Get Your Squarespace Site Into Google AI Overviews

The user-facing shape of AIO is consistent across queries: a paragraph or two of synthesised answer, often with a bulleted breakdown, and a set of source cards beneath or beside it. The cards are clickable and they carry referrer data, so AIO traffic appears in Google Search Console under the standard Search performance report — unlike ChatGPT, where most citation traffic arrives without a referrer. That measurement difference matters: AIO is the easiest AI surface to track, which makes it the right place to start an AI-search install.

Search Engine Land's 2026 GEO research¹ places AIO's reach at 2 billion monthly users by early 2026 and notes Gartner's projection of a 25% decline in traditional search volume this year as users redirect queries through AI-answer surfaces. BrightEdge's 12-month tracker through February 2026 measured AIO appearing on roughly 48% of tracked queries — up from about 31% a year earlier, a 58% relative increase⁸. Coverage is uneven by industry: education-related searches trigger AIO around 83% of the time, B2B tech around 82%, restaurants around 78%, while highly commercial transactional queries still skew lower.

The Squarespace-specific framing matters because the platform's defaults intersect with AIO's signals in three places: heading hierarchy on section-based templates, FAQ schema discipline after the May 2026 rich-result deprecation, and the author-entity layer that AIO uses as a credibility filter. The rest of this hub names each gap and points to the leaf that fixes it.

The mechanical chain is short. Googlebot crawls the page. The page enters the Search index. For a given query, Google's ranking system returns a candidate set. Gemini, the generative layer behind AIO, reads the top of that candidate set and synthesises an answer, citing three to six of the pages it drew from. Google's Search Quality Rater Guidelines⁵ — the 182-page document last updated on 11 September 2025 — defines the framework raters use to score quality, and E-E-A-T sits at the centre of that framework. Trust is the most important indicator.

The implication for Squarespace owners: every fix that helps your AIO appearance rate also helps your classical Google ranking and your ChatGPT citation rate. Restructured passages, named-source citations, clean schema, and a real author entity work across all three surfaces. The work is not duplicated; it is one investment that compounds.

The heading-hierarchy issue is the one most owners do not realise exists. Squarespace's own help center⁶ states that blog post titles render as H1 tags both on individual post pages and on the blog index page where they list. The platform's framing — "Google doesn't penalise you for having multiple H1 tags" — is technically accurate for classical ranking, but it understates the AIO impact. AIO's passage extraction reads cleaner DOM trees more reliably; a blog index with twelve H1 elements offers twelve competing signals about what the page is "about," and the engine picks one less predictably than from a page with one H1 and a clean H2 cascade. The full breakdown and the Code Injection patch live in the heading hierarchy leaf.

The schema gap is narrower but consequential. Squarespace's auto-emitted Article markup carries a headline, a date, and a basic publisher block. It does not include the Person entity with sameAs links that Google's Search Quality Rater Guidelines describe as a trust signal. The fix is supplemental Code Injection: a JSON-LD block that defines the founder Person on /founder/ with verifiable sameAs links to LinkedIn, GitHub, and any public profile, then references that Person URL from every article's author field. The pattern is covered in detail in the E-E-A-T leaf.

The fourth platform behaviour worth naming — less a gap than a missed lever — is Squarespace's lack of native control over the meta robots directive. AIO honours max-snippet, max-image-preview, and max-video-preview values that govern how much of your page the engine can use in the synthesised answer. Squarespace does not surface these in the SEO panel; they go in via Code Injection on Business plan or above. Setting them generously (max-snippet:-1, max-image-preview:large, max-video-preview:-1) opens the door to longer extracted passages, which correlates with longer citation cards.

Each of the next five sub-sections names one layer, the Squarespace-specific implementation, and the leaf where the technical depth lives. None of them is novel as a tactic — the same five appear in every serious AIO playbook. What this hub adds is the Squarespace-specific application: where Squarespace's editor lets you do the work natively, where it requires Code Injection, and where you have to work around a platform default.

Squarespace's official position⁶ is that Google does not penalise multiple H1 tags. The position is correct for classical ranking but optimistic for AIO — passage extraction is more reliable from pages with one clear primary heading. The html-structure leaf ships the audit script, the Code Injection patch, and the editor workflow for keeping new pages clean as you build them.

The 134-167 word band is the practical implementation of the "clear, direct answer at the top of every section" pattern Search Engine Land's 2026 GEO research¹ identifies. Long enough to carry a full thought, short enough to fit a citation card. The Squarespace blog-template walkthrough lives in the passages leaf; the same passage rule applies to ChatGPT citations and is covered in the ChatGPT content-format leaf.

Google's canonical FAQ documentation² states the eligibility restriction explicitly. Search Engine Journal's coverage of the change³ confirms the prior Google guidance that unused structured data does not harm Search. There is no canonical Google statement either confirming or denying continued FAQ schema influence on AIO citations — treat any blog post that promises a "3.2x citation lift from FAQ schema" with appropriate skepticism. The page-by-page Squarespace rule lives in the FAQ schema leaf.

 <script type="application/ld+json"> { "@context": "https://schema.org", "@type": "FAQPage", "mainEntity": [ { "@type": "Question", "name": "Does Squarespace support llms.txt natively?", "acceptedAnswer": { "@type": "Answer", "text": "Not at the root. The workaround is URL Mappings: create a Squarespace page at slug /llms, then add a 301 redirect /llms.txt -> /llms." } } ] } </script> 

The full pattern, including the exact JSON-LD block and the Squarespace Code Injection placement, is in the E-E-A-T leaf. The Person target page is already live at /founder/. Google's Article schema documentation⁴ is explicit about including type and URL/sameAs on author markup to "help Google understand the author better."

Squarespace's auto-emitted Article markup carries the original publish date but does not reliably bump dateModified when only the body content is edited. The fix is supplemental Code Injection on pages where freshness matters: a JSON-LD block that re-declares dateModified with the actual edit date. The meta strip on every editorial page in this site renders both dates and a "verified every 6 weeks" line, which is the visible side of the same discipline. Pair quarterly source-manifest re-verification with the modifiedISO bump and AIO will treat the page as actively maintained.

Search Console does not yet split AIO impressions from classical Search impressions in a dedicated report, but the click-through patterns differ enough that an attentive owner can tell the difference. AIO citations tend to produce high-impression, lower-CTR rows; classical top-three rankings produce lower-impression, higher-CTR rows. A 30-day comparison on your top twenty queries usually reveals which are AIO-cited and which are classical-ranked.

Squarespace's own AI Visibility tool¹⁰ runs branded and non-branded prompts on a plan-dependent cadence: every 14 days on Core and Business, every 7 days on Plus, Advanced, and Commerce. The tool primarily tracks ChatGPT mentions, not AIO citations, but the non-branded prompt results give you a directional signal for whether your category content is being read by the model layer. Use it as a triangulation source, not a primary metric. The full diagnostic pass is in the diagnose leaf and the 18-point reference list is in the checklist.