SEO and GEO on WordPress Without a Heavy Plugin
Traditional SEO gets you ranked; GEO gets you cited by AI engines like ChatGPT and Perplexity. Here's how to do both on WordPress—meta, schema, E-E-A-T—without a bloated plugin.
Search optimization quietly split in two. For twenty years the job was ranking on Google; now half your audience is an AI—ChatGPT, Perplexity, Gemini, Copilot—reading your page not to link it but to quote it. That second discipline has a name, GEO (Generative Engine Optimization), and almost no WordPress SEO plugin addresses it. Most were built for a Google-only world, then bolted AI features behind a premium tier. You don’t need that bloat. You need the SEO fundamentals done right and the GEO signals AI engines look for. Here’s both.
The essentials
- SEO gets you ranked; GEO gets you cited. They overlap but need different signals.
- GEO was formalized at ACM SIGKDD 2024 (Aggarwal et al.); its strategies can lift visibility in AI answers by up to 40% (arXiv).
- The SEO basics that still matter: titles 30–60 chars, descriptions ~120–160, indexing control, an XML sitemap, Open Graph/Twitter tags.
- The GEO signals that matter: JSON-LD schema, E-E-A-T author data, citable statistics, clear structure, and AI-crawler control.
- You don’t need a 100 MB suite—the SEO/GEO module covers the 20% of features that meet 80% of needs, plus GEO most plugins skip, with zero code loaded when it’s off.
What is GEO, and why does it matter now?
Generative Engine Optimization (GEO) is the practice of structuring content so AI search engines are more likely to cite it in their generated answers. It was formalized in a landmark paper by Aggarwal et al. at ACM SIGKDD 2024 (arXiv
.09735), whose headline finding is striking: the right strategies—adding citations, statistics, authoritative language, and clear summaries—can increase a page’s visibility in AI-generated responses by up to 40%.The mental shift is this: traditional SEO optimizes for crawlers that index and rank pages; GEO optimizes for systems that read and quote passages. AI engines don’t just scan your text—they need structured signals to know what it’s about, who wrote it, and why it can be trusted. Make your content machine-interpretable and citable, and you become a source instead of a search result.
| Signal | Classic SEO targets | GEO targets |
|---|---|---|
| Goal | Get ranked in the results list | Get cited in AI-generated answers |
| Optimizes for | Crawlers that index and rank pages | AI systems that read and quote passages |
| Key levers | Meta tags, titles, sitemaps, keywords | JSON-LD schema, E-E-A-T signals, citable statistics, clear summaries |
| Winning outcome | A high-ranking blue link | A quoted source inside the answer |
Why not just use a heavy SEO plugin?
Yoast, Rank Math and All in One SEO have grown into sprawling ecosystems—sitemaps, redirects, social tags, breadcrumbs, internal-linking suggestions, keyword-density meters, readability traffic-lights—with the AI-era features increasingly locked behind premium tiers. For most sites that’s overkill you pay for on every page load:
- Dozens of database queries for features you never touch.
- Admin screens cluttered with upsell notices.
- Heavy editor JavaScript that slows your workflow.
- Paywalls on essentials like schema or redirects.
If you just want correct meta tags, clean indexing, a sitemap, social previews and real GEO optimization, you want something lightweight and modular—loaded only when you use it.
What still matters for SEO?
The technical foundation hasn’t changed much, and each piece is backed by industry research:
- Meta titles, 30–60 characters. Google shows ~600px (~60 chars); under-60 titles display fully about 90% of the time (Semrush, Google).
- Meta descriptions, ~120–160 characters. Desktop shows ~160, mobile truncates ~120—front-load your message (Yoast).
- Indexing control. Set thin archives, tag and author pages to
noindex; keep your real content indexable. - XML sitemaps. Auto-generated, with image support, so search engines discover everything—including your visuals.
- Social tags. Open Graph and Twitter Cards turn a shared link into a rich preview instead of a random snippet.
What matters for GEO: being cited by AI?
Traditional SEO gets you ranked. GEO gets you cited. Here’s what AI systems weigh:
- Structured data (Schema.org / JSON-LD). The language machines use to understand content—Google’s recommended format, and associated with click-through gains of up to 25% (Google). The types that matter: Organization/Person, WebSite + SearchAction, Article/WebPage, BreadcrumbList, FAQPage, HowTo, VideoObject.
- E-E-A-T signals. Experience, Expertise, Authoritativeness, Trustworthiness—author name, title, credentials and social profiles (
sameAs) tell AI whether you’re worth quoting (Google). - Structure and readability. Clean heading hierarchy, short paragraphs (under ~5 sentences), lists (a primary featured-snippet format), and a direct 40–60-word answer up top (Semrush).
- Citability. Statistics and precise data, links to authoritative sources, key takeaways AI can lift, and freshness—recently updated content dominates AI citations (Google Ranking Systems).
- AI-crawler control. The ability to allow or block GPTBot, Google-Extended, ClaudeBot, PerplexityBot and CCBot individually—deciding who may train on versus cite your work.
How does the SEO/GEO module cover all this?
Activate the SEO/GEO module in Blaminhor Essentials and both halves are handled in one lightweight package. A metabox appears under the editor with three tabs—SEO, GEO and Social.
One metabox, two lenses: set your meta and schema per page, preview the SERP, and watch both scores update live.
