If Your Service Pages Hide Pricing, AI Search Won't Cite You
You added structured data. You cleaned up your Google Business Profile. Your name, address, and phone number match across a dozen directories. And still, when someone asks ChatGPT or Perplexity which HVAC company to hire in your city, your name doesn't appear.
The problem may have nothing to do with schema.
The Four Conditions That Eliminate Your Citation Odds
A paper accepted at SIGIR 2026 -- arXiv:2605.25517, "What Gets Cited: Competitive GEO in AI Answer Engines" -- ran 252,000 controlled citation trials across six language models. Each trial put two web pages in head-to-head competition for the same citation slot, changing one factor at a time. The researchers identified four conditions that eliminated citation probability regardless of a page's other strengths:
1. **Topic mismatch** -- the page doesn't answer the specific query 2. **Price not mentioned** -- for queries where cost is a primary concern, absent pricing information is disqualifying 3. **Old timestamp** -- stale content loses to recent content 4. **Low list position** -- a low retrieval rank overrides content quality
These aren't factors to balance against each other. Failing any single one cancels out everything else. A page with strong schema, fresh content, and solid directory signals still loses the citation slot if it trips on a gatekeeper condition.
Timestamp and retrieval position get talked about in AEO circles. The pricing condition doesn't -- and it's the one most service businesses are silently failing.
What "Price Not Mentioned" Actually Means
When a buyer asks an AI engine "how much does it cost to replace an HVAC system in Burlington" or "how much does emergency plumbing cost on a weekend," the engine retrieves candidate sources and runs an implicit comparison. A page from your competitor that says "furnace replacement typically runs $5,000--$12,000 depending on unit size and home square footage" wins over your page that says "contact us for a free estimate."
Both pages may rank similarly in Google. In that AI head-to-head, your page fails the gatekeeper condition. The engine treats absent pricing information as a signal that this page can't answer the buyer's cost question -- and routes the citation to the page that can.
In our November 10, 2026 methodology review (Scout Session 195), we cross-referenced the SIGIR findings against Sourcepull's current audit template. The result: our fix plans cover NAP corrections, schema markup, directory presence, and content freshness. None of them currently include a check for whether the client's service pages contain any pricing information.
For a plumber whose buyers are constantly asking "how much does it cost to fix a burst pipe," that's not a minor optimization gap. It's a gatekeeper failure on the query type that drives most of their business.
Which Categories Are Most Exposed
The gatekeeper applies to queries where cost is a primary buyer concern. Based on the SIGIR paper's framing and the query sets we see across Sourcepull audits, the highest-exposure categories are:
- HVAC (replacement estimates, repair-vs-replace decisions, service call rates) - Plumbing (emergency rates, pipe repair and replacement ranges) - Roofing (replacement estimates, insurance vs. out-of-pocket scope) - Electrical (panel upgrade costs, hourly rates) - Dental (exam fees, crown and procedure pricing, insurance coverage gaps) - Legal (consultation fees, hourly rates by matter type) - Moving services (hourly vs. flat rate, distance-based pricing) - Home cleaning (per-visit vs. recurring, by square footage)
If your buyers ask AI engines about cost as part of how they decide who to hire -- and they do -- your service pages need to meet them on that question.
What to Add to Your Service Pages
The fix doesn't require publishing a full rate card if your pricing is variable. It requires giving the AI engine enough pricing context to treat your page as a viable answer to a cost query.
For most service businesses, that means including at least one of the following on each primary service page:
- **A price range:** "Furnace replacement typically costs $4,500--$10,000 in Ontario, depending on equipment type and home size." - **Cost factors:** "HVAC service call pricing varies based on the type of repair, parts required, and whether it's a weekend or after-hours call." - **A starting-at figure:** "Drain cleaning starts at $150 for standard residential lines." - **A direct cost framing:** "Most clients pay between $X and $Y for this service. Exact pricing depends on scope -- we provide written estimates before any work starts."
This content belongs on your primary service pages. Not on a separate pricing page, not in a buried FAQ three clicks away. The AI engine retrieves pages -- it doesn't navigate your site's information architecture. If pricing context isn't on the page it fetches for the cost query, it doesn't count toward passing the gatekeeper.
The Broader Pattern: Silent Disqualifiers
The pricing gatekeeper is one example of a category of AEO problems that share a structure: conditions that silently eliminate citation odds without triggering any error, warning, or ranking drop.
Another example from our September 2026 methodology review (Scout Session 142, 2026-09-18): Google's AI Mode opt-out toggle. A single Search Console setting blocks a site from AI Overviews, AI Mode, and AI Overviews in Discover simultaneously. A developer who follows generic "protect organic traffic" advice and enables that toggle has effectively removed the site from Google's AI surfaces entirely -- with no error message, no notification. The site ranks normally in traditional search. It just stops appearing in AI-sourced answers.
Old timestamps work the same way. A service page last updated in 2023 doesn't trigger an error. It consistently loses to competitor pages updated within the past year.
These silent disqualifiers are the hardest part of AEO to diagnose because nothing breaks visibly. Schema validates. The page ranks. The GBP is claimed. But in the AI citation competition, the page is failing a necessary condition -- and the only sign is the continued absence from AI recommendations.
What to Check Before Optimizing
Before investing in structural content rewrites or additional schema markup, run a diagnostic pass against the four gatekeeper conditions on your highest-priority service pages:
1. Does each service page contain pricing information relevant to the cost queries buyers actually ask? 2. When was the page last updated? Stale pages lose regardless of other quality signals. 3. Have you verified that the Google AI Mode opt-out toggle is turned off in Search Console? 4. Does the page rank in the top 10 organically for the query it's meant to answer? Low retrieval rank is itself a gatekeeper -- content optimization cannot substitute for it.
That diagnostic order follows directly from the SIGIR research: check gatekeeper failures before optimizing for citation quality. A page failing a necessary condition cannot be fixed with better FAQ formatting or more schema. The gatekeeper has to be cleared first.
Sourcepull's Signal Check runs the query test against your domain and surfaces which conditions you're currently hitting. It's free and takes about two minutes.
See how your business scores on AI platforms.
Check your score — free