Skip to content

SEO

SEO In 2026: What Still Works, What Is Dead

Search behaviour is splitting between Google and AI engines. The SEO playbook that worked in 2022 is half-dead and the new half is what most operators have not built yet.

SEO in 2026 looks nothing like SEO in 2022. The old playbook (build backlinks, target high-volume keywords, write 1,500-word articles for the top three positions) still works for some pages, but it covers maybe 40% of the search behaviour your buyers actually do. The other 60% has moved to AI engines and to behaviours Google itself rewards differently now.

What still works. Topic-cluster site architecture, where one cornerstone page anchors a network of supporting pages all interlinked around a tight topic. Schema.org structured data on every page, especially Article, FAQPage, Organization, LocalBusiness, Service. Real first-party expertise (E-E-A-T signals: experience, expertise, authoritativeness, trust). Page speed and Core Web Vitals (Google rewards them and AI engines weight them). Internal linking to your highest-converting pages.

What is dead. Keyword stuffing. Link-buying schemes that Google now detects and penalises. Thin "blog post for the sake of having a blog post" content. Generic listicles regurgitated from the top three results. AI-generated content with no human-edited point of view (Google explicitly downranks this).

What is new. Getting cited by ChatGPT, Gemini, Perplexity, and Claude. These engines are now serving roughly 15% to 30% of informational queries that used to go to Google. The way they decide who to cite is different from PageRank. They favour clear, structured, opinionated, well-attributed sources. A page that says "Here is the framework, here is why it works, here is the proof, here is who it works for" gets cited more often than a page that says "There are many ways to think about X."

The Ignis SEO playbook in 2026 is two-tracked. Track one is classical SEO for the high-intent commercial keywords. Track two is AI search optimisation: schema graphs, structured opinion-led content, llms.txt, IndexNow, and a real point of view on every cornerstone page. The compounding effect is that both Google AND ChatGPT start citing the same pages, doubling the inbound discovery layer for the same content investment.