AI SEO Statistics: Escape Rooms (2026-07 edition)
40 questions · 120 AI responses · 3 models · measured 2026-07-06
The question bank
The questions we tested — sampled from real buyer journeys in escape rooms.
Each model answered every question once, same wording, same day. These are the prompts behind every percentage on this page.
Show all 40 questions
Model by model
16-point average divergence: which AI you ask changes the answer.
The divergence index is the average gap between the most and least likely model per behavior. Higher = the models disagree more about escape rooms buyers.
| ChatGPT | Claude | Gemini | Consensus | |
|---|---|---|---|---|
| Recommends hiring a professional | 20% | 23% | 10% | 85% |
| Suggests DIY first | 10% | 5% | 0% | 90% |
| Names specific providers | 5% | 8% | 25% | 75% |
| Gives price or cost info | 13% | 13% | 13% | 88% |
| Tells to check reviews | 23% | 23% | 10% | 73% |
| Tells to verify credentials | 3% | 3% | 0% | 98% |
| Mentions case studies / portfolio | 0% | 3% | 0% | 98% |
| Mentions local proximity | 23% | 33% | 15% | 58% |
| Gives selection criteria | 48% | 50% | 33% | 50% |
| Warns about red flags | 10% | 23% | 8% | 83% |
| Asks a clarifying question | 60% | 70% | 5% | 18% |
| Recommends multiple quotes | 3% | 3% | 0% | 98% |
By model
How each assistant handled Escape Rooms questions.
Reading the 120 answers model by model shows how differently the three assistants treat the same escape rooms questions. On the most consequential behavior — whether to send the buyer to a professional at all — the rate ranged from 22.5% (Claude) down to 10% (Gemini), a 13-point gap on an identical question set.
Across the 40 escape rooms answers it produced, ChatGPT recommended hiring a professional in 20% of them and suggested a DIY approach first 10% of the time. It named a specific provider in 5% of answers (about 0.3 distinct providers per answer) and included price or cost information 12.5% of the time. ChatGPT asked a clarifying question before answering in 60% of cases, warned about red flags or scams in 10%, and told the buyer to verify credentials in 2.5%, averaging 448 words per answer. On the remaining cues it told the buyer to check reviews in 22.5%, pointed to case studies or a portfolio in 0%, and framed the choice around local proximity in 22.5%; a selection-criteria checklist appeared in 47.5% of its answers and a recommendation to gather multiple quotes in 2.5%.
Across the 40 escape rooms answers it produced, Claude recommended hiring a professional in 22.5% of them and suggested a DIY approach first 5% of the time. It named a specific provider in 7.5% of answers (about 0.3 distinct providers per answer) and included price or cost information 12.5% of the time. Claude asked a clarifying question before answering in 70% of cases, warned about red flags or scams in 22.5%, and told the buyer to verify credentials in 2.5%, averaging 262 words per answer. On the remaining cues it told the buyer to check reviews in 22.5%, pointed to case studies or a portfolio in 2.5%, and framed the choice around local proximity in 32.5%; a selection-criteria checklist appeared in 50% of its answers and a recommendation to gather multiple quotes in 2.5%.
Across the 40 escape rooms answers it produced, Gemini recommended hiring a professional in 10% of them and suggested a DIY approach first 0% of the time. It named a specific provider in 25% of answers (about 0.6 distinct providers per answer) and included price or cost information 12.5% of the time. Gemini asked a clarifying question before answering in 5% of cases, warned about red flags or scams in 7.5%, and told the buyer to verify credentials in 0%, averaging 279 words per answer. On the remaining cues it told the buyer to check reviews in 10%, pointed to case studies or a portfolio in 0%, and framed the choice around local proximity in 15%; a selection-criteria checklist appeared in 32.5% of its answers and a recommendation to gather multiple quotes in 0%.
Taken together, Claude is the assistant most likely to route an escape rooms buyer to a professional (22.5%) and Gemini the least (10%). ChatGPT produced the longest answers, at 448 words on average. Specific providers were named most often by Gemini (25%) — even there, roughly one answer in 4 carried a name.
Where they disagree
The behaviors where the choice of model changes the answer.
The divergence index for this study is 16.1 points — the average distance between the most and least likely model across the coded behaviors. The gaps below are where which assistant an escape rooms buyer happens to ask matters most:
- Asks a clarifying question: from 5% (Gemini) to 70% (Claude) — a 65-point spread.
- Names a specific provider: from 5% (ChatGPT) to 25% (Gemini) — a 20-point spread.
- Mentions local proximity: from 15% (Gemini) to 32.5% (Claude) — a 18-point spread.
- Gives selection criteria: from 32.5% (Gemini) to 50% (Claude) — a 18-point spread.
- Warns about red flags or scams: from 7.5% (Gemini) to 22.5% (Claude) — a 15-point spread.
The widest single gap — asks a clarifying question, 65 points — means an escape rooms buyer can receive materially different guidance on the same question depending only on which assistant they happen to open, so any visibility strategy built on a single model's behavior describes only part of the escape rooms market.
Where they agree
The points of near-consensus in Escape Rooms.
