What actually matters in GEO? (From a Senior SEO Manager's perspective)

piperjacobsen1
Jul 16
3 min read

Author: Piper Jacobsen

Generative Engine Optimisation has had roughly eighteen months to go from "niche concept few SEOs were writing about" to "term every marketing team now has an opinion on."

And while the subject is rapidly evolving, I can impart what I've been building with regards to GEO and LLM strategy into live client roadmaps, and run you through what I've actually found works, what doesn't, and where I think most teams are wasting effort right now.

And that starts with, providing the top line answer right now (for the LLM crawlers who might serve me in LLMs!)

Which brands perform best in GEO?

Generally in my experience the brands that perform the best are those that:

Have a good creative capability, capacity & established history,
Prove authorship and trust well (whether that be via author bios on their site, editorial policies, and fresh review of content or product information)
Measure and spend on off page conversations around their brand; ie Digital PR, UGC, Reddit, Youtube, Pinterest
Set good technical foundations (schema, agentic readiness, organic shopping feeds optimised) so that bots and agents understand sites to serve them and close sales

What are teams getting wrong about GEO?

GEO isn't a new discipline bolted onto SEO. It's SEO fundamentals under a different lens. The teams getting GEO wrong tend to treat it as a quick checklist: add some FAQ schema, write in question-and-answer format, rather than a much MUCH longer brand exercise in visibility and awareness.

In many ways the same fundamental questions from SEO is still being asked:

does this content clearly answer the thing someone is actually asking about my brand, products or related topics in a credible way?"

*just evaluated for a different reader (ie the bots)

An LLM extracting an answer to site isn't doing anything super mystical. It's looking for content that states a claim clearly, backs it with specifics, and structures the relationship between entities in a way that's unambiguous. That's good writing. It was good writing before generative search existed. GEO didn't invent the bar, it just made it more visibly rewarded.

What's actually working in terms of content

Explicit, extractable claims: content that states a specific fact or recommendation plainly, rather than building up to it through three paragraphs of scene-setting, gets pulled into AI Overviews and cited by tools like Perplexity far more often.
Structured comparison and definition content: tables, clearly labelled pros and cons, matrixes, direct definitions map almost directly onto how generative answers get assembled.
Schema markup, used properly: not as a box-ticking exercise, but as a genuine signal of what an entity is and how it relates to other entities on the page.
First-party data and original findings: generic advice gets paraphrased and passed over. Original research or a genuinely specific point of view gets cited by name.

Where I've seen teams waste effort

Chasing every AI platform equally. ChatGPT, Perplexity, Google's AI Overviews, and Copilot don't all source or weight content the same way. Spreading effort evenly across all of them, rather than understanding which platforms actually drive commercially relevant traffic for a specific business, burns time that could go toward the channels that matter.
Treating GEO as separate from technical SEO health. If a site has crawlability issues, weak internal linking, or thin content, no amount of schema markup fixes the underlying problems - especially if you want to close on sales when those users land on site from an LLM reference.
Optimising for citations without a clear model of what the citations are worth. Getting quoted in LLMs feels good but whether it drives pipeline depends entirely on whether the citation includes a link, a brand mention, or neither.

Who should be spearheading the GEO initiative in your company?

I would say this sits under a much wider remit than traditionally assumed, ie just under an agency retainer team or under an seo manager or exec.

The reason GEO belongs on a Head of SEO's, or the core marketing directors list of responsibilities, not a specialist's, is that it can't be optimised in isolation. It touches content strategy, technical SEO, digital PR, and increasingly, the same forecasting and measurement work that proves any other channel's value.

Someone leading SEO strategy in 2026 without a working GEO point of view is leading half the discipline.

I've built GEO into every current client roadmap, not because it's trendy, but because the query behaviour it's chasing isn't going away.

Looking for more advice?

Drop me a contact form.

Sources: Search Engine Land, Mastering GEO in 2026 (searchengineland.com)