of the report, so this remains unconfirmed volatility, not a verified update.\n\n- Search Engine Roundtable reported a huge spike in SEO community chatter beginning 15 September 2026.\n- Third-party rank-tracking tools also showed a sharp jump in activity over the same hours.\n- Barry Schwartz said he suspected the volatility could mark the start of a new Google update.\n- Schwartz said he doubted Google would confirm the update soon, even if one was underway.\n- Google had not announced or confirmed anything as of the report, so the event stays unverified.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "news", "core-updates" ] }, { "id": "https://seomadman.com/news/september-update-reversed-days-later", "url": "https://seomadman.com/news/september-update-reversed-days-later", "title": "An unconfirmed September update hit big sites, then fully reversed", "summary": "Glenn Gabe tracked large sites dropping hard around September 3-4, 2026, then surging back on September 13. Google confirmed neither move.", "content_text": "Analyst Glenn Gabe tracked a number of large sites losing significant Search and Discover traffic starting around September 3 into September 4, 2026, with evidence pointing to a change dated around September 2. On September 13, the impact on a number of those sites fully reversed, with traffic surging back. Google confirmed neither the drop nor the reversal.\n\n- Glenn Gabe tracked a number of large sites hit hard starting around September 3 into September 4, 2026.\n- Evidence he reviewed pointed to a change dated around September 2, ahead of the visible impact.\n- Some affected sites lost traffic in both Google Search and Google Discover at the same time.\n- On September 13, 2026, the impact on a number of those sites completely reversed and traffic surged back.\n- Google confirmed nothing about this event at any point, neither the drop nor the reversal.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "news", "core-updates" ] }, { "id": "https://seomadman.com/reports/ai-overview-citation-source-overlap", "url": "https://seomadman.com/reports/ai-overview-citation-source-overlap", "title": "How much AI Overview citations overlap with the top 10", "summary": "BrightEdge's own year-long tracking: only 17% of AI Overview citations also rank in the organic top 10, flat since last year, with wide swings by industry.", "content_text": "BrightEdge tracked AI Overview citations against organic rankings across nine industries from February 2025 to February 2026. Only about 17% of cited sources also rank in the organic top 10, a figure that has barely moved in a year. Healthcare and B2B tech sit near 23%, while restaurants and finance sit under 12%.\n\n- BrightEdge found only about 17% of sources cited in AI Overviews also rank in the organic top 10, roughly unchanged from a year earlier.\n- Five out of six AI Overview citations pull from content that does not appear on page 1 of traditional organic results, according to the same tracking.\n- Top-10 overlap ranges from 24.0% in healthcare down to 9.3% in restaurants, a gap of more than 14 percentage points across the nine industries BrightEdge tracks.\n- Entertainment, travel and e-commerce saw the largest gains in overlap over the year, rising from single digits to the mid-teens even as the overall average barely changed.\n- AI Overview presence itself grew from roughly 30% to 48% of BrightEdge's tracked queries over the same twelve months, a 58% increase.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "reports", "ai-search" ] }, { "id": "https://seomadman.com/reports/ai-referral-traffic-by-industry-2026", "url": "https://seomadman.com/reports/ai-referral-traffic-by-industry-2026", "title": "AI referral traffic to publishers doubled in a year, Similarweb finds", "summary": "Similarweb measured 770.7 million monthly AI referral visits worldwide, up 117.4% year over year, led by marketplaces and news.", "content_text": "Similarweb's traffic panel measured 770.7 million average monthly visits sent from AI platforms to websites worldwide between June 2025 and May 2026, up 117.4% year over year. ChatGPT sent over 80% of that referral traffic to the top 1,000 global domains. Marketplaces led every tracked industry at 46.8 million monthly visits, growing 237.3%, with news and travel tied just behind.\n\n- Similarweb's worldwide traffic panel measured 770.7 million average monthly AI referral visits between June 2025 and May 2026, more than double the prior year's figure.\n- ChatGPT accounted for over 80% of AI referral traffic to the top 1,000 global domains, per Similarweb, with Gemini, Claude, Perplexity and Grok splitting the rest.\n- Marketplaces received the most AI referral traffic of any tracked industry, at 46.8 million average monthly visits, up 237.3% year over year.