Your best product page ranks third for its main keyword and still doesn't appear when a prospect asks ChatGPT which vendors to shortlist. The buyer reads a written answer with a few brand names and a handful of source links, and yours isn't among them. Fewer people now click the blue links under that answer. In Pew Research Center's browsing-panel study of 900 U.S. adults, people who saw a Google AI summary in March 2025 clicked a traditional result on 8% of visits. People who saw no summary clicked on 15%.
AEO services are one way to get your brand into those answers. Hiring makes sense when competitors keep getting cited for prompts you should own and nobody on your team has the hours to fix it. If you already run a content team, a platform your team operates keeps the prompt sets, company context and tracking history with your team after a contract ends.
What AEO services are
AEO services are paid work, delivered by an agency or through a platform, that aims to get your brand named and cited in AI-generated answers. AEO stands for answer engine optimization. The full definition and core tactics are covered in our guide to Answer Engine Optimization (AEO) services.
Most engagements bundle the same core deliverables:
- An AI visibility audit with a baseline prompt set
- Entity, schema and crawlability fixes
- Answer-first rewrites and new pages
- Off-page authority work on the sources engines cite
- Ongoing prompt and citation tracking with regular reports
The engines in scope, and the vocabulary around them, decide how you judge every one of those deliverables.
The AI engines AEO services target
A credible AEO service covers six surfaces: ChatGPT search, Perplexity, Gemini, Claude, Google AI Overviews and Google AI Mode. Each one picks sources differently, so a brand can lead in one engine and be absent from another on the same prompt.
Two outcomes count. A mention is your brand named in the answer text, with no link. A citation is your URL attached to the answer as a source. You can win the mention and lose the citation. When that happens, the buyer hears your name but clicks through to a review site or a competitor's comparison page.
How AEO relates to SEO, GEO and LLMO
Teams use SEO to earn rankings and clicks. They use AEO to earn mentions and citations inside the answer itself. The two share most of their foundations, and we've laid out where they split in AEO vs. SEO.
GEO (generative engine optimization) traces to a paper by Aggarwal and colleagues at Princeton, first published in 2023 and presented at KDD 2024. The authors found that adding citations, quotations and statistics to a page raised its visibility in generative engine responses. LLMO (large language model optimization) describes the same goal from the model side. That means getting a model to retrieve, summarize and cite your content accurately.
Agencies sell these labels together, and the work behind them overlaps almost completely. Google's AI optimization guide treats AEO and GEO as names for visibility work in AI search and says that optimizing for generative AI search is still SEO. We use AI visibility as the umbrella term. AEO is the work you do, and AI visibility is what you measure.
What an AEO service includes
Providers split the scope into technical work that makes your pages readable and eligible, on-page work that makes them quotable, and off-page work that builds the third-party record engines draw on, plus reporting. The order below matches how a sound engagement runs, starting with a baseline.
AI visibility audit
An audit sets the baseline every later report is compared against. The provider builds a prompt set of non-branded questions your buyers actually ask, runs each one across the target engines, and records whether you're mentioned, whether you're cited, and which sources the engine cited instead.
A good audit also measures competitor share of answers on the same prompts. Then it lists the specific pages, missing pages and third-party sources behind each gap. Insist on repeated runs per prompt, because a single run captures one draw from a variable system. Our walkthrough on AI visibility auditing shows how to scope one yourself before you pay for it.
Entity and schema optimization
Providers use entity work to help engines identify your company. That means one consistent company name, category description and product naming across your website, your profiles and the directories that list you.
Structured data is a baseline technical deliverable, but it has limits. Google's optimization guide says structured data isn't required for generative AI search and that no special schema.org markup is needed. Markup still helps with rich result eligibility, and it has to match the visible text on the page. Ahrefs, a vendor with its own tracking product, matched pages that added JSON-LD between August 2025 and March 2026 against control pages. Ahrefs found no major citation uplift in AI Overviews, AI Mode or ChatGPT.
Crawlability matters more. OpenAI's crawler documentation says websites that opt out of OAI-SearchBot won't be shown in ChatGPT search answers. Check robots.txt for OAI-SearchBot, PerplexityBot, Claude-SearchBot and Googlebot before anyone writes a new page. Treat llms.txt as optional, because Google's guide states that Google Search doesn't use it.
Answer-first content
Writers creating answer-first content put the direct answer in the opening sentences under a question-led heading, then support it with specifics. AI engines write their own answer from the sources they retrieve, and this format gives them a clear passage to summarize.
Most teams should rewrite before they create. If a page already ranks or gets cited somewhere, restructure it and keep the URL so its history carries over. Build a new page only when nothing on your website answers a prompt buyers keep asking.
