# How Should You Structure a Website for SEO and AI Search? > For the complete documentation index, see [llms.txt](https://attensira.com/llms.txt). > Every page on this site is also served as Markdown: append `.md` to any URL, or send `Accept: text/markdown`. > How to structure a site so Google and AI engines find, understand and cite it, with hubs, URLs and internal links, plus how we restructured our own site. - Canonical page: https://attensira.com/blog/site-architecture-seo - Family: blog - Last updated: 2026-10-04 - Author: Karl-Gustav Kallasmaa (https://attensira.com/blog/author/karl) - Claims as data: https://attensira.com/claims.json?slug=site-architecture-seo **In short:** Good site architecture gives every question one owner page, groups owners under a few hub pages, keeps important pages within three clicks of the home page, and links related pages with descriptive anchor text. That lets crawlers find everything, tells Google which page answers which query, and gives AI engines one clear page to cite instead of five similar ones. **Section hubs AI crawlers fetched most · attensira.com · AI-crawler requests by the user agent each request claimed; 8 section hubs of at least 657 pages fetched · 5 Sept 2026 to 4 Oct 2026, n=143,601 requests** | Name | requests | | --- | --- | | /industries | 3,723 | | /blog | 3,442 | | /best | 3,149 | | /guides | 3,041 | | /bots | 3,033 | | /tools | 2,999 | | /compare | 2,925 | | /glossary | 2,807 | _attensira.com workspace AI traffic, read over the Attensira MCP · 4 Oct 2026 · 143,601 requests, none IP-verified_ Site architecture is how your pages are organised and connected: which pages exist, how their URLs are named, which pages link to which, and how many clicks it takes to reach each one. It decides what crawlers find, how Google understands which page answers which query, and increasingly which single page an AI engine picks to quote. This post covers the principles, a practical way to plan a structure, the AI-search angle most guides skip, and what we learned restructuring our own site this month. ## What makes a site architecture good for SEO? Five properties, in order of how often they are the problem. ### One owner page per question Every question your buyers ask should have exactly one page on your site that answers it best. When three blog posts, a glossary entry and a landing page all half-answer "what is X", search engines have to guess which one to show, and each page competes with the others. Pick the owner, make it the best answer, and merge or link the rest into it. ### Hubs that group owners Group owner pages under a small number of hub pages: a features hub, a guides hub, a compare hub. A hub lists its children, says in a sentence what each one answers, and links to all of them. Hubs give crawlers a short path to everything and give readers a map. ### Shallow depth Important pages should sit within about three clicks of the home page. Depth is not a ranking factor by itself, but deep pages get crawled less, linked less and found less. If a page you care about is five clicks down, link it from its hub and from related pages. ### Descriptive URLs and folders Use readable URLs with words that describe the page, and group similar topics in folders. Google says grouping similar topics in directories can help it learn how often the URLs in each directory change. Keep URLs stable: a renamed URL loses its signals until the new one earns them back. ### Links that explain themselves Internal links carry meaning. Google says anchor text tells users and Google something about the page being linked to, so "see our pricing" beats "click here". Link where a reader would naturally want to go next, not to every page from every page. ## How do you plan a site structure? Work from questions, not from your org chart. 1. **List the questions.** Pull them from Search Console queries, sales calls, support tickets and the questions you track in AI answers. Group duplicates, and note what each searcher wants from the page ([search intent](/blog/what-is-search-intent) decides whether the owner is a definition, a guide, a comparison or a pricing page). 1. **Assign one owner per question.** For each group, decide which page owns it, existing or new. 1. **Group owners into hubs.** Usually by job: what the product does, who it is for, how it compares, how to learn the topic, tools and reference. 1. **Map the URLs.** Hub at /{hub}, owners at /{hub}/{owner}. Avoid more than two levels for anything that matters. 1. **Draw the links.** Every owner links up to its hub, across to three to eight related owners, and on to the next step a reader would take. 1. **Decide what to retire.** Pages that answer nothing and earn nothing go. Pages that duplicate an owner merge into it. Questions with no owner at all are your gaps; a [content gap analysis](/blog/seo-content-gap-analysis) ranks them. ## What changes for AI search? Two things, and both push in the same direction as good classic architecture. **Engines cite passages from pages that clearly answer one question.** When an AI engine looks for a source, it wants a page whose opening answers the question directly. A structure with one owner per question produces exactly those pages. A structure with five overlapping posts produces five weak candidates and often none gets cited. **Engines reach beyond page one.