\n- News and travel sites tied for second, each receiving 44.5 million average monthly visits, both up 115.6% year over year.\n- Beauty had the smallest tracked AI referral volume at 2.0 million monthly visits, despite the fastest relative growth of any industry at 312.5%.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "reports", "ai-search" ] }, { "id": "https://seomadman.com/reports/core-web-vitals-by-cms-platform-2025", "url": "https://seomadman.com/reports/core-web-vitals-by-cms-platform-2025", "title": "Core Web Vitals by CMS, 2025: Duda and Wix lead, WordPress lags", "summary": "The 2025 Web Almanac's CMS chapter breaks Core Web Vitals pass rates down by platform: Duda at 85%, Wix up 14 points to 74%, WordPress flat at 45%.", "content_text": "HTTP Archive's 2025 Web Almanac measured Core Web Vitals pass rates for ten CMS platforms using July 2025 CrUX field data. Duda led at 85% of mobile sites passing, TYPO3 followed at 79%, and Wix jumped 14 points to 74%. WordPress stayed near the bottom at 45%, and Weebly was the only platform to decline, slipping to 47%.\n\n- Duda had the highest mobile Core Web Vitals pass rate of any CMS platform in the 2025 Web Almanac, at 85%, followed by TYPO3 CMS at 79%.\n- Wix posted the largest year-over-year gain of any platform, rising 14 percentage points from 55% to 74% of mobile sites passing all three metrics.\n- WordPress passed Core Web Vitals on only 45% of mobile sites, among the lowest of the ten platforms measured, though it still improved roughly 4 points year over year.\n- Weebly was the only platform in the study to lose ground, declining to a 47% pass rate.\n- The gap is widest on INP: Squarespace led at 96% while 1C-Bitrix and Weebly tied for last at 71%, a 25-point spread the Almanac calls unsolved.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "reports", "technical-seo" ] }, { "id": "https://seomadman.com/reports/schema-org-type-adoption-web-almanac-2024", "url": "https://seomadman.com/reports/schema-org-type-adoption-web-almanac-2024", "title": "WebSite leads schema.org type adoption at 12.73% of mobile pages", "summary": "The 2024 Web Almanac breaks JSON-LD down by schema.org type: WebSite tops out at 12.73% of mobile pages, Organization at 7.16%, Product at just 0.77%.", "content_text": "HTTP Archive's 2024 Web Almanac structured-data chapter broke JSON-LD adoption down by individual schema.org type across its June 2024 crawl of nearly 17 million websites. WebSite led at 12.73% of mobile pages, followed by Organization at 7.16% and BreadcrumbList at 5.66%, while Product reached only 0.77% and Article just 0.18%.\n\n- The 2024 Web Almanac's structured-data chapter found WebSite the most common JSON-LD type, appearing on 12.73% of mobile pages.\n- Organization followed at 7.16%, BreadcrumbList at 5.66%, and LocalBusiness at 3.97%, all well ahead of content and commerce types.\n- Product structured data reached only 0.77% of mobile pages and Article just 0.18%, despite both types being central to Google's rich-result eligibility.\n- FAQPage adoption rose slightly on desktop, from 0.2% in 2022 to 0.6% in 2024, even after Google removed FAQ rich results from search in August 2023.\n- These per-type figures come from HTTP Archive's June 2024 crawl of 16,935,953 websites, the same dataset behind the chapter's headline 41% overall JSON-LD adoption rate.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "reports", "structured-data", "technical-seo" ] }, { "id": "https://seomadman.com/reports/serp-volatility-by-category-september-2026", "url": "https://seomadman.com/reports/serp-volatility-by-category-september-2026", "title": "Google SERP volatility by category, mid-September 2026", "summary": "Semrush Sensor's live index reads 2.1 out of 10 for mid-September 2026, but News and Finance run far hotter than Science or Health.", "content_text": "Semrush Sensor put Google's overall SERP volatility at 2.1 out of 10, inside its normal range, on September 16, 2026. That average hides a wide spread: News sat at 5.0, Finance at 4.1 and Food & Drink at 3.7, while Science and Health held at 0.8, showing a calm headline number can mask a rough day in one vertical.\n\n- Semrush Sensor's all-categories score read 2.1 out of 10 on September 16, 2026, which the tool itself labels normal range.\n- The News category scored 5.0 on the same day, more than double the all-categories figure and the highest of the 25 categories shown.\n- Finance scored 4.1 and Food & Drink scored 3.7, both well above the sitewide average despite the calm headline number.\n- Science and Health both scored 0.8, the lowest readings on the page, showing volatility is not evenly distributed across verticals.\n- Semrush's own methodology note says the score comes from a proprietary formula over a fixed daily keyword set, not a disclosed weighting anyone can reproduce.