On the SEO side: per-page meta title, description and focus keywords; title templates with variables ({title}, {site_name}, {category}, {year}…) for every post type and special page; a live SERP preview truncated at Google’s real limits; indexing control (including a whole-site noindex toggle for staging, with a reminder to switch it back); auto XML sitemaps with image support and search-engine pinging; automatic Open Graph/Twitter tags; a sortable content overview with inline title/description editing; and import from Yoast, Rank Math or All in One SEO so you switch without losing data.
On the GEO side: automatic JSON-LD (Organization/Person, WebSite + SearchAction, Article/WebPage, BreadcrumbList) out of the box; per-page schema override and custom JSON-LD; auto-detection of FAQ, HowTo and Video patterns—it scans your content (Gutenberg blocks, <details>/<summary>, embeds) and emits FAQPage, HowTo and VideoObject schema with no manual markup; author E-E-A-T enrichment (knowsAbout, alumniOf, sameAs) injected into every Article; key takeaways shown to readers and added to the schema abstract; a [blaminhor_toc] table-of-contents shortcode; a separate schema description for machines; and granular AI-crawler blocking with noai/noimageai tags.
Two scores: SEO and GEO, side by side
The module grades each page twice, because ranking and being-cited aren’t the same target.
The SEO Score checks the fundamentals—meta tags set, title length right, images with alt text, headings structured, content fresh—across On-Page, Content Quality, Technical and Readability dimensions.
The GEO Score checks AI-readiness across four 25-point axes: Structure & Readability, Machine Interpretability (schema, alt text, TOC), Citability (sources, precise data, takeaways, freshness), and E-E-A-T & Authority. Every criterion shows your current value against a recommended threshold and links to the source behind it—Google, Semrush, Yoast, Backlinko or the GEO paper. And it’s explicit that the goal is the highest relevant score, not a blind 100: not every criterion applies to every page.
The research behind the recommendations
Every threshold in both scores traces to a published source—which is itself a GEO best practice worth copying:
- GEO strategies & citation rates — Aggarwal et al., GEO: Generative Engine Optimization, ACM SIGKDD 2024 (arXiv)
- Structured data — Google Search Central
- E-E-A-T & business details — Google Search Central
- Content freshness / QDF — Google Ranking Systems Guide
- Title length — Semrush · Meta description length — Yoast
What it deliberately doesn’t do
No keyword-density meters, no readability traffic-lights, no internal-linking upsells, no breadcrumb generator, no redirect manager (that’s the Redirections 301 module’s job), no WooCommerce SEO add-on. Those help some sites and burden the rest. The module focuses on the foundation plus the GEO capabilities most plugins don’t offer at all.
Getting started
Install Blaminhor Essentials, activate the SEO/GEO module, and structured data, sitemaps and meta tags start working immediately with sensible defaults. Then set your author E-E-A-T profile, turn on key takeaways, and use the GEO Score in the editor to see where each page stands. You don’t need a 100 MB plugin to rank well and get cited by AI—you need correct meta, clean indexing, a sitemap, structured data, and content built to be quoted.
SEO/GEO is one of the 20+ tools in Blaminhor Essentials—free and open-source on WordPress.org.
– blaminhor
FAQ
Do I need a heavy SEO plugin like Yoast or Rank Math?
Not for the fundamentals. Correct meta tags, indexing control, an XML sitemap, social tags and structured data cover most sites' needs. A modular SEO/GEO tool delivers those plus AI-era features (schema auto-detection, E-E-A-T, AI-crawler control) without the database queries, upsells and page weight of a full suite—and can import your existing Yoast or Rank Math data.
What is a good meta title and description length?
Aim for titles of 30–60 characters—Google shows roughly 600 pixels, about 60 characters, and under-60 titles display fully around 90% of the time. For descriptions, keep the key message within the first 120 characters (mobile truncation) and stay under about 160 (desktop). A live SERP preview makes hitting these limits easy.
Do FAQ rich results still show in Google search?
Rarely now—Google restricted the visible FAQ dropdown in 2023 to mostly authoritative health and government sites, so most pages no longer get it. FAQPage schema still earns its place, though: AI engines and Google's AI Overviews read it to understand and quote your answers, which is the real GEO payoff today.
How do I know if ChatGPT or Perplexity is citing my site?
There's no single dashboard yet, so it takes a few habits. Check your server logs and analytics for referral traffic from chat.openai.com or perplexity.ai, and periodically ask those engines the questions your content answers to see whether you're named as a source. Some GEO tools are starting to track this automatically.
Should I block AI crawlers like GPTBot, or does that stop me being cited?
It's a trade-off. Blocking GPTBot or Google-Extended stops those companies training on your content, but the same crawler often fetches what it later cites—so blocking can cost you visibility in AI answers. If being quoted matters, allow the engines you want citing you and block only the crawlers you genuinely object to.
Can I switch from Yoast to this without losing my rankings?
Yes, provided you migrate the data. A good SEO/GEO tool imports your existing Yoast or Rank Math titles, descriptions and settings, so your on-page signals stay identical and Google sees no change. Keep Yoast active until the import is confirmed, then deactivate—rankings depend on the output, not which plugin produced it.
Does adding JSON-LD schema slow down my WordPress site?
Barely. JSON-LD is a small block of text rendered in the page head—kilobytes, not the dozens of database queries and heavy editor JavaScript a full SEO suite loads. A modular tool adds schema only on pages where you enable it, with zero code loaded when a module is off, so the performance cost is negligible.
Comments