{"id":51709,"date":"2026-05-16T22:29:10","date_gmt":"2026-05-17T02:29:10","guid":{"rendered":"https:\/\/overcentral.com\/en\/ahrefs-study-schema-markup-has-minimal-impact-on-ai-search-visibility\/"},"modified":"2026-05-17T10:46:23","modified_gmt":"2026-05-17T14:46:23","slug":"schema-markup-ai-search-impact","status":"publish","type":"post","link":"https:\/\/overcentral.com\/en\/schema-markup-ai-search-impact\/","title":{"rendered":"Ahrefs Study: Schema Markup Has Minimal Impact on AI Search Visibility"},"content":{"rendered":"\n

Schema markup has long been considered a foundational element of technical SEO, with the promise of helping search engines understand content more deeply and display it more prominently. But as artificial intelligence reshapes how users find information, a critical question has emerged: does structured data actually boost visibility in AI-powered search results? A large-scale study from Ahrefs<\/a>, analyzing nearly 1,900 webpages over an eight-month period, suggests the answer is far more nuanced than many in the industry expected. The research, which examined citation patterns across Google AI Overviews, Google AI Mode<\/a>, and ChatGPT<\/a>, found that adding JSON-LD schema markup produced no meaningful increase in AI citations<\/a> on any of the platforms tested. While pages that are cited by AI systems are roughly three times more likely to include JSON-LD schema than those that are not cited, the study’s authors emphasize that this is a correlation, not a causal relationship. Websites that invest in structured data tend to invest in technical SEO more broadly, publish stronger content, and build greater domain authority. The implication is clear: schema markup is often a signal of overall quality rather than a direct driver of AI visibility.<\/p>\n\n\n\n

Breaking Down the Numbers Across Three Major AI Platforms<\/h2>\n\n\n\n

Ahrefs ran a carefully controlled experiment isolating the effect of schema markup on AI citations. The test group consisted of 1,885 webpages that had JSON-LD schema added between August 2025 and March 2026, while a control group of 4,000 URLs from other domains with similar citation levels but no schema changes served as a baseline. using a difference-in-differences statistical approach, the researchers measured citations 30 days before and 30 days after schema implementation. The results varied by platform but shared a central theme: none showed a material positive impact.<\/p>\n\n\n\n

Google AI Overviews Showed a Minor Decline<\/h3>\n\n\n\n

Perhaps the most surprising finding came from Google AI Overviews, where pages that added schema markup experienced a small but statistically significant decline of 4.6 percent in citations compared to the control group. The study advises against overinterpreting this drop, however, as the absolute numbers were minimal, averaging roughly 12 lost citations per day. Critically, both the test and control pages were already in a strong downward trend before schema was implemented, suggesting external factors were driving the decline. The slight acceleration in the test group may reflect noise rather than a genuine negative effect of schema.<\/p>\n\n\n\n

Google AI Mode and ChatGPT Registered Negligible Gains<\/h3>\n\n\n\n

On Google AI Mode, the addition of schema markup corresponded with a 2.4 percent increase in citations, while ChatGPT showed a 2.2 percent uptick. In both cases, the results were statistically indistinguishable from zero, meaning the observed changes could easily be attributed to random variation. For practical purposes, adding structured data did not move the needle on either platform. These findings align with separate tests conducted by searchVIU, which demonstrated that AI systems like ChatGPT and Google AI Mode extract only visible HTML text during real-time page retrieval and ignore hidden schema formats such as JSON-LD. If AI models are not even reading the markup during the citation process, there is little reason to expect a direct impact.<\/p>\n\n\n\n

\"Ahrefs
Ahrefs Study: Schema Markup Has Minimal Impact on AI Search Visibility<\/figcaption><\/figure>\n\n\n\n

Why Correlation Does Not Equal Causation in Schema and AI Citations<\/h2>\n\n\n\n

The Ahrefs data confirms that cited pages are more likely to feature JSON-LD schema, but digging into the underlying dynamics reveals why this pattern exists. Websites that implement structured data are typically those that also prioritize technical excellence across other dimensions. They tend to have clean site architecture, fast load times, well-optimized metadata, and strong internal linking. They publish authoritative, well-researched content and often have robust backlink profiles. These factors collectively contribute to a site’s trustworthiness and relevance, which is what AI models actually use to determine whether to cite a source. Schema markup, in this context, is a marker of a well-maintained site rather than a direct signal that influences AI citation decisions.<\/p>\n\n\n\n

Practical Implications for SEOs and Website Operators<\/h2>\n\n\n\n

For professionals managing content strategies in an AI-driven search landscape, the study offers several actionable takeaways. The most immediate is that if a page is already being crawled and cited by AI systems, adding JSON-LD schema will not further expand its reach in AI search<\/a> results. The effort is better directed elsewhere, such as improving content depth, authority, or topical coverage. However, this does not mean schema markup has lost its value. There remain strong, well-documented reasons to implement structured data, particularly for traditional Google search. Rich results such as recipe carousels, product stars, event listings, FAQ accordions, and breadcrumb enhancements all depend on proper schema markup and can meaningfully improve click-through rates in standard search results. Schema also plays a role in voice assistant discoverability and can strengthen a site’s presence in Google’s Knowledge Graph, which influences entity recognition across the ecosystem.<\/p>\n\n\n\n