Off-page authority building
Providers target the sources engines already trust in your category through off-page work. Search Engine Land's coverage of a Peec AI study reports that Reddit was the most-cited domain across ChatGPT, Google AI Mode, Gemini, Perplexity and AI Overviews, followed by YouTube and LinkedIn. Review platforms such as G2 and Yelp showed up often on recommendation queries.
In practice the work covers review generation, inclusion in third-party category comparisons, genuine participation in communities, and earned coverage in trade press. Engines can name a brand based on forums, reviews, listicles and other third-party articles rather than owned content.
Citation and prompt tracking
Tracking proves the work and catches regressions. A provider should run the prompt set on a fixed schedule (daily or weekly), report monthly, and hold the prompt list steady between reports so changes reflect the market rather than a new question set. Model updates can shift visibility across every prompt at once, so a monthly report should flag them.
How much AEO services cost
AEO pricing clusters around the engagement type. First Page Sage's October 2026 pricing report, built from US-dollar agency quotes, gives the clearest ranges:
| Engagement | Typical price | What it usually buys |
|---|---|---|
| One-off audit | $1,500–$12,000, median $4,300, over 2–6 weeks | Baseline prompt results, citation gaps, technical blockers, prioritized fixes |
| Project or sprint | $5,000–$40,000, median $14,500, over 30–90 days | Audit plus a defined round of implementation |
| Custom monthly retainer | $3,000–$15,000/mo, median $7,200, typical 6-month term | Prompt tracking, content, technical fixes, PR and reporting |
| SaaS tracking tool | Semrush AI Visibility Toolkit is $99/mo per domain, billed annually, with 25 custom prompts | Monitoring only. Your team does the work |
| Platform with expert guidance (GrowthX) | From $6,000/mo | Your team operates the platform. GrowthX configures it and advises |
The same report also lists tiered package retainers ranging from $2,500/mo to $25,000/mo (median $6,300/mo) and performance-based pricing with a median base of $2,600/mo.
Retainer tiers buy different things. Lower tiers mostly buy monitoring and reporting, and higher tiers add content production and digital PR. If a retainer doesn't include a tracking tool, budget for one on top.
Agency, platform or in-house delivery
The delivery model decides who holds the knowledge when the contract ends. For a CMO, that matters more than the monthly fee. Here's how the four common models compare:
| Model | Cost | Team effort | Outputs | Who owns the knowledge after the contract |
|---|---|---|---|---|
| Agency retainer | Monthly fee, usually a 3–6 month minimum | Low. Your team reviews and approves | Audits, content, PR, reports | The agency, unless the contract transfers prompt sets, playbooks and tracking history |
| Platform with expert guidance (GrowthX) | Subscription | Medium. Your team makes daily decisions, reviews and publishes | Company context, scored page inventory, opportunities, drafts sent to your CMS | Your team. Context and writing rules sit in your workspace, and published pages live in your CMS |
| Tracking tool plus in-house team | Tool fee plus existing salaries | High. Your team runs every fix | Visibility data only | Your team |
| Full in-house build | New headcount plus tools | Highest | Whatever the team builds | Your team |
Pick the model that matches your constraint:
- Agency: Fits when you have no content team, need results inside a quarter, and accept that the playbook leaves with the agency.
- Platform with expert guidance: Fits when you have a content team that keeps rebuilding briefs across disconnected tools and you want the context to compound in a system you operate.
- Tracking tool plus in-house: Fits when you have strong SEO operators and only lack measurement.
- Full build: Fits large enterprises with AI search as a standing board priority and budget for dedicated roles.
Tools for answer engine optimization and tracking
Tools fall into three categories. A tracker tells you where you stand without changing a single page.
- Visibility trackers: Profound, Peec AI, the Semrush AI Visibility Toolkit, Ahrefs Brand Radar and OtterlyAI run prompt sets on a schedule and report mentions, citations and share of voice. Engine coverage varies by plan. OtterlyAI's plans include AI Overviews, ChatGPT, Perplexity and Copilot, and charge extra for AI Mode, Gemini and Claude.
- Schema and structured data tools: Schema App and WordLift manage markup and knowledge graphs at scale. Screaming Frog validates JSON-LD, Microdata and RDFa across a full crawl.
- AI content platforms: AirOps, Surfer, Clearscope and Writer handle drafting and optimization. Surfer, Clearscope and AirOps now bundle some prompt tracking as well.
CheckThat sits apart from that set. CheckThat is GrowthX's free, open AI visibility index for B2B software brands. It tracks how ChatGPT, Claude, Perplexity and Google AI answer buyer questions across software categories, with human editorial review on every prompt. Use it to see who wins your category's prompts before you buy any tracker.
How to measure AEO results
Three KPIs cover most reporting needs, with GA4 setup below to isolate AI referrals:
- Citation rate: The share of tracked answers that cite one of your URLs as a source.