** Ahrefs found that in March 2026 only 37.9% of URLs cited in Google AI Overviews also ranked in the top 10 for the same query. The engine fans a question out into related searches and picks pages that answer the sub-questions. Owner pages for the sub-questions, linked from a hub, are what it finds. There is one more practical point: make sure the main text and the links are in the HTML your server sends. Most AI crawlers do not run JavaScript, so navigation built only in scripts can hide your whole structure from them. ## How did we restructure our own site? We are rebuilding attensira.com right now, and the architecture decisions were made from our own measurements. The numbers below are our own workspace's page citations on ChatGPT and Perplexity, read over the Attensira MCP on 4 October 2026. **We are cutting the site to one owner per question.** Every URL gets one of three fates: it owns a question and is rewritten, it repeats a question another page owns and is deleted, or a question has no owner yet and a new page is written. Deleted pages simply go away; their best content is absorbed by the owner page that answers the same question. Roughly a third of the site's URLs are being removed. **Most posts earned nothing.** Of the 310 blog URLs in our page inventory, 3 received any ChatGPT or Perplexity citation in the 90 days to 4 October 2026, a separate, longer read than the month below. Many of the rest answered the same few questions with slightly different titles. That is the failure mode one-owner-per-question is designed to prevent. **The home page, hubs and comparisons earned the citations.** Across 4,524 ChatGPT and Perplexity answers from 4 Sep to 4 Oct 2026 (answers read on 24 of 31 days), the most-cited pages on our site were the home page at www (72 citations), /compare (61), /alternatives/profound and /pricing (58 each) and the head-to-head /compare/attensira-vs-profound (47). Engines leaned on pages that answer one clear question: what are the options, what does it cost, how does it compare. **Crawlers come in through the hubs.** The chart at the top of this page is the crawler side of the same picture. In the 30 days to 4 October 2026, AI crawlers made 143,601 requests to attensira.com, by the user agent each request claimed, and each section hub drew between 2,807 and 3,723 of them. No single blog post was among the 30 most-fetched pages. A hub is the short path a crawler takes to everything under it, which is why every owner page needs a link from its hub. **Some sections were never fetched at all.** In the same 30 days, at least 121 of our pages drew no AI-crawler request, against at least 657 that drew one or more. The /api-platform hub and all 18 pages under it were in that unfetched list. When a whole section is missing from the crawler log, start with its hub: is it linked from the home page and the navigation, and is it in the sitemap? **Architecture problems showed up in the citations.** In the same 4,524 answers, the home page was cited under two host names, 72 times at www and 47 at the bare address, splitting one page's citations across two addresses. And two pages we are retiring, /use-cases/saas-startups and /use-cases, were still cited 41 and 34 times. Our rule for that case: the owner page must carry the answer the old page was cited for, and we read the citations again after the change. This rewrite is part of the same work: this post now owns "site architecture SEO", and it links to the owners of the neighbouring questions instead of repeating them. ## Which pages did AI crawlers never fetch? Coverage, not volume. In the same 30 days of crawler traffic, AI crawlers fetched at least 657 of our pages and never fetched at least 121. One whole section sat in the unfetched list: the /api-platform hub and the 18 pages under it. A page no crawler fetches cannot be cited, whatever its content says. - **657** pages AI crawlers fetched at least once (143,601 requests, 5 Sep to 4 Oct) - **121+** pages with no AI-crawler fetch at all (at least 121 in the same 30 days) - **19** /api-platform hub and its 18 child pages, all unfetched (0 requests each in 30 days) attensira.com workspace AI traffic, read over the Attensira MCP on 2026-10-04 · by the user agent each request claimed, none IP-verified, 5 Sept 2026 to 4 Oct 2026. ## What does a good hub page contain? A hub is not a list of links with a heading on top. It is a page in its own right, and often the page an engine cites when the question is broad ("what are the options for X"). - **An answer to the broad question.** One paragraph that says what the hub covers and how the options differ, so the hub can stand alone as a source. - **Every child, with one line each.** The line says which question the child answers, in the reader's words, so a reader can pick without opening five tabs. - **An order that means something.** Most asked first, or the order a buyer goes through them. Alphabetical order is rarely the most useful. - **The way out.** The next step for a reader who has seen enough: a tool, a comparison, a trial. Keep hubs few. A site with forty hubs has the same problem as a site with no hubs: the reader cannot see the shape. ## What does a hub fix look like as a pull request? A hub update is a small diff, not a redesign. This example adds the paragraph that answers the broad question, then gives every child a link with a line saying which question it answers, most asked first. That is the shape an engine can lift and cite for a broad question. **example-co/site #87: Guides hub: answer the broad question and describe every child page** `app/guides/page.tsx` ```diff

Guides

-

Read our latest guides.

