{{header-meta}}
AI Search Consulting
Most of what gets your brand named in ChatGPT and AI Overviews is SEO done well. The rest is knowing what’s actually different about AI search, and measuring whether it’s working.
The work
AI search visibility is included in every engagement I run. Not a separate product, not an add-on, and not a second line on the invoice. Most of what earns visibility in ChatGPT or an AI Overview is SEO you’d recognize, and splitting it out would mean charging you twice for one body of work. If someone is selling you AI search as its own service, it’s worth asking what it includes that good SEO doesn’t.
What’s new is the measurement, and the work that follows from what it shows. Here’s what that covers:
- Prompt research grounded in how real buyers ask about your category, producing a set of roughly 25 prompts rather than keyword phrases converted into questions, revisited as your priorities and the category shift.
- Monthly tracking of how often your brand is named in an answer, and how often your own pages are the source, across the major AI assistants and AI-generated answers in search. Your competitors are tracked against the same prompt set.
- An action plan built from where each prompt currently stands, since being named without being cited calls for different work than not appearing at all.
- A tiered strategy for the prompts that matter most to the business, covering what it would take to become part of the answer consistently.
- Production handoffs written to hand straight to whoever’s implementing: on-page changes as they come up, content direction that feeds the monthly briefs, and off-site targets worth pursuing.
What’s different about AI search
AI systems ground their answers in search results, which means organic ranking is the floor. That’s Google’s own guidance, Bing has published the same, and what I’m seeing across client accounts lines up with it.
A handful of things behave differently enough to need their own attention:
The questions don’t look like search queries. Longer questions, comparisons between named options, and requests filtered through a list of conditions like budget, company size, or integrations. That’s why a prompt set has to be built from how buyers actually talk rather than from keyword phrases with question marks bolted on.
AI systems run fan-out queries you never see. When someone asks a question, the system breaks it into related subqueries, gathers sources for each, then assembles an answer from what it finds. Ranking well for the obvious term isn’t enough. The pages that get cited tend to rank across the cluster of questions sitting underneath it.
Being named and being cited are separate problems. Your brand can appear in an answer with no link back to your site, which means someone else’s content is describing you. Your page can also be the source without your brand being mentioned at all, which means you did the work and someone else gets the credit. Those call for different fixes, and you can’t tell them apart without measuring both.
Crawler access isn’t the same question as Googlebot access. AI assistants use their own user agents, and the rules that govern them live in robots directives, CDN configuration, and bot management settings that often get set by technical teams without marketing in the room. Many AI crawlers also don’t render JavaScript, so content that’s visible in Google can be invisible in an AI answer.
What other sites say about you carries more weight than it used to. AI answers tend to assemble from multiple sources agreeing with each other. Brand mentions on relevant, credible sites now matter in a way that goes beyond the link itself.
Why the SEO foundation comes first
The early months of an engagement build in order: keyword research, then on-page recommendations, then the technical audit, then content that fills the keyword and topic gaps the research surfaced. That’s the SEO consulting work this sits on top of, and the sequence isn’t arbitrary.
Measuring citation rates for pages that can’t rank yet tells you very little. If a page isn’t competitive in organic search, it’s unlikely to be the source an AI system reaches for, and the citation data will confirm that month after month without telling you anything you can act on. The foundation is what makes the AI measurement worth reading.
This is also where I’d push back on a lot of what’s being sold right now. There’s a wave of AI-specific tactics with thin evidence behind them, and some of them run against practices that have held up for years. Others are the AI-era version of things that got sites into trouble in earlier eras. I’d rather recommend something that still works in two years than something that spikes this quarter. When the evidence changes, so will the recommendations.
What this looks like when it works
Three engagements where the AI visibility came out of ordinary SEO work rather than a separate AI search project, measured against real competitive sets:
The Grossman Group, a leadership and internal communications firm, is one of the two most cited sources across the questions buyers ask in their category, alongside YouTube and ahead of LinkedIn, Forbes, Reddit, Harvard Business School, and every competitor.
Vibes, a B2B SaaS mobile marketing platform, is the single most cited domain across every question we track in their category, ahead of YouTube, Reddit, and every competitor benchmarked. In overall AI search visibility they rank third, behind two publicly traded companies many times their size.
Culturelle, a consumer health brand, earned AI Overview citations across dozens of primary keywords during a two-year engagement, and a new traffic channel from AI assistants grew from nothing to roughly 1,800 visits a month over the course of the engagement, about 6% of their organic traffic. Over the same period, most sites in their category lost 25 to 30% of organic traffic. Culturelle held its ground.