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "reports", "core-updates", "technical-seo" ] }, { "id": "https://seomadman.com/research/ai-overview-ctr-field-experiment", "url": "https://seomadman.com/research/ai-overview-ctr-field-experiment", "title": "A randomized field experiment on AI Overviews and organic clicks", "summary": "A randomized Chrome-extension experiment hid AI Overviews for 1,065 searchers. This piece analyzes Agarwal and Sen's SSRN study of the CTR impact.", "content_text": "Agarwal and Sen's field experiment randomly assigned 1,065 Prolific participants to see or not see Google AI Overviews across 68,089 real searches in January and February 2026. Outbound organic clicks fell 39.8% when an AI Overview appeared, zero-click searches rose 34.5%, and a companion survey found no measurable gain in perceived search quality.\n\n- The first randomized (not observational) field experiment on AI Overviews found outbound organic clicks fell 39.8% when an AI Overview appeared on the results page.\n- Zero-click searches rose 34.5% in the AI Overview condition, while sponsored ad clicks and overall search frequency stayed statistically unchanged.\n- Effect size was not uniform: roughly 88% of the click loss was concentrated in cases where the AI Overview occupied the very top of the page.\n- A companion satisfaction survey found no significant difference in perceived information quality between the AI Overview group and the group with it hidden.\n- The study used a custom Chrome extension to randomize exposure per search, a design that avoids the before/after confound of comparing traffic across a rollout date.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "research", "ai-search" ] }, { "id": "https://seomadman.com/research/author-byline-eeat-searchpilot-ab-test", "url": "https://seomadman.com/research/author-byline-eeat-searchpilot-ab-test", "title": "What happened when someone actually A/B tested author bylines", "summary": "SearchPilot ran a real controlled SEO split test on author bylines and bios. Result: no detectable ranking gain, and possibly a small loss.", "content_text": "SearchPilot, an agency known for statistically controlled SEO split tests, ran two separate tests adding author bylines, photos and bios to category pages on a review site to see if authorship content would lift organic traffic. Neither test found a detectable positive effect; the first test's forecast pointed toward a small negative impact instead.\n\n- SearchPilot ran a controlled SEO A/B split test, comparing control pages against variant pages with author bylines added, not a before-and-after comparison.\n- The first test added a prominent author image, byline and bio to category pages and found no detectable traffic impact, with the forecast leaning toward a small negative effect.\n- A second, smaller version of the byline without an author image also produced no detectable impact.\n- SearchPilot's own write-up states the test does not prove authorship content is worthless everywhere, only that it did not measurably help on this site.\n- The test is more methodologically rigorous than a correlation study, since it compares matched control and variant pages rather than an observational dataset.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "research", "content-strategy" ] }, { "id": "https://seomadman.com/research/canonical-vs-redirect-duplicate-signal-searchpilot-test", "url": "https://seomadman.com/research/canonical-vs-redirect-duplicate-signal-searchpilot-test", "title": "SearchPilot test: canonical alone matched a redirect for duplicates", "summary": "A SearchPilot split test added 301 redirects atop an existing canonical tag for trailing-slash duplicates. The result stayed inconclusive.", "content_text": "SearchPilot ran a controlled split test asking whether redirecting duplicate trailing-slash URLs consolidates ranking signal better than a canonical tag alone. The pages already carried a canonical to the non-trailing-slash version. Redirecting them produced an expected traffic drop on the redirected URLs and only a marginal, statistically inconclusive gain on the canonical target.\n\n- SearchPilot tested adding 301 redirects from trailing-slash URLs that already carried a canonical tag pointing to the non-trailing-slash equivalent.\n- Redirected URLs lost organic sessions as expected once their traffic funneled through to the canonical target instead.\n- The canonical target pages saw only a marginally positive traffic change, and SearchPilot itself calls that result inconclusive at a statistically significant level.\n- The client deployed the redirect site-wide anyway, citing alignment with best practice rather than a proven statistical gain.