AI Visibility Series · September 2026 · RevParGenius Intelligence · RevParGenius Editorial Team · 6 min read
A boutique heritage hotel in Camperdown, Victoria ran RevParGenius's AI Visibility Check to see how often ChatGPT, Perplexity, Gemini and Grok actually recommend it to travellers asking real questions. The result was a solid-looking score — and a much more interesting story underneath it.
The property showed up in two-thirds of the AI answers it was checked against, with mostly positive sentiment. But it was completely invisible on two of the exact queries a heritage-focused, budget-conscious Great Ocean Road traveller would actually type. That gap — not the headline score — is the part worth understanding if you run a hotel anywhere near a major regional tourist route.
A single AI Visibility Check run scored this Victoria property 78/100, with a 67% mention rate across 12 tracked prompts. It was strongest on generic local-accommodation searches and completely absent on two specific, high-intent queries about heritage stays and budget access to the Twelve Apostles. A strong AI visibility score doesn't confirm bookings are healthy — and the gaps that matter most are usually hiding inside the prompts you'd expect to win.
This Property's Results
What Does an AI Visibility Check Actually Measure?
AI Visibility Check runs a hotel's name against a set of realistic traveller prompts — "best boutique hotels in [town]," "where to stay for couples," "heritage accommodation near [landmark]" — across ChatGPT, Perplexity, Gemini and Grok, and records whether the property gets mentioned, where it ranks in the answer, and whether the mention reads positive, neutral or negative.
It is a proxy for one specific question: does this hotel show up when a traveller asks an AI assistant something they would plausibly actually ask? It is not a measure of website traffic, booking conversion, occupancy, or overall marketing performance — a point worth keeping in mind before drawing conclusions from any single run.
What Did the Check Actually Find for This Property?
On broad, generic queries — best boutique hotels in the area, where to stay for couples, where to stay for families, luxury accommodation, top-rated hotels — the property was mentioned positively across nearly every responding engine, often ranking first. That is a genuinely good result: it means the hotel is not invisible in the kind of everyday search an AI assistant increasingly handles instead of a Google results page.
Sentiment on the mentions it did receive was strongly positive — 20 of 24 mention cells came back positive, one neutral, three with no sentiment data returned. So where the property showed up, it showed up well.
One technical note worth flagging plainly: one of the four AI engines returned an error on every single tracked prompt for this run. That is why the mention-rate calculation is based on 36 valid checks (three working engines × 12 prompts) rather than the full 48 cells. This is not a strike against the hotel — it is a reminder that a single failed engine can quietly reduce your sample without changing your headline percentage.
Where Did the Gaps Show Up — and Why Do They Matter?
The property was invisible — across all three responding engines — on two specific prompts: "unique heritage accommodation near the Great Ocean Road" and "where can I stay near the Twelve Apostles without paying coastal prices." Both are exactly the kind of long-tail, intent-rich question a real traveller comparing regional Victoria options would ask when they are close to a booking decision.
Coverage was also patchier on the Colac-to-Warrnambool corridor search and on "best places to stay that can be booked online immediately" — both practical, decision-stage queries rather than generic discovery ones. The pattern is consistent: this property is well covered for "does this town have a nice place to stay?" and much less covered for the specific angle — heritage character, value versus the coast, route-based positioning — that would actually tip a comparison in its favour.
A hotel showing up for "hotels in [town]" is table stakes. The prompts that actually influence a booking decision tend to be the specific ones — near a landmark, cheaper than the coast, heritage character, pet-friendly, self-contained. Those are also the prompts a generic listing or thin website page is least likely to answer well, which is exactly what showed up in this run.
What Should You Double-Check Before Trusting Your Score?
For anyone reading their own report for the first time, a few things are worth verifying rather than taking at face value. This run listed two name variants of what was almost certainly the same competitor business as separate entries in the visibility table — effectively double-counting one property. Confirm and merge duplicate competitor names before drawing conclusions about who is ahead of whom.
The report also noted this property's OTA citation share (24.2%) sits well below a 2025 industry baseline (55.3%), and framed that as a sign of a healthier direct-booking pipeline. That is a reasonable hypothesis — but citation share alone does not prove it. It would take actual booking-channel data to confirm whether that gap reflects strong direct bookings or simply fewer OTA listings being surfaced in the AI engine responses checked.
