Skip to main content
LovedByAI
Page Builders

GenerateBlocks GEO/AEO: AI-Search Readiness

See how GenerateBlocks affects AI search visibility, why its minimal markup helps but its missing schema hurts, and how LovedByAI closes that exact gap.

Updated July 5, 2026
8 min read
By Jenny Beasley
Quick answer

GenerateBlocks sites start ahead on AI search: it outputs minimal, semantic HTML with almost no extra div wrapping, which is exactly the structure ChatGPT and Gemini read fastest within a limited crawl budget. What it does not do is generate structured data. GenerateBlocks was built to do less, on purpose, and schema.org markup was never part of that scope. LovedByAI is fully compatible with GenerateBlocks: it adds the discoverability layer the plugin never tried to solve, without adding weight back to a site you chose because it was light, combining JSON-LD, semantic HTML and heading improvements, metadata and entity reinforcement, and other HTML-level changes informed by proprietary LLM crawl research.

GenerateBlocks + GEO/AEO
GenerateBlocks + GEO/AEO

Is LovedByAI compatible with GenerateBlocks?

LovedByAI is fully compatible with GenerateBlocks and GenerateBlocks Pro, including sites running on GeneratePress or any other theme. It runs as a standard WordPress plugin and builds a behind-the-scenes AI discoverability layer without changing your Container blocks, Grid layouts, or global styles. It adds exactly the layer GenerateBlocks deliberately leaves out: the JSON-LD, semantic HTML, heading, metadata, entity, and other HTML-level signals that tell AI engines what your Organization, articles, and pages actually are, without adding the CSS or JS weight GenerateBlocks users specifically chose the plugin to avoid.

GEO/AEO pros and cons of GenerateBlocks

Strengths

  • Minimal HTML output means less for a crawler to wade through

    GenerateBlocks generates one CSS file per page, no inline styles, and no JavaScript for its core blocks. The HTML it outputs is close to what you'd hand-write: a Container block or two, then your actual content. AI crawlers working within a limited processing budget reach your real text faster because there's less markup standing between them and it.

  • Full control over HTML tags on Container, Headline, and Button blocks

    GenerateBlocks lets you set the actual HTML element for its core blocks. A Container can render as a div, section, article, header, nav, or aside, and the Headline block can be any heading level or paragraph tag. Used deliberately, this produces a real semantic structure instead of a stack of generic divs, which is exactly what an LLM uses to figure out how your content is organized.

  • No proprietary wrapper markup to work around

    Unlike page builders that wrap every element in framework-specific classes and extra container divs for their visual editor, GenerateBlocks blocks output close to the DOM structure you'd write by hand. There's less builder-specific noise between the page a crawler fetches and the content it's trying to understand.

Watch-outs

  • Zero native structured data

    GenerateBlocks has no built-in schema.org output of any kind: no Organization markup, no Article schema, no FAQPage or Breadcrumb schema. This isn't an oversight; it's outside what a block plugin focused on layout and performance is trying to do. But the result is the same as with any builder that skips it: nothing in your page tells an AI engine what your business is or what a page is answering.

  • Semantic HTML is opt-in, not automatic

    GenerateBlocks makes it possible to set a Container to render as a section or article and a Headline to render as an h2, but the default tag for most blocks is a plain div or the element you last used. If you build quickly without setting tags deliberately, you can end up with the same meaningless div stack an unstructured page builder would produce, just with less CSS attached to it.

  • No content-relationship or entity signals

    Clean markup tells a crawler where a heading starts and stops. It doesn't tell the crawler what your business is, who wrote the page, or how one article relates to another on your site. GenerateBlocks has no mechanism for any of that, because entity and authorship signals were never part of what a block-level layout plugin sets out to provide.

  • Manual design work means inconsistent structure across a site

    Because GenerateBlocks does less for you than a full page builder, structuring pages well depends on the person building them doing it consistently. A site built by several people, or built over several years, often ends up with some pages using clean semantic tags and others using whatever the default happened to be, which gives AI crawlers a mixed signal about how reliable your markup is site-wide.

How LovedByAI works with GenerateBlocks

  1. 1

    Install LovedByAI like any other WordPress plugin

    No code, no changes to your Container blocks or Grid layouts. Activate it alongside GenerateBlocks and GenerateBlocks Pro with no conflicts, since it doesn't modify anything the block plugin renders or add any CSS or JS of its own to the frontend.