\n- SearchPilot does not publish the page count, test duration, or a confidence interval for this test, a gap consistent with its other published case studies.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "research", "technical-seo" ] }, { "id": "https://seomadman.com/research/internal-linking-crawl-budget-botify-case-study", "url": "https://seomadman.com/research/internal-linking-crawl-budget-botify-case-study", "title": "What a 10M-page site shows about links and crawl budget", "summary": "Botify's case study of a US auto marketplace ties internal-link depth to crawl activity: a 19x increase in crawling once linking was fixed.", "content_text": "Botify's published case study of a 10 million-page US auto marketplace found 99% of pages uncrawled by Google, largely because 1.3 million-plus pages carried only a single internal link. After restructuring internal links, breadcrumbs and the sitemap, crawl activity to strategic pages rose 19 times over six weeks, and organic traffic doubled within three months.\n\n- Botify's case study covers a US online vehicle marketplace of roughly 10 million pages, of which 9.9 million had never been crawled by Google.\n- More than 1.3 million indexable pages carried only a single internal link, a structural cause Botify identifies directly for the site's crawl gap.\n- After Botify's team restructured homepage links, breadcrumbs and the sitemap, crawl activity to the site's strategic pages rose 19 times within six weeks.\n- Known URLs fell by roughly half in the same six-week window, and the average number of inlinks per page rose, both measured directly rather than estimated.\n- Organic search traffic doubled within three months of the internal-linking changes, per Botify's own reporting of the engagement.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "research", "technical-seo" ] }, { "id": "https://seomadman.com/research/title-tag-rewrite-definitions-compared", "url": "https://seomadman.com/research/title-tag-rewrite-definitions-compared", "title": "Why title rewrite rate studies disagree: 33% vs 61% vs 76%", "summary": "Ahrefs found a 33.4% title rewrite rate, far below Zyppy's 61.6% and Search Engine Land's 76%. The gap traces to how each study defines a match.", "content_text": "Ahrefs' 2021 study of 953,276 top-10 pages found a 33.4% title rewrite rate, less than half Zyppy's 61.6% in 2022 and Search Engine Land's 76% in 2025. Ahrefs counted punctuation and brand-name changes as a match, not a rewrite; the later studies counted any wording change. That definitional choice, not just elapsed time, explains much of the spread.\n\n- Ahrefs' November 2021 study of 953,276 pages ranking in the top 10 found Google rewrote title tags 33.4% of the time, using the original tag the other 66.6%.\n- Ahrefs treated punctuation differences and added brand names as a match, not a rewrite, while later studies counted any wording change as a rewrite.\n- When Ahrefs did find a rewrite, it traced 50.76% of replacement titles back to the page's H1 tag, making the H1 the dominant fallback source Google pulls from.\n- Ahrefs found rewrite rate tracked pixel width, not character count: titles over 600 pixels wide were rewritten 46.12% of the time, up from 29.45%.\n- Three studies measuring the same behavior between 2021 and 2025 produced rewrite rates of 33.4%, 61.6% and 76%, too wide a spread to blame on elapsed time.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "research", "technical-seo" ] }, { "id": "https://seomadman.com/studies/coinstats-core-web-vitals-organic-traffic", "url": "https://seomadman.com/studies/coinstats-core-web-vitals-organic-traffic", "title": "CoinStats fixed a Core Web Vitals bug, tripled search impressions", "summary": "DebugBear's case study: CoinStats found Base64 images were bloating 15,000 pages, fixed it, and saw Good URLs and search impressions both increase 3x.", "content_text": "CoinStats, a crypto portfolio tracker, had poor Largest Contentful Paint on many pages because it was embedding Base64 images directly in its HTML. After switching roughly 15,000 pages to conventional img src URLs, HTML download size was cut in half. DebugBear's published case study reports both the share of Good URLs and search impressions on those pages increased 3x.\n\n- CoinStats, a cryptocurrency price and portfolio tracking site, had a slow Largest Contentful Paint and a high bounce rate across many of its pages before the fix.\n- The root cause was Base64-encoded images embedded directly in the HTML document instead of standard img src URLs, which bloated page weight.\n- CoinStats updated around 15,000 pages to use img src URLs, which cut the HTML document's download size in half.\n- DebugBear's published case study reports both the share of \"Good URLs\" in Search Console and search impressions on those pages increased 3x after the fix rolled out.