+

Choosing a CRM comes down to who uses it daily, what it must connect to, and what it costs after year one. Each guide below answers one of those questions.

``` _Example · example hub update_ ## How do you handle audiences, countries and languages? Audience pages ("for agencies", "for ecommerce") are owners too: each answers "is this for someone like me?". Give each one its own question and its own evidence, or merge them. A page that swaps the industry name into the same paragraphs answers nothing new and competes with its siblings. Countries and languages are different. A German page answering the German question is a separate owner, linked to its English counterpart with hreflang, not a duplicate. Translate the owners that matter commercially first; a translated long tail of thin pages multiplies the sprawl problem in every language. ## What are the common architecture mistakes? - **Topic sprawl.** A new post for every keyword variation, each thin, all competing. - **Orphans.** Pages nothing links to, usually created for a campaign and forgotten. - **Faceted URL explosions.** Filters and parameters that create thousands of crawlable near-duplicates. - **Navigation in JavaScript only.** Crawlers that do not run scripts never see the links. - **Restructuring for tidiness.** Moving live URLs to cleaner folders without a reason costs signals that take months to come back. - **Deleting without checking.** Removing a page that answers still cite, with no owner carrying its answer. - **Breadcrumbs that disagree with the URL.** A page whose breadcrumb says one hub while its URL and links say another sends mixed signals about where it belongs. Pick one parent and use it everywhere. - **Pagination that buries content.** Long archives split into page 2, page 3 and beyond push older posts deep. Link the evergreen owners from their hub instead of relying on the archive. ## How do you check your architecture is working? Watch three things after a change, and run the structural checks in our [site audit checklist](/blog/site-audit-checklist) (redirects, sitemap, click depth, orphans). In Search Console, the owner pages should gather the impressions their merged pages used to split. In a crawl, every owner should be reachable within a few clicks with no orphans. And in AI answers, the owner pages should be the ones cited for their questions. Attensira records which of your pages each answer cites, per engine, and re-reads your pages for changes, so you can see whether a restructure moved the citations or only the URLs. For how the page inventory and citation counts are kept, see [how Attensira works](/how-it-works). ## FAQ ### What is a flat site architecture? A structure where every important page is a few clicks from the home page, usually three or fewer, through hub pages. It is not literally flat; it is shallow. Deep chains of category pages hide content from crawlers and readers alike. ### Do URL folders matter for SEO? They help. Google says grouping similar topics in directories can help it learn how often URLs in each directory change, and descriptive words in a URL are useful to people. Do not restructure live URLs only to tidy folders; every moved URL loses signals for a while. ### How many internal links should a page have? There is no number to hit. Link where a reader would want to go next, with anchor text that says what they will find. Every page that matters should be linked from its hub and from a few related pages, and no page you care about should be an orphan. ### Should I delete thin pages or keep them? Merge pages that answer the same question into one owner, and delete pages that answer nothing. Before you remove anything, check whether Google or AI answers still cite it, and make sure the owner page carries the answer they were citing. ### Does site architecture matter for AI search? Yes. AI engines lift passages from pages that clearly answer one question. Five near-identical pages split the signal and give the engine five weak candidates. One owner page with a clear answer