Who I work with
Small to mid-sized businesses, marketing teams, and agencies with some internal resources, whether writers, developers, designers, or outside partners, to help carry the work forward.
The companies best suited to this work usually have one thing in common: they’ve noticed something changing in how their category shows up in AI search. Competitors named in ChatGPT answers where they aren’t. AI Overviews sitting above results they used to own outright. A traffic line that’s drifting without an obvious cause.
A few of the most common scenarios:
In-house marketing teams watching the ground shift have the people and budget to execute, and an SEO program that’s been running for a while, but nobody measuring whether any of it is landing in AI answers. I take over the SEO thinking and bring the measurement with it.
Founders and operators who own the marketing function can see that their buyers are researching inside AI assistants now, and that their brand isn’t part of those conversations. They need someone thinking about search and AI visibility together, not one or the other.
Agencies and marketing partners who want AI visibility handled as part of their client SEO work. Most often I’m introduced openly as the SEO partner and work alongside the agency’s client-facing team. Either way, I do the SEO, the agency owns the client relationship.
How engagements run
AI search visibility is included in every retainer at no additional cost. There’s no separate scope and no upsell.
Retainers are where this work lives, because AI visibility is a trend line rather than a snapshot. The prompt set gets tracked monthly, the picture shifts as competitors publish and models update, and the recommendations follow the movement.
If your site hasn’t had consistent SEO attention for a year or two, AI search work is premature. Not because the tactics are difficult, but because the measurement will keep telling you the same thing: your pages aren’t competitive enough to be chosen as sources. Fix that first and the visibility follows. I’d rather tell you that on a first call than bill you for six months of tracking a number that can’t move yet.
Everything runs through me. Every call, every report, every recommendation. Twenty years in SEO, twelve as an independent consultant, and seven or eight clients at a time by design.
Questions I get asked
Is this the same thing as AEO, GEO, or AI SEO?
Mostly, yes. Answer Engine Optimization, Generative Engine Optimization, AI SEO, LLMO, and AI search optimization all describe the same objective: getting your brand named and your pages cited when someone asks an AI assistant a question in your category. The labels are still settling and different people use them differently. I use “AI search” because it describes what’s happening without asking you to learn an acronym. If you came here looking for an AI SEO consultant, an AEO consultant, or an AI search optimization consultant, you’re in the right place.
Can I hire you for AI search work on its own?
No, and I’d rather explain why than take the engagement. The measurement is straightforward to set up. Acting on it is where the work is, and almost everything the data points to is SEO: pages that need to rank for the questions underneath the question, content gaps, technical issues blocking crawlers, structure that makes a passage quotable. Selling that as a separate service would mean charging you twice for one body of work. What I do sell is SEO consulting, with this included.
We already work with an SEO agency. Can you handle the AI search side?
Not as a split. AI visibility rests on the same technical foundation, content, and rankings the SEO work produces, so measuring it while someone else owns those levers means finding problems I can’t fix. If what you’re seeing in AI search is the reason you’re rethinking the arrangement, that’s a conversation worth having, and it’s a conversation about all of it rather than a slice.
Does this cost extra on top of an SEO retainer?
No. It’s part of the scope, and it’s in the agreement in writing.
How do you actually measure AI visibility?
A fixed prompt set, tracked monthly, benchmarked against a defined competitive set. Two things get measured: how often your brand is named in an answer, and how often your own pages are the cited source. Those move independently, and the gap between them is usually where the opportunity is. Your competitors run against the same prompt set, which is what makes the number mean something.
What actually changes in the work compared to regular SEO?
Less than the industry would have you believe, and more than nothing. The main shifts are targeting the cluster of related questions around a term rather than the term alone, structuring key answers so they hold up when quoted out of context, checking crawler access for AI user agents separately from Googlebot, and treating brand mentions on credible sites as worth pursuing in their own right.
What about llms.txt and the newer AI-specific tactics?
There’s no evidence yet that llms.txt does anything, and I’m not going to recommend work on that basis. Some of the more advanced tactics being discussed are legitimate and worth watching. Others are already producing the familiar pattern of a visibility spike followed by a collapse. I’ll tell you plainly what’s supported by evidence and what isn’t.
How quickly can this move?
Faster than traditional rankings, in a lot of cases. Google’s AI answers draw on a much shorter window of data than rankings do, and AI systems reach deeper into results than a top-three position requires. A page you improve this month can show up in an AI answer well before it climbs the rankings for the same term.
The slower pieces are the ones that were always slow: building real coverage across the questions surrounding a topic, and earning mentions on sites other people trust. Those run on the usual timeline.
Client work
“
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.
Name