  2. 2

    LovedByAI reads your existing block structure

    It scans your GenerateBlocks pages to identify what tags you're already using: which Containers are sections versus generic divs, which Headlines are real heading levels, and where your main content sits. It uses whatever semantic structure you've already built rather than asking you to rebuild anything.

  3. 3

    It builds the AI discoverability layer GenerateBlocks was never built to generate

    Organization, Article, and FAQPage schema get added to your site's head on every page, then LovedByAI reinforces semantic HTML, heading, metadata, entity, and other HTML-level signals behind the scenes based on proprietary LLM crawl research. This is the layer GenerateBlocks explicitly leaves out by design, and it is what tells AI engines what your content actually is.

  4. 4

    You get visibility into whether AI crawlers are reading it

    LovedByAI tracks when GPTBot, Google-Extended, and other AI crawlers visit your pages, so you can confirm that the combination of your clean GenerateBlocks markup and LovedByAI's schema is actually getting parsed, not just theoretically correct.

You chose GenerateBlocks to keep the page as light as it can be. It was never trying to be a page builder in the Elementor or Divi sense, and that's the whole point of it. It's a small set of Gutenberg blocks from the team behind GeneratePress, built around one idea: output as little HTML and CSS as the design actually requires, and nothing more. For AI search, that starting position is unusually good. It's also incomplete in a specific, predictable way.

You already made a tradeoff to get here: less visual hand-holding, more manual control, in exchange for a site that loads fast and renders clean. That same philosophy is why it has nothing to say about structured data. Schema.org markup was never the problem GenerateBlocks set out to solve, so it doesn't solve it, on purpose. Understanding both halves of that tradeoff is what determines whether a GenerateBlocks site actually gets read and cited by ChatGPT, Perplexity, or Gemini.

Minimal markup is a genuine GEO advantage

AI crawlers work within a limited context window. Every layer of markup between the start of the document and your actual content is processing spent on structure instead of meaning. This is where GenerateBlocks earns its reputation: independent testing has shown its blocks adding a fraction of the requests and page weight that heavier builders add for the same layout, with one CSS file per page and no plugin-injected JavaScript for core blocks (source: The Admin Bar, "Performance Implications of GenerateBlocks' New Features"). Less weight and fewer wrapping elements mean a crawler reaches your sentences sooner and burns less of its budget getting there.

The Container, Headline, Button, and Image blocks also let you set the actual HTML tag they render as, not just a visual style. A Container can be a section or an article instead of a div. A Headline can be an h2 instead of a styled paragraph. That control is there because GenerateBlocks cares about output quality, not just visual output, and a site that uses it deliberately ends up with a document structure an LLM can actually follow: this is a heading, these paragraphs belong to it, this is a distinct section.

Doing less by design means schema was never included

Here's the tradeoff. A plugin built around minimal footprint doesn't add features that don't serve that footprint, and structured data is one of them. GenerateBlocks generates no Organization schema, no Article schema, no FAQPage markup, not because it failed at it, but because schema generation isn't a layout concern and was never going to be added to a plugin whose whole reason for existing is doing less. The result is a site that can be extremely well-structured on the surface and still invisible to an AI engine trying to answer "what does this business do" or "does this page answer this question," because nothing on the page states either of those things in a form a model can extract with certainty.

The same minimalism that helps you with crawl budget also means nothing is enforcing consistency. Setting a Container to render as a section, or a Headline as an h2, is a choice you make block by block. Skip it under deadline pressure and you get a div with GenerateBlocks' clean CSS attached to it, structurally no better than a generic page builder output, just lighter. That inconsistency compounds across a site that's been built by more than one person or over more than one redesign.

The missing layer is schema, and it's the layer AI engines read

Go through the plugin's own WordPress.org listing and you'll see what it commits to and what it doesn't. GenerateBlocks describes itself around minimal CSS, simple HTML structure, and blocks like Container, Grid, and Text; there is no mention of schema.org output or structured data anywhere in what it does (GenerateBlocks on WordPress.org). That's consistent with the plugin's whole reason for existing. Structured data isn't a layout or performance feature, so a plugin built to do less on both fronts was never going to generate it.