- Share of voice: Your brand mentions divided by all brand mentions across the prompt set, times 100. Vendors use different denominators, so compare share of voice only within one tool and one prompt set.
- LLM referral traffic: Sessions arriving from AI assistants in your analytics, tracked against conversions.
Answers vary from run to run. SparkToro's repeated-run tests, run in November and December 2025, found AI tools rarely returned the same list of brands twice for the same prompt. Report the percentage of runs that mention you over several weeks, and treat single-day readings as noise.
Set up an AI channel in GA4
GA4 now ships an AI Assistant channel in its default channel group. Google says the channel covers referrals from sources like ChatGPT, Gemini, Deepseek, Copilot and Grok, and excludes AI Overviews and AI Mode. Google doesn't publish its full referrer list, so build a custom channel as a backstop:
- Go to Admin, then Data display, then Channel groups.
- Open the three-dot menu on the default channel group and choose "Copy to create new."
- Click "Add new channel," name it, set the condition to Source, choose "matches regex," and paste the pattern below.
- Click Save channel, then use Reorder to move the AI channel above Referral. GA4 assigns traffic to the first matching rule, so a lower position lets Referral swallow AI visits.
Analytics Mania's AI traffic pattern covers chatgpt.com, perplexity.ai, gemini.google.com, claude.ai and copilot.microsoft.com, among others:
chatgpt\.com|chat\.openai\.com|gemini\.google\.com|deepseek\.com|perplexity(?:\.ai)?|claude\.ai|copilot\.microsoft\.com|deepl\.com|character\.ai|(?:\w+\.)?meta\.ai|grok\.x\.com|grok\.com|x\.ai|bard\.google\.com|(?:\w+\.)?mistral\.ai|writesonic\.comGA4 still undercounts this traffic. Many AI platforms strip the referrer, mobile apps often hand clicks to a browser with no origin, and copied URLs arrive as Direct.
Clicks from AI Overviews and AI Mode land in organic Google traffic. Add a "how did you hear about us" field to demo forms to recover what analytics misses.
Then check the pages themselves. Look at whether the pages that answer your target questions get impressions, clicks and AI referrals, and whether those visits convert.
Every page you publish adds data. Use it to expand your question map with directly relevant buyer questions and the adjacent ones your market asks.
For board reporting, CheckThat's measurement model groups AI visibility into Presence, Reputation, Perception and Influence. Put each on a slide with the prompt set, engines and collection period stated, so the board compares like with like quarter to quarter.
When to hire AEO services
Hire when two or more of these signals show up at once:
- The board asks what AI says about you: Nobody can answer with a measured baseline, only screenshots.
- Competitors get cited instead: Your rivals appear on buyer-intent prompts you should own. Seer Interactive's 2026 CTR update, drawn from its agency client data, reports a full-year average organic CTR of 2.07% on informational queries for brands cited in an AI Overview, against 0.94% for brands not cited.
- Clicks fall while rankings hold: Positions look stable in Search Console, but traffic to key pages keeps sliding.
- No one has the hours: Your content team is already at capacity, and AEO would sit at the bottom of every sprint.
If only the first signal applies and you have a content team, buy a one-off audit and run the fixes yourself. A retainer earns its fee when the gap spans technical, content and off-page work at once.
Before you sign, ask for a before/after case with the client named, the prompt set, the engines, the baseline and the measurement window. A percentage lift can't be assessed without a baseline, so require all five before you weigh the result.
How to choose an AEO provider
Judge providers on four criteria:
- Engine coverage: Confirm which of the six engines they track on the plan you'd buy, and whether Google AI Mode and Claude cost extra.
- Tracking method: Ask how many runs per prompt, how often, from which location and login state, and whether the prompt set stays fixed between reports.
- Data ownership: Get it in writing that the prompt set, tracking history, briefs and content transfer to you when the contract ends.
- Realistic promises: Expect commitments on work delivered and metrics reported. Treat outcome guarantees as a warning.
Google's hiring guidance for SEO providers says to be wary of guarantees of #1 rankings, check references and request an audit. The same test applies to AI answers. Walk away when you see any of these red flags:
- Guaranteed citations or a guaranteed position in ChatGPT or AI Overviews
- Paid "brand placement" packages sold as a mention-rate lift
- llms.txt or special AI markup pitched as the headline fix
- Results shown as single screenshots instead of rates across repeated runs
- New acronyms with no plain definition of the deliverable behind them
- No written methodology for the prompt set or the share-of-voice formula
When a provider clears all four criteria and none of the red flags, start with a scoped audit and judge them on the gaps they find. If your team already does this work and keeps losing context between the audit, the brief, the draft and the report, the GrowthX website portfolio management platform keeps company context, a scored page inventory, opportunities and drafts in one workspace your team operates, and AI visibility is tracked on segments GrowthX configures for you. Book a demo to see it running on your own pages.
