Types of SEO

Modern SEO is best understood as eight disciplines that compound, not a checklist of tactics or a single practice with optional add-ons. Pages that rank do well across multiple disciplines simultaneously. Pages that fail almost always fail at one specific discipline — and the failing one is rarely the one the team has been working on.

This article defines each discipline, shows where it owns leverage, and points out which combinations matter for which business models. What is SEO covers the practice end-to-end; this one is the taxonomy.

“Knowing the eight types is the easy part. Knowing which three or four to execute first — and how they reinforce each other — is where the revenue lives.”

The eight disciplines at a glance

1. Technical SEO — the foundation

The infrastructure layer. If technical is broken, nothing else matters — engines can’t crawl, can’t render, can’t understand the page. The 2026 non-negotiables: Core Web Vitals passing on mobile (LCP under 2.5s, INP under 200ms, CLS under 0.1), JSON-LD schema on every template, an XML sitemap segmented by content type, HTTPS with valid cert, JavaScript rendering that doesn’t hide content from crawlers, and a robots.txt that reflects actual indexation intent (including allow rules for GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, Google-Extended).

2. Content SEO — the strategy layer

Decides what to write, how to structure it, and when to refresh it — separate from the per-page on-page detail layer. Owns search intent matching, topic-cluster architecture, E-E-A-T signals (author + Person schema, sameAs identity, editorial standards), content audits, and refresh cycles for time-sensitive content. The discipline that compounds slowest but pays the longest tail.

3. On-Page SEO — the per-page layer

The optimisations applied to individual pages. Title tags, meta descriptions, H1-H6 hierarchy, image alt, internal linking, URL structure, schema markup. Most teams half-do this because they treat it as a static checklist instead of a design discipline. The 2026 test: open any article on your site, read the first 150 words, and check whether the page’s primary query is answered before the user scrolls.

4. Off-Page SEO — the authority layer

The signals engines use to decide whether your domain is trustworthy. Historically backlinks; in 2026 the definition expanded to include unlinked brand mentions in editorial coverage, podcast quotes, conference talks, and Wikipedia citations. Volume stopped mattering around 2018; quality and editorial weight are the actual signals. One mention in The New York Times or the Australian Financial Review weighs more than a thousand directory links.

5. Local SEO — different scoring model

Different ranking system from classic SEO. Google Business Profile, NAP consistency across citations, review velocity, local schema with proper LocalBusiness subtype (Restaurant, Dentist, LegalService, etc.), proximity-based scoring. A national-level technical foundation doesn’t carry over — you can rank position one nationally and still lose the map pack to a single-location competitor with a 4.9-star GBP. Treat as a separate discipline or you’ll under-invest.

6. AI SEO (GEO) — the new layer

Getting cited inside ChatGPT, Google AI Overview, Gemini, Perplexity, and Claude. The signals overlap heavily with classic SEO (schema, entity clarity, content depth, authority) but the measurement is different — share of voice across engines, not SERP position — and some signals weigh differently. Passage-level structure matters more for AI; backlinks weigh slightly less. Read how search engines work for the synthesis stage that drives AI citation.

7. International SEO — multi-region complexity

Hreflang implementation, ccTLD vs subfolder vs subdomain strategy, locale-specific content (not machine-translated), geo-targeting in Search Console, and the operational discipline of running multiple regional programmes without cannibalising. The discipline most generalist guides skip because the failure modes are subtle — wrong hreflang annotations don’t error, they just silently waste 30% of your international visibility.

8. Specialised — ecommerce, programmatic, automation, analytics

The cluster that catches the disciplines that don’t fit cleanly into the seven above. Ecommerce SEO covers product pages, faceted nav, Product schema, shopping feeds. Programmatic SEO covers pages generated at scale from data with thin-content safeguards. SEO automation covers the operational layer that ties the other disciplines together (briefs, audits, rank tracking, AI mention tracking). Analytics + KPIs covers GA4 + Search Console + the measurement frame.

How they actually compound

The mistake is treating these as a checklist where each gets ticked off independently. The compound effect comes from running them together. A few examples of how they reinforce:

• Technical + content. Schema markup (technical) makes content (content SEO) more legible to AI engines. Shipping schema without good content is wasted; shipping good content without schema leaves AI citations on the table.
• Off-page + AI. Brand mentions in editorial coverage (off-page) become entity-recognition signals (AI SEO) when AI engines retrain. The same investment serves both disciplines.
• Local + on-page. Local schema (local SEO) only triggers rich results if on-page elements (title, address, NAP) match. A LocalBusiness graph with inconsistent NAP across the page reads as broken to Google.
• International + technical.Hreflang (international) can’t fix a slow LCP (technical). The international layer compounds only on top of a healthy technical foundation.

What to prioritise by business model

The mistake universal across all four columns: skipping technical because “the site looks fine.” Every other discipline depends on a healthy technical foundation. Skip technical and the rest underperforms regardless of investment.