CAJOLE Start a conversation
Answer engine optimization · AI search visibility

Be the answer inChatGPT

ChatGPT Google AI Overviews Perplexity Copilot Gemini

AEO — being the answer AI engines give when your customers ask. We measure how often, and how accurately, your brand appears when AI assistants answer your customers' questions. We do the work that improves it. We report every number with its method disclosed.

Fixed scope. Fixed fee. Maximum results.

Changing times
Before
Ten links.
The buyer chooses.
Now
One answer.
Something else chose.

Most of what you already do still matters. What changed is the unit of optimization: from ranked position to inclusion and framing inside a synthesized answer, measured per engine.
A page can rank first and still be missing from the answer.

94% of business buyers used AI at some point during their most recent purchase.
Forrester survey of ~18,000 buyers, published January 2026. Measures any AI use during a purchase, not AI search specifically.
One major analyst forecast a 25% decline in traditional search volume by 2026.
Gartner, February 2024 — a forecast, not observed fact; single source.
Evidence

Ranking first is no longer the same as being cited.

38%from ~76%
The share of sources cited in Google's AI Overviews that come from top-ten-ranking pages fell to 38%, from ~76% a year earlier.
Ahrefs study of 863,000 search results pages; vendor-published; 2026.
~40%visibility lift
In peer-reviewed experiments, content built to be quoted — direct answers, real statistics, cited sources — lifted visibility by up to ~40%.
Aggarwal et al., ACM SIGKDD 2024.
68%/54%/~10%
68% of organizations are changing strategy; 54% have handed AI search to their SEO team; only ~10% of marketers feel fully prepared.
BrightEdge survey, 750+ marketers; 2025.
What we do

We measure it, do the work, and report it with the method attached.

01 — Measure
Share of answer, per engine

How often you appear in answers to an agreed question set, against named competitors — plus citation rate. Same question set, same panel, every run.

02 — Do the work
Content, entity and citation work

Answer-first structure, real facts and figures, cited sources, a clean entity footprint. Fixed scope — never hourly.

03 — Report
Every number with its method

Re-measured against the same baseline, under the reporting standard below, with a change log of engine and model events in the window.

The tools in this market largely give you software to operate — they measure, you fix. The market is young and split between rebrandable software on one side and invisible fulfilment on the other. We combine the two: we measure and do the work, at fixed scope, with every metric's methodology disclosed — directly for companies, and wholesale under agencies' brands.

Programs · Assessments · Solutions
Baseline assessment
2–3 WKS

"A fixed-scope read on how your brand actually appears in AI answers today — which questions surface you, which surface your competitors, where the gaps are, and the 90-day plan to close them."

Content & entity sprint
4–6 WKS

"We rebuild a defined set of your pages so AI answers can actually quote them — answer-first structure, real facts and figures, cited sources, and a clean entity footprint — then re-measure against the same baseline."

Measurement foundation
2–3 WKS

"We switch on everything that can be measured about AI-driven visits, and write down plainly what still cannot be seen."

Managed program
MONTHLY

"An ongoing program that watches how AI answers treat you, ships the work that changes it, and reports monthly with the method attached."

The reporting standard
Every metric we report discloses: the questions asked (prompt universe), the denominator, the engines, geography and language, collection method, run count, cadence, date window, competitor set, and confidence. Trend reports carry a change log of engine and model events in the window — so you can tell our work apart from platform noise.
Prompt universe Denominator Engines Geography & language Collection method Run count Cadence Date window Competitor set Confidence
What we commit to
The work, to the agreed scope, plus measurement reported under this standard.
What no one can honestly promise
Rankings, citations, traffic, or revenue. If someone promises you those, ask them for their method.
The honest boundary

What can be measured — and what genuinely cannot.

Measurable today
Google Search Console has a dedicated generative-AI performance report, launched 3 June 2026 — covering Google Search and Discover surfaces only. No equivalent exists for Bing, ChatGPT, or Perplexity.
Official Google launch.
Crawler and bot activity, per platform, based on each platform's own published policy — not a blanket claim about AI crawler behavior.
Not measurable today
An estimated 60–70% of AI-influenced visits are misclassified in standard analytics.
An industry estimate, not a measurement.
No tool fully closes that gap. Anyone who says otherwise is selling you something.
We also tell you what not to buy
llms.txt has reached ~10% adoption across roughly 300,000 domains — with no measurable correlation to being cited. We don't sell it.
SE Ranking study. Describes current behavior, which could change.
The strongest study on technical schema markup found AI citations barely moved. We keep it tidy; we do not sell it as the lever.
Ahrefs difference-in-differences study, 1,885 pages, May 2026; vendor-published with method described. Evidence against markup as a primary driver.
A widely-repeated practitioner claim puts this work at "80% good, fundamental SEO" — we treat that as a practitioner claim, not a finding, but the direction is right.
Practitioner claim (uSERP), widely repeated; not an empirical finding.
Partner Programs

Partnering with you, empowering your brand.

AI search visibility, delivered under your brand. Your client, your contract, your retail. Non-circumvention in writing.

We measure
Share of answer against named competitors, per engine.
We execute
Content, entity and citation work at fixed scope — never hourly.
You own the client
Wholesale to you; your retail, brand and contract.
The next step · roughly 90 days
One client. A baseline, then one sprint.
Weeks 1–3 Baseline Weeks 4–9 One sprint Weeks 10–11 Re-measure, same panel Week 12 Agency-facing debrief

Your contract with your client, your brand throughout, pilot terms agreed up front. If it works, you have a case study of your own. If it doesn't, you have a disclosed measurement of why — which is worth having either way.