The Potential Role of Schema for New or Low-Visibility Pages<\/h3>\n\n\n\n

For pages that have not yet been picked up by AI systems, schema markup may serve a different function. While the study did not find a direct boost to citations for already-visible pages, there is a plausible argument that structured data could help previously unnoticed content get crawled and indexed by AI systems in the first place. This is speculative but worth testing, particularly for sites operating in competitive verticals where every incremental advantage matters. The safest approach is to run a controlled experiment on your own site. Select five to ten pages that already receive some AI citations, implement JSON-LD schema, and compare their performance over 30 days against a matched control group making no changes. This site-specific testing is the only way to determine whether schema has any effect given your particular content, domain authority, and audience.<\/p>\n\n\n\n

Understanding the Study Design and Its Limitations<\/h2>\n\n\n\n

The robustness of the Ahrefs findings rests on a methodology designed to isolate the causal effect of schema markup. By using a large sample of test pages and an even larger control group, the researchers controlled for broad fluctuations in AI platform behavior that could otherwise mask or mimic the impact of schema. The difference-in-differences analysis is a well-established econometric technique that compares the change in citations for treated pages against the change for control pages, effectively differencing out common trends. Still, the study carries an important limitation. All of the pages analyzed had at least 100 AI citations before the experiment began. They were already in the consideration set of large language models. For sites that are not yet generating AI citations at scale, the findings may not generalize. Schema might still play a gatekeeping role for initial discovery, even if it does not amplify visibility once a page is established.<\/p>\n\n\n\n

\"Ahrefs
Ahrefs Study: Schema Markup Has Minimal Impact on AI Search Visibility<\/figcaption><\/figure>\n\n\n\n

Where Schema Markup Still Delivers Unquestionable Value<\/h2>\n\n\n\n

Despite the minimal impact on AI citations, the broader case for schema markup remains strong. In the traditional Google search ecosystem, structured data directly enables enhanced search features that can dramatically increase organic visibility. A product page with valid schema can appear with ratings, price, and availability right in the search results, often achieving higher click-through rates than standard listings. Event pages can surface dates and locations in an eye-catching format. Recipe pages can display cook time, calorie counts, and user reviews. These rich results are not speculative; they are proven traffic drivers. Schema also supports Google’s understanding of entities and relationships, which feeds into how the search engine interprets a site’s authority on specific subjects. For SEO professionals focused on long-term sustainable performance, maintaining up-to-date structured data is still a best practice, even if it does not directly influence ChatGPT citations.<\/p>\n\n\n\n

The Bottom Line on Schema and AI Search Visibility<\/h2>\n\n\n\n

The Ahrefs study delivers a clear verdict for the SEO community: adding schema markup will not, on its own, increase how often AI systems cite your content. The days of treating structured data as a shortcut to AI visibility should end. At the same time, the research should not be read as a reason to abandon schema altogether. The value of structured data for traditional search, rich results, voice assistants, and knowledge graph integration is well established and independent of AI citation performance. Website operators should continue implementing schema where it supports user experience and standard search goals, but they should set realistic expectations about what it can do for AI-driven discovery. The real levers for AI visibility remain content quality, topical authority, and domain trust, the same fundamentals that have always mattered in search. Schema is a supporting player, not a game changer, and treating it as such is the most strategic path forward.<\/p>\n\n\n\n

\"Ahrefs
Ahrefs Study: Schema Markup Has Minimal Impact on AI Search Visibility<\/figcaption><\/figure>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

Schema markup has long been considered a foundational element of technical SEO, with the promise of helping search engines understand content more deeply and display it more prominently. But as artificial intelligence reshapes how users find information, a critical question has emerged: does structured data actually boost visibility in AI-powered search results? A large-scale study […]<\/p>\n","protected":false},"author":6,"featured_media":85557,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"fifu_image_url":"https:\/\/cards.overcentral.com\/cards\/en\/51709.png","fifu_image_alt":"Ahrefs Study: Schema Markup Has Minimal Impact on AI Search Visibility","footnotes":""},"categories":[31],"tags":[],"class_list":["post-51709","post","type-post","status-publish","format-standard","has-post-thumbnail","category-technology"],"fifu_image_url":"https:\/\/cards.overcentral.com\/cards\/en\/51709.png","fifu_image_alt":"Ahrefs Study: Schema Markup Has Minimal Impact on AI Search Visibility","_links":{"self":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts\/51709","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/comments?post=51709"}],"version-history":[{"count":0,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts\/51709\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/media\/85557"}],"wp:attachment":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/media?parent=51709"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/categories?post=51709"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/tags?post=51709"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}