is easier to find, trust and cite. ## Sources Each entry below is one claim on this page. `retrieved` is the date the exporter or a human last saw the source say this, and nothing more: it is not a publication date and not a promise of re-verification. `cite` is the anchor on the canonical page where this claim is rendered. ### google-directories-grouping Google says using directories to group similar topics can help it learn how often the URLs in individual directories change, and recommends descriptive words in URLs. - source: https://developers.google.com/search/docs/fundamentals/seo-starter-guide (developers.google.com) - retrieved: 2026-10-04 - cite: https://attensira.com/blog/site-architecture-seo#claim-google-directories-grouping > Using directories (or folders) to group similar topics can help Google learn how often the URLs in individual directories change. ### google-anchor-text Google says anchor text tells users and Google something about the page being linked to. - source: https://developers.google.com/search/docs/fundamentals/seo-starter-guide (developers.google.com) - retrieved: 2026-10-04 - cite: https://attensira.com/blog/site-architecture-seo#claim-google-anchor-text ### ahrefs-aio-top10-38 37.9% of URLs cited in Google AI Overviews also appeared in the top 10 organic results for the same query; 863K keyword SERPs, 4M AI Overview URLs. - source: https://ahrefs.com/blog/ai-overview-citations-top-10/ (ahrefs.com) - retrieved: 2026-10-04 - cite: https://attensira.com/blog/site-architecture-seo#claim-ahrefs-aio-top10-38 ### dogfood-blog-citations-90d attensira.com workspace, read over the Attensira MCP on 2026-10-04 (page citations, 90 days, /blog/ pages), 310 blog URLs were in the page inventory and 3 of them received any citation in ChatGPT or Perplexity answers. - source: https://mcp.attensira.com/mcp (mcp.attensira.com) - retrieved: 2026-10-04 - cite: https://attensira.com/blog/site-architecture-seo#claim-dogfood-blog-citations-90d ### dogfood-cited-pages-hubs attensira.com workspace, read over the Attensira MCP on 2026-10-04 (4,524 answers, 2026-09-04 to 2026-10-04, 24 of 31 days read), page citations: www.attensira.com/ 72, /compare 61, /alternatives/profound 58, /pricing 58, bare attensira.com/ 47, /compare/attensira-vs-profound 47, /use-cases/saas-startups 41, /use-cases 34. - source: https://mcp.attensira.com/mcp (mcp.attensira.com) - retrieved: 2026-10-04 - cite: https://attensira.com/blog/site-architecture-seo#claim-dogfood-cited-pages-hubs ### dogfood-hub-crawler-fetches attensira.com workspace AI traffic, read over the Attensira MCP on 2026-10-04 (2026-09-05 to 2026-10-04, 143,601 AI-crawler requests by claimed user agent, none IP-verified): /industries drew 3,723 requests (7 agent tokens seen), /blog 3,442 (7), /best 3,149 (6), /guides 3,041 (5), /bots 3,033 (5), /tools 2,999 (6), /compare 2,925 (6), /glossary 2,807 (6). - source: https://mcp.attensira.com/mcp (mcp.attensira.com) - retrieved: 2026-10-04 - cite: https://attensira.com/blog/site-architecture-seo#claim-dogfood-hub-crawler-fetches ### dogfood-top-fetched-no-posts attensira.com workspace AI traffic, read over the Attensira MCP on 2026-10-04 (2026-09-05 to 2026-10-04): AI crawlers fetched at least 657 pages, and no individual blog post was among the 30 most-fetched. - source: https://mcp.attensira.com/mcp (mcp.attensira.com) - retrieved: 2026-10-04 - cite: https://attensira.com/blog/site-architecture-seo#claim-dogfood-top-fetched-no-posts ### dogfood-unfetched-pages attensira.com workspace AI traffic, read over the Attensira MCP on 2026-10-04 (2026-09-05 to 2026-10-04, 143,601 AI-crawler requests by claimed user agent, none IP-verified): at least 121 attensira.com pages had zero AI-crawler fetches, including /api-platform and 18 /api-platform/* pages (19 pages, 0 requests each), while at least 657 pages were fetched at least once. - source: https://mcp.attensira.com/mcp (mcp.attensira.com) - retrieved: 2026-10-04 - cite: https://attensira.com/blog/site-architecture-seo#claim-dogfood-unfetched-pages