\n- The case study does not publish a timeframe, exact before and after scores, or specific URLs, so treat the multiplier as directional, not a precise rate.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "studies", "technical-seo" ] }, { "id": "https://seomadman.com/studies/homescience-tools-content-pruning-case-study", "url": "https://seomadman.com/studies/homescience-tools-content-pruning-case-study", "title": "Home Science Tools: pruning 200 blog pages, organic sessions up 104%", "summary": "Inflow's case study on HomeScienceTools.com: removing roughly 200 low-value blog pages preceded a 104% rise in sessions and 102% rise in transactions.", "content_text": "Agency Inflow's case study on client HomeScienceTools.com describes pruning about 200 blog pages, roughly 10% of the blog, chosen for having little organic traffic, few pageviews, no conversions and no backlinks. In the months after the August 2018 pruning, the agency reports organic sessions up 104%, transactions up 102%, and strategic content revenue up 64%.\n\n- Agency Inflow pruned roughly 200 pages, about 10% of client HomeScienceTools.com's blog, in early August 2018.\n- Pages were selected for removal based on having little or no organic traffic, pageviews, conversions, or backlinks.\n- Inflow reports a 104% increase in organic sessions and a 102% increase in transactions following the pruning.\n- The agency also reports a 64% increase in what it calls strategic content revenue over the same period.\n- The site's organic keyword footprint dipped slightly right after pruning, then grew substantially in the 90 days that followed, per the case study.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "studies", "content-strategy" ] }, { "id": "https://seomadman.com/studies/ifttt-internal-linking-traffic-case-uproer", "url": "https://seomadman.com/studies/ifttt-internal-linking-traffic-case-uproer", "title": "IFTTT's internal linking fix and its 33% YoY traffic gain", "summary": "Uproer's case study on IFTTT: searchbots reached only 59 pages via internal links until a fix, after which the site saw 33% YoY organic traffic growth.", "content_text": "Agency Uproer's published case study on IFTTT.com found searchbots could crawl only 59 pages through the site's internal links, because those links were JavaScript-rendered. An HTML sitemap, then server-side rendering and new navigation links, followed. Indexed pages rose nearly 3x in two weeks, and IFTTT saw 33% year-over-year organic traffic growth.\n\n- Uproer's case study describes client IFTTT, the SaaS automation platform, having hundreds of thousands of indexable pages but internal links rendered only in JavaScript.\n- A Screaming Frog crawl mimicking a searchbot's link-only path found just 59 pages reachable, a crawlability problem the agency treated as the root internal linking issue.\n- Launching an HTML sitemap so those links existed in raw HTML delivered, per the case study, nearly a 3x increase in indexed pages within two weeks.\n- The agency later added server-side rendering plus breadcrumb, sidebar and footer navigation as a longer-term site-wide interlinking fix, over a 6-month engagement.\n- Uproer attributes 33% year-over-year traffic growth to the engagement, but its own write-up says linking was prioritized alongside a broader audit, not tested alone.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "studies", "technical-seo" ] }, { "id": "https://seomadman.com/studies/marshfield-clinic-schema-markup-ctr-case-study", "url": "https://seomadman.com/studies/marshfield-clinic-schema-markup-ctr-case-study", "title": "Marshfield Clinic: 454% CTR gain from schema markup", "summary": "Schema App's customer story on Marshfield Clinic reports a 454% CTR increase when a Review Snippet was awarded, plus 32% from entity linking.", "content_text": "Schema App's customer story on Marshfield Clinic Health System states click-through rate rose 454% across physician URLs when a Review Snippet was awarded, and rose 32% on physician pages after entity linking at scale. The story names the client and vendor, but discloses no measurement dates or baseline click numbers behind either percentage.\n\n- Schema App's customer story states CTR increased 454% across Marshfield Clinic's physician URLs when a Review Snippet rich result was awarded.\n- The same story states a separate 32% CTR increase on physician pages after Schema App's External Entity Linking feature was rolled out across page categories.\n- Marshfield Clinic began the project after observing a 30% decline in organic traffic between 2021 and 2022, which is the stated trigger for the engagement, not a result of it.\n- The schema types involved are physician pages with Review Snippet and Local Business markup, plus entity linking to Wikipedia, Wikidata and Google's Knowledge Graph.