Because this was a single run, there is no trend yet. A second check, ideally a few weeks out and after addressing any specific content gaps identified, is what turns one score into a signal you can actually act on.
RevParGenius Take
A good score can still be hiding your two most valuable gaps.
This property is not invisible — it is genuinely well covered for generic search. The value of running the check was not the 78/100. It was finding the exact two prompts where a heritage-and-value-seeking traveller near a major tourist route currently gets recommended a competitor instead. That is a fixable content gap, not a fundamental visibility problem. But you only find it by looking at the specific prompts, not the average score.
Should Every Hotel in Regional Victoria Run This Check?
Regional Victoria has dozens of towns competing for the same Great Ocean Road, Twelve Apostles and Grampians traffic — Camperdown, Warrnambool, Colac, Port Fairy, Ballarat, Halls Gap and many more. According to Skift Research's 2024 AI Travel Planning Adoption Survey, 56% of US travellers now use AI tools to plan trips — and the pattern is spreading rapidly to Australian inbound and domestic travellers alike. If AI assistants are increasingly where trip planning starts, the question is not whether your hotel exists online. It is whether it gets named on the specific questions that lead to a booking, not just the generic ones.
You can run the same check on your own property for free — same four engines, your own set of prompts, same score and gap breakdown shown above. It will not tell you why bookings are up or down, but it will tell you, in about ten minutes, exactly which real traveller questions you are currently missing.
Frequently Asked Questions
What does an AI Visibility Score of 78 actually mean for a hotel?
An AI Visibility Score of 78 out of 100 reflects how often a property was mentioned, how positively, and how prominently across a set of tracked prompts run against ChatGPT, Perplexity, Gemini and Grok in a single check. It is a snapshot of AI search presence — not a measure of bookings, traffic, or overall marketing health. Two hotels with the same score can have very different gap profiles depending on which specific prompts they are missing.
Does a high AI mention rate mean a hotel doesn't have a marketing problem?
No. Being mentioned by an AI assistant does not establish search volume, website conversion, or occupancy. A property can be well covered in AI answers and still have weak bookings for reasons the visibility check simply does not measure — pricing, website UX, channel mix, or seasonal demand. The check answers one question: is this property being named by AI when travellers ask? It does not answer whether those naming events are converting.
Why would an AI engine show an error instead of a result?
Occasional failed responses from a given engine are normal in any automated check run across multiple AI providers. When one engine errors on every prompt in a run, it reduces the total valid sample size — which is why this run is based on 36 valid checks rather than 48. Worth checking before comparing mention-rate percentages across different runs or properties.
How often should a hotel re-run its AI Visibility Check?
A single run gives a snapshot. A second run a few weeks later — especially after addressing a specific content or schema gap identified in the first run — is what turns the score into a trend you can act on. Monthly checks are a reasonable baseline for most regional properties; quarterly is the minimum if you want to track whether changes you have made to your website or listings have shifted your AI presence.
What is the fastest way to improve AI visibility for a specific prompt where a hotel scores zero?
The fastest fix for a zero-mention prompt is usually a content gap: a page, blog post, or listing description that directly and specifically answers the question the traveller is asking. AI engines surface properties that are explicitly described in terms matching the query — "heritage accommodation near the Great Ocean Road" needs those exact words in credible, indexed content. Writing a 400-word page or blog post targeting the exact phrasing of the missed prompt, then re-running the check three to four weeks later, is the standard way to test whether the fix worked.
Related Reading
→ Camperdown vs Warrnambool: what the STR demand data shows
→ Best AI visibility tools for hotels in 2026
→ How AI Visibility Check works — full methodology
Free AI Visibility Check
See Where Your Hotel Stands on ChatGPT, Perplexity, Gemini and Grok
Run the same check on your own property — your score, your mention rate, and exactly which traveller questions you are currently missing.
Run Your Free AI Visibility CheckRead All Hotel Intelligence Reports
No commitment. Live data at revpargenius.com
Data sources: RevParGenius AI Visibility Check report, run 18 September 2026 07:18 UTC (single run, not independently re-run for this article); Cloudbeds 2025 OTA citation-share baseline as cited in the source report; Skift Research 2024 AI Travel Planning Adoption Survey. Hotel identity withheld by request. RevParGenius is an independent hotel market intelligence platform — not affiliated with any OTA, revenue management system, or hotel chain.