What the market is telling agencies
70% of agencies have raised or plan to raise prices partly because of AI search.
BrightEdge vendor survey; low-to-moderate reliability.
55% of marketers have budget earmarked for this work.
Scribewell/VML survey via Digiday, cited July 2026 — self-reported intent.
54% of US marketers plan to implement within 3–6 months, while only 23% currently invest in measurement.
eMarketer, January 2026, via a vendor intermediary.
AI-referred retail traffic converted 54% better than non-AI traffic in May 2026, with revenue per visit +53% — from 38% worse a year earlier.
Adobe Analytics (Q3 2026 report); vendor-published, not independently audited; US retail only.
Power your agency with white-label AI search visibility
Learn · the words people use

Many names for one shift.
Here's what it all means.

Answer Engine OptimizationAEO

Being the answer AI engines give when your customers ask. The discipline of getting included, and framed correctly, inside a synthesized answer — not just ranked on a page of links.

Generative Engine OptimizationGEO

Optimizing how generative AI systems find, quote, and frame your content. Used interchangeably with AEO by most of the market; we use both, because buyers search both.

AI SEOAI search optimization

Acceptable synonyms. Much of this work is good, fundamental SEO carried forward — what is new is the unit of optimization, measured per engine.

AI search visibilityPlain language

The phrase we use when we explain what we do: how often, and how accurately, your brand appears when AI assistants answer your customers' questions.
A word about "AIO"

You will see "AIO" used for at least three different things — AI Overviews, AI optimization, and artificial-intelligence operations. It is not a category label, and we don't use it as one. If someone sells you "AIO," ask them which of the three they mean.

The two metrics we sell in
Share of answer
How often you appear in answers to an agreed question set, against named competitors.
Citation rate
How often your pages are the source the answer credits.
Insights & Research
We are building a publication for this work: what we measure, what we find, and what we get wrong. Nothing published yet, and we won't pretend otherwise.
Shared with clients
Questions, answered plainly

What is answer engine optimization (AEO)?

AEO is the work of being the answer AI engines give when your customers ask. It covers measuring how often and how accurately a brand appears in AI-generated answers, then changing the content, entity and citation signals those answers draw on.

How is AEO different from SEO?

The unit of optimization changes: from ranked position to inclusion and framing inside a synthesized answer, measured per engine. A page can rank first and still be missing from the answer. Most fundamentals carry over.

What is generative engine optimization (GEO)?

GEO is optimizing how generative AI systems find, quote, and frame your content. In practice the market uses GEO and AEO interchangeably; so do we, because buyers search both.

What does "AIO" mean?

It has at least three live meanings: Google's AI Overviews, AI optimization, and AI operations. Because it is ambiguous, we don't use it as a category label. Ask any vendor using it which one they mean.

What is share of answer?

Share of answer is how often your brand appears in answers to an agreed question set, compared with a named competitor set, measured per engine. Citation rate is how often your pages are the credited source.

Can you guarantee we will be cited in ChatGPT?

No. Nobody can honestly promise rankings, citations, traffic, or revenue. We commit to the work, to the agreed scope, plus measurement reported under our disclosed standard. If someone promises you outcomes, ask them for their method.

Which AI engines can actually be measured today?

Google publishes a dedicated generative-AI performance report in Search Console, launched 3 June 2026, covering Google Search and Discover only. No equivalent exists for Bing, ChatGPT, or Perplexity, so those are measured by our own disclosed panel. An estimated 60–70% of AI-influenced visits are misclassified in standard analytics — an industry estimate, not a measurement.

Does llms.txt or schema markup get us cited?

On current evidence, no. llms.txt has reached ~10% adoption across roughly 300,000 domains with no measurable correlation to being cited (SE Ranking) — we don't sell it. The strongest study on schema markup found AI citations barely moved (Ahrefs, 1,885 pages, May 2026); we keep markup tidy and do not sell it as the lever.

Do you work with agencies?

Yes — wholesale and white-label. You keep the client, the contract, and your retail pricing; the deliverables carry your brand; non-circumvention is in writing. Direct company engagements and agency partnerships run side by side.

Do you have case studies?

Not yet. We are a new venture, we have no case studies, and we are not going to invent any. What we can show you is the method we would use and exactly what it will and will not tell you. [Client examples — coming with our first pilots]

How is this priced?

Everything is fixed-scope or recurring — never hourly. We scope in a short conversation and quote against that scope; we don't publish rate cards we can't stand behind.

What is CAJOLE?

CAJOLE is an answer engine optimization (AEO) and AI search visibility company. We measure how often — and how accurately — a brand appears when AI assistants like ChatGPT, Google AI Overviews, Perplexity, Copilot, and Gemini answer customers' questions, we do the work that improves it, and we report every number with its method disclosed.

Who does the work
Online-marketing practitioners
The fulfilment itself — content, entity and citation work.
Business consulting and strategy operators
Packaging, scope discipline, partner success.
AI engineers and developers
Measurement infrastructure and delivery automation.
Start here

See how AI answers talk about you.

A baseline assessment is a fixed-scope read on how your brand actually appears in AI answers today — and the 90-day plan to close the gaps. Two to three weeks. No pricing games, no promised outcomes.

Contact
cajole.cc · North America