On other behaviors the three models move almost in lockstep — the points of near-consensus for escape rooms, where all three landed within a few points of each other:
- Gives price or cost information: 12.5% across all three models.
- Tells the buyer to verify credentials: 0%–2.5% across all three (a 3-point spread).
- Mentions case studies or portfolio: 0%–2.5% across all three (a 3-point spread).
- Recommends multiple quotes: 0%–2.5% across all three (a 3-point spread).
Measured question by question, the three assistants coded a response the same way most consistently on "tells the buyer to verify credentials" (identical coding in 97.5% of questions) and least consistently on "asks a clarifying question" (17.5%).
Every behavior, measured
All twelve coded behaviors for Escape Rooms, averaged across the three models.
The behaviors AI models reproduce most often for escape rooms are asks a clarifying question (45% on average), gives selection criteria (43.3%) and mentions local proximity (23.3%); the rarest are mentions case studies or portfolio (0.8%), recommends multiple quotes (1.7%) and tells the buyer to verify credentials (1.7%). Each figure below is the share of a model's 40 answers in which the behavior appeared at least once, averaged across the 3 models with the full per-model range in parentheses:
- Asks a clarifying question: 45% on average (ChatGPT 60%, Claude 70%, Gemini 5%) — a 65-point spread.
- Gives selection criteria: 43.3% on average (ChatGPT 47.5%, Claude 50%, Gemini 32.5%) — a 18-point spread.
- Mentions local proximity: 23.3% on average (ChatGPT 22.5%, Claude 32.5%, Gemini 15%) — a 18-point spread.
- Tells the buyer to check reviews: 18.3% on average (ChatGPT 22.5%, Claude 22.5%, Gemini 10%) — a 13-point spread.
- Recommends hiring a professional: 17.5% on average (ChatGPT 20%, Claude 22.5%, Gemini 10%) — a 13-point spread.
- Warns about red flags or scams: 13.3% on average (ChatGPT 10%, Claude 22.5%, Gemini 7.5%) — a 15-point spread.
- Names a specific provider: 12.5% on average (ChatGPT 5%, Claude 7.5%, Gemini 25%) — a 20-point spread.
- Gives price or cost information: 12.5% on average (ChatGPT 12.5%, Claude 12.5%, Gemini 12.5%).
- Suggests a DIY approach first: 5% on average (ChatGPT 10%, Claude 5%, Gemini 0%) — a 10-point spread.
- Tells the buyer to verify credentials: 1.7% on average (ChatGPT 2.5%, Claude 2.5%, Gemini 0%) — a 3-point spread.
- Recommends multiple quotes: 1.7% on average (ChatGPT 2.5%, Claude 2.5%, Gemini 0%) — a 3-point spread.
- Mentions case studies or portfolio: 0.8% on average (ChatGPT 0%, Claude 2.5%, Gemini 0%) — a 3-point spread.
Trust signals
How well the models protect the escape rooms buyer.
Beyond whether to hire, the rubric codes how carefully each assistant protects the escape rooms buyer once a decision is made. Telling the buyer to check reviews or ratings appeared in 18.3% of answers on average. Verifying credentials or certifications appeared in 1.7%. Warning about red flags or scams appeared in 13.3%.
On structuring the decision, a selection-criteria checklist showed up in 43.3% of answers on average and a recommendation to gather multiple quotes in 1.7%. The single least-reproduced protective signal for escape rooms is "tells the buyer to verify credentials" at 1.7% on average — the clearest opening for content that supplies it, since the models are not yet reliably surfacing that guidance on their own.
Referral behavior
Do AI models name Escape Rooms providers?
For service providers the decisive question is whether these systems name anyone at all. Across 120 escape rooms answers, a specific provider was named in 12.5% of responses on average — roughly 0.4 distinct providers per answer. In practice the assistants behave far more as an explanatory layer than as a referral engine for escape rooms: visibility comes from being the reasoning a model reproduces, not from being the named recommendation.
The question set
What these 40 Escape Rooms questions cover.
The 40 questions behind every percentage on this page were drawn from real escape rooms (professional services; buyer hiring decisions for this specific service) buyer journeys. Each was put to all 3 models once, with identical wording, so the rates above describe how the assistants handled this exact escape rooms question set — not a general prior or a hand-picked subset. The full list is shown earlier on this page; the coded percentages are what those specific questions produced.
How to read this
A note on the numbers.
A percentage here is the share of a model's 40 answers in which the behavior appeared at least once — not a confidence score. Because each model answered every question exactly once on 2026-07-06, the figures describe this specific escape rooms question set and snapshot rather than a general prior. The full protocol and coding rubric are documented in the study methodology.
Methodology
A controlled snapshot, documented end to end.
40 standardized buyer questions per industry, one response per model per question (ChatGPT (gpt-5-mini), Claude (claude-sonnet-5), Gemini (gemini-3-flash-preview)), collected 2026-07-06, coded against a fixed 12-behavior rubric with human QA. AI outputs vary with model version, location and time — figures describe this sample and window, and are refreshed each edition. Read the full methodology →