Service
AEO (Answer Engine Optimization)
Make your site easy for AI answer engines to reach, read, and quote accurately.
The short answer
AEO (Answer Engine Optimization) is our name for a way of working, not a platform feature. Google's own guidance says the fundamentals of search still apply to AI experiences, and that no special AI schema is required, so AEO here means doing those fundamentals deliberately for a buyer question: crawler access for the bots that feed AI answers, indexable and server-rendered pages, a clear factual answer near the top of the page, consistent facts across every place your company is described, and original evidence worth quoting. We then observe which prompts actually surface you, and report what we see.This page covers how aeo (answer engine optimization) is run: deliverables, cadence, and process. Package inclusions and prices live on the pricing page, and category-specific detail lives on the industry pages.
Deliverables
What you actually get.
- Crawl access review: whether Googlebot, Google-Extended, GPTBot, OAI-SearchBot, ClaudeBot, and PerplexityBot can reach your pages, and what each choice costs you
- Indexability and rendering fixes so the content exists in the HTML response, not only after JavaScript runs
- Answer-first content structure: a direct factual answer, clean heading hierarchy, scannable definitions
- Factual consistency pass so pricing, product names, positioning, and company facts match everywhere
- Structured data (Organization, Product, Service, FAQ, HowTo) for machine clarity and search feature eligibility, not as a citation lever
- Original evidence: benchmarks, methodology, and data other people have a reason to quote
- Observed citation tracking: a fixed prompt set, checked on a schedule, reported with the raw outputs
- Optional experiments we run and measure, including llms.txt, with results reported honestly
Outcomes
What changes in the business.
- AI systems can reach, read, and correctly parse the pages that matter
- The facts they find about you are consistent and current
- You have observed prompt data instead of assumptions about AI visibility
Process
How we run it.
01
Baseline
Record crawl access, indexability, factual inconsistencies, and current outputs for a fixed prompt set.
02
Fix access
Clear the technical barriers first, since nothing else matters if the content cannot be reached or read.
03
Earn the quote
Publish clear answers and original evidence on the questions your buyers actually ask.
04
Observe
Re-run the prompt set monthly and report what changed, including when nothing did.
Where this sits
AEO (Answer Engine Optimization) is one stage of a connected system.
Run on its own it can help. Run inside the system, it compounds with the other seven stages. Step through to see how they connect.
Stage 03
Search and AI discovery
SEO, AEO, and GEO run together so buyers and answer engines both find and describe you correctly.
AEO vs GEO
Two halves of one program, in our own words.
AEO and GEO are industry terms and our internal operating labels, not official platform categories. Google states that its search fundamentals still apply to AI experiences, so most of this is those fundamentals done deliberately.
AEO: reachable, readable, quotable
On-site work: crawler access, indexable server-rendered pages, answer-first structure, consistent facts, structured data for machine clarity, and original evidence worth quoting. It improves the odds a page can be used as a source. That is this page.
GEO: better public evidence
Mostly off-site work: consistent public facts, accurate listings and profiles, corrected outdated claims, and earned authoritative mentions. No one outside the model providers controls the output, so we improve the evidence and report what we observe. See GEO (Generative Engine Optimization) or read the GEO vs AEO guide.
Every Momentence tier includes both, because they share a prompt set, a content pipeline, and an entity graph. Running one without the other leaves you either cited but poorly described, or well described and never quoted.
Measurement
The engines we track, and how often.
AEO is only real if the citation is measured. Every tier runs a fixed prompt set against named engines on a schedule, with verbatim answers archived so change is auditable.
ChatGPT
Chat answers and ChatGPT search results, with and without browsing
Google AI Overviews
AI Overviews and AI Mode on the queries your ICP runs
Perplexity
Answers and the cited source list, including follow-up prompts
Gemini
Gemini answers and Google's generative surfaces
Claude
Claude answers with and without web access
| Tier | Prompt set | What is covered |
|---|---|---|
| Ignite | 5 observed prompts | Category and brand prompts in one market and language, checked monthly. |
| Momentum | 20 observed prompts | Category, comparison, and problem prompts across assistants, checked monthly with observed mention rates. |
| Category Leadership | 50+ observed prompts | Full buyer-stage prompt set, competitive mention comparison, and a correction workflow for wrong or outdated claims. |
| Custom | Prompt set per statement of work | Multi-product, multi-brand, or multi-market prompt sets sized to the portfolio. |
- Weekly written update in a shared Slack channel: what shipped, what moved, what is next.
- Monthly performance report covering observed citations, observed mentions, ranking movement, and pipeline, including months with no movement.
- Quarterly strategy review that re-scores the prompt set and retires prompts that stopped mattering.
No tasks, no seats, no overage fees or surprise usage metering. The prompt counts above are scope, not a meter. See published pricing.
Sample scorecard
What lands in your inbox every month.
This is the reporting artifact, metric by metric. Nothing here requires a login to a dashboard we own.
| Metric | Definition we use | How it is reported |
|---|---|---|
| Observed citation rate | Share of the observed prompt set where a page on your domain appeared as a cited source, by assistant, on the dates we checked. | Percentage plus the prompt-level list, with the raw answers archived |
| Observed mention rate | Share of the observed prompt set where your brand was named at all, cited or not, against a fixed competitor set. | Percentage against up to five named competitors |
| Prompt coverage | Share of your observed prompt set that has a clear, quotable answer published on your own domain. | Covered, partially covered, and uncovered counts |
| Description accuracy | Whether the sentence an assistant generated about your category position, product, and pricing was factually correct. | Verbatim quotes with each inaccuracy and its likely source |
| Engine split | The same observations broken out per assistant, because changes rarely arrive on all of them at once. | Per-engine table with month-over-month change |
| Pipeline endpoint | Sessions, conversions, and qualified conversations attributable to AI search referrers and to the pages we build. | Reconciled to your analytics and CRM, with the attribution limits stated |
Metric definitions matter more than the chart. We publish ours in the AI visibility metrics guide so you can hold any vendor, including us, to the same wording.
Stack
The tooling we operate on your behalf.
Tooling is necessary and insufficient. Here is what we run, so you know the work is instrumented rather than manual.
Search and ranking data
Google Search Console, Semrush, and server log analysis for crawl and index behavior.
AI answer monitoring
A fixed prompt set run against each assistant on a schedule, with verbatim answers and cited sources archived so month-over-month change is auditable. We report what we observed, and we do not claim the outputs are controllable.
Models in production
GPT, Claude, and Gemini class models used for research, drafting support, and structured data generation, always with a human editor before publish.
Publishing targets
Webflow, WordPress, Contentful, Sanity, HubSpot, and modern React codebases, or we build and own the site ourselves.
Measurement
GA4, one attribution tool, and your CRM as the reporting endpoint rather than a dashboard we control.
Comparing us to a self-serve visibility platform? Read platform vs agency and the honest AEO tools landscape.
How this differs by industry
AEO (Answer Engine Optimization) is not run the same way in every category.
The queries buyers run, the trust signals they need, and the review process we work inside change by category. Each industry page sets out what changes and what stays the same.
B2B SaaS
Horizontal SaaS built for teams, sold through pipeline.
Vertical SaaS
Industry-specific SaaS: legal, healthcare, construction, hospitality, and more.
Developer Tools
Developer-first products with PLG motion, docs, and community.
Fintech
Regulated financial software: payments, lending, wealth, and infrastructure.
AI & ML Products
AI-native products, model APIs, and ML platforms.
Consumer Software
Secondary capability. Consumer apps where search, content, and paid drive growth.