\n- Neither CTR percentage is paired with a disclosed baseline rate or exact measurement window, so both read as vendor-reported lifts, not reproducible figures.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "studies", "structured-data" ] }, { "id": "https://seomadman.com/studies/simms-fishing-magento-shopify-migration", "url": "https://seomadman.com/studies/simms-fishing-magento-shopify-migration", "title": "Simms Fishing's Magento to Shopify migration: the real numbers", "summary": "Inflow's case study: Simms Fishing saw organic traffic up 126% YoY and non-brand category traffic up 5,642% after a Magento to Shopify move.", "content_text": "Agency Inflow published a named case study on Simms Fishing's January 2022 Magento to Shopify migration. Organic traffic was up 126% year over year roughly 10 months out. Within the first 4 months, non-brand category page traffic was up 5,642% and the non-brand keyword footprint was up 1,095%, driven by splitting consolidated category pages into narrower, more specific ones.\n\n- Inflow's case study reports Simms Fishing's organic traffic grew 126% year over year in the roughly 10 months following its January 2022 Magento to Shopify migration.\n- In the first 4 months after launch, non-brand category page traffic rose 5,642% and the non-brand keyword footprint rose 1,095%, per the agency's published figures.\n- The gains are attributed mainly to splitting broad category pages, such as one combined \"fishing shirts\" page, into narrower pages for hoodies, T-shirts and flannels.\n- Inflow frames the result as going beyond ordinary migration maintenance, crediting new content and URL structure work done alongside the platform switch.\n- The case study does not publish raw session or keyword-count numbers behind the percentages, so the figures here are relative gains, not absolute traffic volumes.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "studies", "technical-seo" ] }, { "id": "https://seomadman.com/tools/faceted-navigation-parameter-analyzer", "url": "https://seomadman.com/tools/faceted-navigation-parameter-analyzer", "title": "Faceted navigation parameter analyzer", "summary": "Check a filtered listing URL's query parameters against Google's own documented faceted navigation guidance, and get a suggested robots.txt block.", "content_text": "This tool parses a faceted or filtered listing URL with URLSearchParams and checks it against Google's crawling infrastructure documentation on faceted navigation: parameter separator style, parameter count, session identifiers, and repeated path segments. It also drafts a robots.txt block for the facet parameters it recognizes, all inside your browser.\n\n- Google's own documentation names faceted navigation and session identifiers as the leading cause of the infinite URL spaces that waste crawl budget.\n- Google recommends the standard & separator between parameters, because commas, semicolons and brackets are hard for crawlers to detect as separators.\n- Google publishes no exact parameter-count threshold for when a faceted URL space becomes a problem, so this tool's count check is a labelled heuristic, not an official rule.\n- Blocking facet parameters in robots.txt is one documented option; using rel=canonical toward the unfiltered page is a documented but weaker alternative.\n- This tool only reads the one URL you paste; it cannot crawl your site to see how many real combinations your filters actually generate.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "tools", "technical-seo" ] }, { "id": "https://seomadman.com/tools/international-targeting-strategy-planner", "url": "https://seomadman.com/tools/international-targeting-strategy-planner", "title": "International targeting strategy planner", "summary": "Score a ccTLD, subdomain, subdirectory or URL-parameter structure against Google's own documented geotargeting pros and cons for your constraints.", "content_text": "This planner scores four URL structures, ccTLD, subdomain, subdirectory and URL parameters, against the pros and cons Google's own documentation publishes for geotargeting. Enter your country count, domain budget and server needs, and it ranks the options using Google's documented tradeoffs rather than a generic checklist, and explains why each one moved up or down.\n\n- Google's documentation lists four URL structures for geotargeting and explicitly marks URL parameters as not recommended.\n- A country-code domain gives the clearest geotargeting signal but can only target one country and costs more to run at scale.\n- A subdirectory is the lowest-maintenance option because it stays on one host, but it cannot support different server locations per region.\n- A subdomain is the middle ground: easy to set up and able to use different server locations, at the cost of a less obvious URL for users.