AI search is rewriting your traffic graph in real time.
Google AI Overviews now appear on 30%+ of commercial queries. ChatGPT and Perplexity are referrers. If your content isn't structured for citation, you're invisible in the next traffic medium. We rebuild technical SEO, content depth, and AI-citability in one engagement.
SEO (Search Engine Optimisation) is the practice of making content discoverable in classical search results — Google, Bing, Yahoo. GEO (Generative Engine Optimisation) is the parallel practice for AI-powered surfaces — ChatGPT, Perplexity, Google AI Overviews, Gemini, Bing Copilot. The two share technical foundations (crawlability, indexability, schema, Core Web Vitals including INP) and content principles (entity-rich writing, topic-cluster architecture, internal linking) but diverge on output: SEO competes for ranked positions, GEO competes for citations. Modern programs run them together, layering passage-level citability scoring, llms.txt policy, FAQPage / HowTo / Product schema, and AI-crawler access on top of classical SEO. Brands that ignore the AI layer in 2026 lose discoverability in the surfaces buyers are increasingly using to research before they ever hit Google.
What's broken in most SEO programs
5,000 thin posts targeting head terms with no topical depth. AI engines parse entities and relationships — keyword-stuffed content gets ignored. Rankings fall, and the team blames Google's helpful-content update.
Article schema only. No FAQPage, HowTo, Product, Organisation, sameAs. AI Overviews and Perplexity rely heavily on structured data — un-schemaed content cannot be cited cleanly.
INP replaced FID in 2024. Most sites haven't measured it. LCP > 4s on mobile. CLS > 0.25 on key pages. CrUX field data tells the truth that lab data hides.
Brands either block ChatGPT/Claude/Perplexity by default in robots.txt, or allow them with no policy. Both lose. We set the right access pattern + maintain llms.txt for AI-grounded discovery.
20+ posts chasing the same query. Internal links broken. The strongest page on the topic isn't the canonical one. Editorial calendar is bolting on, not pruning.
What we run
Technical SEO (crawl, index, security, mobile, CWV incl. INP)
Content SEO with topic clusters + hub-and-spoke architecture
Programmatic SEO (templated pages from data)
GEO / AI-search optimisation (ChatGPT, Perplexity, AI Overviews, Gemini, Bing Copilot)
llms.txt + AI-crawler policy + schema for AI discoverability
Local SEO (where relevant) — GBP, NAP, local schema
Ecommerce SEO — product schema, faceted nav, marketplace presence
SaaS SEO — integration pages, alternatives pages, comparison content, docs SEO
How we audit & operate
Forensic audit
Technical, content, GEO, local — full audit across all 9 categories. CrUX field data. Schema validation. AI-crawler access map. Cannibalisation analysis. Delivered as a 30+ page brief with a prioritised action board.
Topic architecture
Hub-and-spoke clusters built from SERP-overlap analysis (not text similarity). Pillar pages defined. Internal link graph rebuilt. Cannibalising pages consolidated.
AI-citability layer
Passage-level citability scoring on every page. Q&A formatting. Entity-rich content. Schema for FAQ, HowTo, Product, Organisation, sameAs. llms.txt published and maintained.
Content engine
Editorial calendar tied to topic clusters + freshness updates + decay detection. Briefs produced via skills + MCPs. Senior editorial review on every piece. AI-citation tracking via dedicated monitors.
What lands in your inbox.
- 30+ page SEO + GEO audit (technical + content + AI-search)
- Topic cluster map + hub-and-spoke architecture
- Editorial calendar with monthly publishing + freshness cadence
- Schema implementation + llms.txt + AI-citability scoring
- Monthly performance dashboard (rankings, traffic, AI citations)
- Quarterly strategy review against ranking + revenue goals
Where this worked.
Organic sessions +186% in 9 months
- Cannibalisation collapse merged 32 posts into 8 hub pages
- Schema rebuild captured AI Overview placement on 14 head terms
- Programmatic landing pages added 22K monthly sessions
ChatGPT + Perplexity citations: 0 → 47/mo
- llms.txt + entity-rich content rebuild
- Comparison pages ("vs / alternatives") now drive 38% of trial signups
- INP fixed from 480ms to 180ms — bounce rate −22%
Questions we actually get asked.
What's GEO and why should I care now?
Generative Engine Optimisation — making your content citable by AI engines (ChatGPT, Claude, Perplexity, Gemini, Bing Copilot, Google AI Overviews). It's a leading indicator: where the citations go, the traffic follows. Brands optimising now will compound an unfair advantage over the next 18 months.
How long until we see SEO results?
Technical fixes lift in 30–45 days. Content lifts in 90–180 days. Topic cluster strategy compounds over 6–12 months. AI citation lift is the fastest — we typically see ChatGPT and Perplexity start citing fixed pages within 30 days of publishing.
Do you write the content yourself?
We run the editorial calendar, write briefs, and produce content with a senior editorial layer + AI-assisted production. For founder-led pieces, we ghostwrite from your interviews. For category content, we produce end-to-end.
What about backlinks?
We don't run link-buying schemes. We earn links through original research, comparison pages, and integration partnerships. If your category requires aggressive link building, we'll tell you up front and refer to specialists we trust.