What AEO actually is, how it differs from SEO, which models get tracked, and what it takes to get started.
AEO is the discipline of getting your firm cited, named and recommended inside AI-generated answers - ChatGPT, Perplexity, Gemini, Claude and the rest - rather than just ranked in a list of blue links. It's built on structured content, consistent authority claims, and citation tracking, not keywords. For a fuller breakdown of what that work actually involves day to day, see what an AI search consultant actually does.
SEO optimises for ranking algorithms and click-through. AEO optimises for extraction: how a language model reads, trusts and quotes a page when constructing an answer. The two overlap, but the content structure, the metrics, and the definition of "winning" are different. For more on the distinction, see SEO and AEO aren't the same job, and on why doing more of the existing work doesn't close the gap, why "more SEO" won't fix an AI search problem.
Five categories cover most of what buyers ask an AI model: best-of and shortlist queries ("best marketing agency for X"), direct comparisons against named competitors, alternative or switching-intent queries, capability-specific positioning (measurement, attribution, a particular vertical), and niche or category-specific searches. I map these to your business before a single article gets written, then track visibility against each one separately.
I track citations and mentions across the eight major models - ChatGPT, Perplexity, Claude, Gemini, Copilot, Grok, Meta AI and DeepSeek - and report visibility and citation volume for each, monthly.
Three things, every month: whether you're visible on the target prompts, how much AI-referred traffic reaches your site, and how models actually describe you when they do cite you - two firms can both get cited and come away with very different framing. On the traffic point, treat analytics as a floor rather than a ceiling: plenty of buyers see you named in an AI answer and type your URL directly instead of clicking through, so the real commercial impact usually runs ahead of what's tracked.
No. The on-site sprint work is fully within my control and doesn't depend on external placements. Off-site authority building - press mentions, directories, Wikipedia - is a genuine amplifier, but it's a separate workstream I'll only recommend once the foundation is in place.
Yes - most clients already have a marketing function in place, and I'm not there to replace it. I run one specific, defined piece of work and hand off cleanly at the edges: your team keeps ownership of brand, campaigns and sales process, I own getting you cited in AI search. I've done this alongside in-house marketing and sales teams before, and I'll agree exactly where the handoffs sit before we start.
A named content contact for approvals - three business days per round, no response counts as approval - and either direct CMS access or a publish-by SLA with an agreed number of business days per article. Thirty minutes a month to review the data with me. Everything else I handle end to end. Existing sales collateral or brand guidelines are a big help if you have them, since they shortcut the pillar-defining work, but they're not required to get started.