Menu
Services/ AEO
Answer engine optimization

We do the work that gets brands named in AI answers

AEO services, meaning answer engine optimization, are the work of getting your brand into the answer when someone asks ChatGPT or Perplexity what to buy. We do it for Shopify brands doing $1M to $5M, inside the search work rather than as a separate retainer. If you have been putting this one off because every article about it sounds like a sales pitch, that is a reasonable read of the situation.

Monthly prompt set 4 assistants
ChatGPTrecorded
Perplexityrecorded
Google AI answersrecorded
Geminirecorded
Same prompts every month. We report what came back, not a score we invented.
On this page
01The mechanism
02The levers
03Pricing, in a way
04Reporting
05Questions
Read0 / 5
The mechanism

How an AI answer picks brands

An answer is not a ranking. There is no list of ten where you sit at four. The model returns a short answer naming two or three brands, and everything outside that is invisible.

Weighs most
What others wrote

Reviews, Reddit threads, comparison articles, press, and category roundups.

Confirms it
What you publish

Product data, structured markup, and pages that answer real questions.

What it names comes from what it has read about you, and most of that was not written by you, because a model treats a brand describing itself as the least reliable source in the room. Your site still matters in a narrower way. It confirms what other sources claim, and lets a model verify that you sell what people say you sell.

That asymmetry is the thing most AEO offers get backwards. Schema markup on your own site is the confirmation step. It is not the reason a model knows you exist.

The levers

What moves it for a Shopify brand

Four things, in the order they matter at your size.

01 Being in the roundupsSomeone writing "best category brands" is producing the source a model reads. Getting into those is PR work, not markup work.
02 Review volume and recencyA product with four hundred recent reviews reads differently to a model than one with twelve from 2023.
03 Product data a machine can parseStructured markup carrying price, availability, and review data, correct on every product rather than on the template.
04 Entity consistencyYour brand name, category, and details matching across every profile that mentions you, so a model resolves you to one thing instead of three.

Notice what is not on that list. There is no trick, no prompt to optimize against, and no submission process. A model cannot be gamed into naming you the way a search engine could once be gamed into ranking you, because there is no index to manipulate. There is only what has been written.

Pricing, in a way

Why this is not a separate retainer

Three of those four are things you would pay a search team to do anyway. PR earns links and citations in the same motion. Review programs run out of email. Structured data is technical SEO with a different name on it.

Buying it separately means two vendors doing overlapping work with neither of them owning the outcome. When your citations do not move, the AEO vendor points at your SEO and the SEO vendor points at your PR.

So it sits inside the search work here. There is no AEO line on the invoice, and the person doing your technical SEO is the person doing your structured data, because it is the same job.

Ecommerce SEO services
Reporting

How we measure it

You cannot rank-track an answer. There is no position two to report.

Instead we run a fixed set of prompts every month across ChatGPT, Perplexity, Google's AI answers, and Gemini. The same prompts each time, written as a buyer in your category would type them. We record whether you appear, where in the answer you land, and which source the model cited.

The useful part

Knowing a model pulled from a specific Reddit thread or a specific roundup tells you exactly where next month's work goes.

The tooling in this area is young and the numbers move for reasons nobody fully controls. We report what the prompts returned rather than a score we invented.

Before you ask

Common questions

01What AEO is
02AEO against SEO
03What GEO is
04What it costs
05Time to appear
06Worth doing yet
07Guarantees
08What we need
What are AEO services?

AEO services, or answer engine optimization, are the work of getting a brand named inside an AI-generated answer rather than ranked in a list of links. The work covers third-party mentions, review programs, structured data, and entity consistency across the web. For an ecommerce brand it overlaps heavily with SEO and PR.

Is AEO different from SEO?

The inputs overlap by most of their surface. The difference is what the system reads for. A search engine reads for keywords and links and returns a list. A model reads for entities and corroboration and returns a short answer. So structured data and third-party mentions carry more weight than they do in a ranking, and there is no position two to occupy.

What is generative engine optimization?

Generative engine optimization, or GEO, is the same work under a different name. Some of the market calls it AEO and some calls it GEO, and the distinction people draw between them is not consistent enough to be useful. If an agency treats them as two products, ask what the second one includes that the first does not.

How much do AEO services cost?

Nothing on its own. It is inside the search engagement, which starts at $2,500 a month after a $2,875 setup. If someone is quoting you a separate AEO retainer, it is worth asking which of the four things above they are doing that regular search work does not already cover.

How long before we show up in AI answers?

Structured data and entity work land soonest, often inside a couple of months, because they are technical changes a model picks up when it next reads the site. The part that takes longer is third-party coverage, since a roundup has to be written before it can be read. Brands with existing reviews and press move faster than brands starting from nothing.

Is this worth doing yet?

Google still sends most of your traffic and we build for that first. What has changed is the step before the search, where some buyers ask an assistant for options and start their comparison from that answer. Since the work that earns a citation is mostly the work that earns rankings, the near-term return arrives through search either way.

Can you guarantee we get cited?

No, and anyone who does is describing something they do not control. There is no submission, no index, and no ranking factor to satisfy. What we control is whether the inputs exist, and we report monthly on what the prompts return so you can see whether they are moving.

Do we need to do anything ourselves?

Two things help and neither is heavy. Reviews, meaning whatever post-purchase flow you have should be asking for them. And access to your product data, so the structured markup describes what you actually sell rather than what the theme guessed.

Two business days

Start with the audit

Part of the free audit is running your category through the assistants and recording what comes back. Two business days later you know whether you appear, who does, and what they have that you do not.

Get your cheat code