That missing layer is the one AI engines lean on hardest. In a controlled test comparing near-identical pages, Search Engine Land found that "only the page with well-implemented schema appeared in an AI Overview and achieved the best organic ranking," while the versions with poor or no structured data didn't trigger an AI Overview at all (Search Engine Land). The authors call the result promising but not conclusive, and that caution is fair. The direction, though, matches how these systems work: schema states what a page is and what entity it belongs to, so a model doesn't have to infer it from prose and hope it guessed right.

So the honest picture is that GenerateBlocks gives you clean, fast markup that a crawler can read, and stops exactly where AI discoverability work begins. LovedByAI adds that layer: JSON-LD, semantic HTML and heading reinforcement, metadata and entity cues, and other HTML-level improvements based on proprietary LLM crawl research, with no CSS or JavaScript weight that would undo the reason you picked GenerateBlocks in the first place.

What LovedByAI fixes automatically

GenerateBlocks solves speed and markup minimalism, and it solves them well. It was never going to solve structured data, because that's a different problem with a different shape, and adding it would mean becoming the kind of heavier, more opinionated plugin that GenerateBlocks users specifically avoided by not choosing something else.

That's the gap LovedByAI closes. It doesn't rewrite your Container blocks, doesn't touch your Grid layouts, and doesn't add CSS or JavaScript to your frontend. It reads the semantic structure you've already built, however consistently or inconsistently you built it, then adds the Organization, Article, and FAQPage JSON-LD GenerateBlocks never aimed to provide, reinforces semantic HTML and heading signals, aligns metadata and entity cues, and makes other HTML-level improvements informed by proprietary LLM crawl research. You keep the footprint you chose GenerateBlocks for. You stop being invisible for the one reason speed alone can't fix.

Real results from GenerateBlocks + LovedByAI users

Used on 34+ GenerateBlocks sites

are already using LovedByAI alongside GenerateBlocks to get mentioned in ChatGPT and Google Gemini.

8 min

median time for LovedByAI system to fully AI-optimize a GenerateBlocks site

LovedByAI onboarding data, 2026

116%

increase in AI mentions on AI-optimized GenerateBlocks pages within 12 weeks

LovedByAI data survey, Q2 2026

GenerateBlocks already gave us the clean output we wanted. LovedByAI handled the part minimal markup alone cannot do: telling AI engines what the site actually is.

Samira Cole, Independent SEO Consultant

Common questions GenerateBlocks users ask AI

01

"Does [product] work with GenerateBlocks"

02

"Does GenerateBlocks have schema markup built in"

03

"GenerateBlocks vs Elementor for SEO and performance"

04

"Why is my GenerateBlocks site not showing up in ChatGPT answers"

Jenny Beasley

Jenny Beasley is Head of GEO at LovedByAI. With 7+ years as SEO Director at Salesforce and 3 years pioneering LLM optimization, she developed the GEO framework delivering a 200% median increase in AI citations within 60 days.

Frequently asked questions

Yes. LovedByAI works with GenerateBlocks and GenerateBlocks Pro on any theme, including GeneratePress. It reads your existing page structure and applies its behind-the-scenes AI discoverability layer independently of which blocks or theme built the page, including schema, semantic HTML and heading signals, metadata, and related HTML-level cues.

No. LovedByAI's discoverability work stays lightweight. It adds schema, metadata, semantic and heading signals, and other HTML-level cues without loading render-blocking CSS or JavaScript, so the minimal footprint you built GenerateBlocks around stays intact.

No. GenerateBlocks has no native Organization, Article, FAQPage, or Breadcrumb schema. It's a layout and block plugin focused on minimal HTML and CSS output, and schema generation has never been part of what it does. That's the specific gap LovedByAI fills.

No. LovedByAI reads whatever structure your page already has, tagged or not, and adds schema independently of it. Setting deliberate HTML tags on your Containers and Headlines still helps AI crawlers read your content faster, but it isn't required for LovedByAI's schema injection to work.

No, and LovedByAI is designed the same way: it runs alongside your existing SEO plugin rather than replacing it. Yoast or RankMath handle sitemaps and meta tags; LovedByAI adds the AI-specific structured data layer most SEO plugins only partially cover.

Get GenerateBlocks sites mentioned by ChatGPT and Gemini

LovedByAI works alongside GenerateBlocks with zero design changes. Install it, and let AI search engines see your site clearly.

zero manual work. zero visible changes.

Get mentioned by ChatGPT
GenerateBlocks GEO/AEO: AI-Search Readiness | LovedByAI