AI Answers vs. Your Pricing Page: Why ChatGPT Quotes the Wrong Price (and How to Fix It)
Ask ChatGPT or Perplexity "how much does [your service] cost?" and you'll often get a confident number. Sometimes it's yours. Often it's a two-year-old figure, a competitor's rate, or a guess stitched together from forum posts. If AI is quoting your prices wrong, you're losing buyers before they ever reach your site.
Why pricing questions are the highest-intent AI queries
"Cost," "price," and "how much" queries come from people close to buying. When an assistant answers them, it frames the whole decision: who is affordable, who is premium, who is "worth it." If your pricing is missing or wrong, the engine fills the gap — and the brand with clear, crawlable pricing gets named instead. This is core answer engine optimization territory.
Where AI engines actually get your prices
Answer engines blend several sources, roughly in this order of trust:
- Your own pricing page — if it's crawlable plain HTML, not a JavaScript calculator or a PDF.
- Structured data —
Offer,PriceSpecification, andProductschema with explicit currency and value. - Third-party listings — directories, review sites, and marketplaces that display price ranges.
- Discussions — Reddit threads, forums, and old blog posts. These are the usual source of stale or wrong numbers.
Common pricing problems and why they happen
| What AI says | Why it happens |
|---|---|
| An outdated price | Old blog posts, cached directory listings, or archived pages outrank your current page. |
| "Pricing isn't publicly available" | Your page says "Contact us" with no ranges, so the engine has nothing to quote. |
| A competitor's number attributed to you | Comparison articles mix up vendors; no clean source contradicts them. |
| A wildly wrong range | Prices live inside a quote widget, image, or PDF the crawler can't parse. |
| You're labeled "expensive" | No context on what's included, so the engine compares only headline numbers. |
"But we don't publish pricing"
You don't need a full rate card. Engines reward any clear, honest anchor: "Most projects run $2,000–$6,000," "Plans start at $49/month," or "Typical install: $8–12 per square foot." A range with what drives the price up or down gives AI something accurate to repeat — and lets you control the framing instead of a stranger on a forum.
The pricing-page fix list (in priority order)
- Publish a range or starting price in plain HTML text near the top of the page.
- Answer the question directly with an FAQ like "How much does X cost?" — a 2–3 sentence answer engines can lift verbatim.
- Explain what's included so you're compared on value, not just the headline figure.
- Add
Offerschema with price, currency, and validity dates. See our schema guide for AI. - Show a "last updated" date — freshness signals help engines prefer your page over old mentions.
- Sync third-party listings so directories and review profiles show the same numbers.
- Reference your pricing page in llms.txt so AI crawlers find the canonical source fast.
Work these into your broader AEO process and tick them off with the AEO checklist.
Check what AI is saying right now
Pricing answers drift as engines recrawl and re-rank sources, so this isn't a one-time fix. Ask the major assistants your top three price questions every month and log the answers. Or let Rooster do it: the free AI Visibility Scan shows how ChatGPT, Perplexity, and Google's AI answers describe your business — including the prices they quote — and the AI Visibility Tracker alerts you when those answers change.