\n- Google states it does not use IP-based content adaptation as a geotargeting signal, recommending hreflang, alternate URLs and explicit links instead.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "tools", "technical-seo" ] }, { "id": "https://seomadman.com/tools/meta-refresh-vs-redirect-checker", "url": "https://seomadman.com/tools/meta-refresh-vs-redirect-checker", "title": "Meta refresh vs. 301 redirect checker", "summary": "Parse a meta refresh tag's delay and target, and see which redirect signal Google actually reads from it: permanent at 0 seconds, temporary above 0.", "content_text": "This tool parses a meta refresh tag's delay and target URL, then classifies it against Google's own redirects documentation. A 0-second meta refresh is read as a permanent redirect signal, identical to a 301; any delay above 0 seconds is read as a temporary signal instead, the opposite of what most people assume a slower, gentler redirect would do.\n\n- Google's documentation places a 0-second meta refresh in the same permanent-redirect category as an HTTP 301, using it as a signal the target should be canonical.\n- A meta refresh with any delay above 0 seconds is read as a temporary redirect signal instead, so the original page, not the target, stays eligible to be treated as canonical.\n- Google's documentation ranks redirect methods by how reliably it can interpret them, with server-side HTTP redirects first and meta refresh second.\n- A JavaScript redirect and a crypto redirect are both named as less reliable than meta refresh, and Google recommends using them only when a server-side redirect is not possible.\n- This tool only parses the tag's text; it cannot check what your server actually returns for a real HTTP redirect.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "tools", "technical-seo" ] }, { "id": "https://seomadman.com/tools/pagination-canonical-conflict-checker", "url": "https://seomadman.com/tools/pagination-canonical-conflict-checker", "title": "Pagination and canonical conflict checker", "summary": "Check a paginated sequence's URLs and canonicals against Google's guidance, and catch the common mistake: consolidating every page to page 1.", "content_text": "This tool checks a paginated sequence, a list of page URLs and their declared canonicals, against Google's own pagination documentation. It catches the single most common mistake: every page's canonical pointing back at page 1, which tells Google to drop pages 2 and beyond from the index. It also flags duplicate URLs and page numbers hidden in a URL fragment.\n\n- Google's documentation states plainly not to use the first page of a paginated sequence as the canonical page for every other page in it.\n- Each page in a sequence needs its own unique URL, because Google treats paginated pages as separate URLs rather than one combined page.\n- Google ignores URL fragments, so a next-page link that only differs after a hash mark may look already-crawled and never get followed.\n- Google stopped using rel=next and rel=prev tags to understand pagination; they are harmless to leave in but do nothing for Google today.\n- This tool only evaluates the URLs and canonicals you type in; it cannot detect a page missing from the sequence entirely.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "tools", "technical-seo" ] }, { "id": "https://seomadman.com/tools/robots-txt-syntax-validator", "url": "https://seomadman.com/tools/robots-txt-syntax-validator", "title": "Robots.txt syntax validator", "summary": "Lint a whole robots.txt file for structural and syntax errors against RFC 9309 and Google's own interpretation of it, line by line.", "content_text": "This validator lints an entire robots.txt file line by line against RFC 9309 and Google's own interpretation of it: colon-separated fields, supported field names, paths that must start with a slash, absolute sitemap URLs, and rule lines with no preceding user-agent. It checks whether the file is well-formed at all, separate from testing one crawler against one path.\n\n- A valid robots.txt line is a field, a colon, and a value; a line with no colon is invalid and gets ignored.\n- Google's documentation supports exactly four fields: user-agent, allow, disallow and sitemap, and states other fields such as crawl-delay are not supported.\n- An allow or disallow path value must start with a slash to designate the root, per Google's documentation.\n- A sitemap value must be a fully qualified, absolute URL, not a path relative to the robots.txt file.\n- This tool checks the whole file's structure; it does not test a specific crawler against a specific URL path, which the site's separate robots.txt tester covers.", "date_published": "2026-09-16T00:00:00.000Z", "date_modified": "2026-09-16T00:00:00.000Z", "tags": [ "tools", "technical-seo" ] } ] }