40 Questions Hospitality and Tourism Should Ask Before Trusting AI
Your revenue management AI might optimise price but kill the repeat business that pays your salary. Your guest communication AI might improve response time while erasing the warmth that made someone book with you instead of your competitor.
These are suggestions. Use the ones that fit your situation.
1When Duetto AI recommends a 35 percent price increase for Friday night, what guest segment loses access and never returns?
2Is your Duetto model trained on your actual repeat guest data, or on industry averages that treat your hotel like any other?
3Which bookings from loyal guests will your AI price out before you see the lifetime value you are sacrificing?
4Does your revenue management system measure the cost of a guest complaint about surge pricing, or only the revenue gained?
5When your AI sets rates based on demand forecasts, who decides whether a 10 percent occupancy gain justifies losing your brand's reputation for fairness?
6Are you comparing your current revenue per available room against the revenue you lose when priced-out guests choose a competitor?
7What happens to staff morale when they field complaints about AI-driven pricing that guests find unreasonable?
8Does your AI recommendation system account for the cost of acquisition for new guests to replace those you priced out?
9How often do you manually override your revenue management AI, and what does that tell you about whether it understands your market?
10If every hotel in your city uses the same Duetto pricing model, how is price the differentiator that wins you bookings?
Guest Communication and Experience
11When ChatGPT generates your pre-arrival email, does it know that your guests book you for a specific style of welcome that ChatGPT cannot replicate?
12Are your guest messages from AI template outputs, or do they reflect your actual property's character and your staff's voice?
13What personalisations is your Salesforce Einstein guest communication missing because it only has booking and transaction data?
14When a guest has a problem, does your AI communication system know when to hand off to a human, or does it resolve service failures on its own terms?
15Does your Revinate AI analysis tell you why guests say you are 'efficient but cold' compared to a lower-rated competitor?
16If your AI generates all guest follow-up messages, how will you know if guests are still choosing you for your brand or for convenience?
17When your AI spots a guest service failure, does it recommend process changes that treat symptoms or changes to staff behaviour that address root causes?
18Are guests actually responding to your AI-generated messages at higher rates, or just responding more quickly to a template?
19Does your AI guest experience system measure delight and loyalty, or only satisfaction scores and response times?
20What guest stories and feedback is your AI missing because it only processes structured data from forms and ratings?
Operations and Staff Decisions
21When your operational AI recommends fewer staff on a slow Tuesday, does it account for the training time and relationship-building that happens on quiet shifts?
22Is your AI suggesting housekeeping or kitchen changes based on what improves checklist compliance, or on what actually improves guest experience?
23Does your AI know the difference between a process that feels efficient to measure and a process that creates the experience guests remember?
24When your system flags a staff member as underperforming against AI-set metrics, are you seeing their actual impact on guest loyalty?
25Are operational metrics you are optimising for actually correlated with repeat bookings and guest referrals, or just easier to count?
26What decisions is your AI making about which guests get priority service, and are those decisions aligned with your hospitality values?
27Does your operational AI recommend changes that reduce labour costs at the expense of the attentiveness that builds guest relationships?
28When your AI suggests automating a task, have you considered what happens when guests lose the chance to interact with a staff member?
29Is your AI operational system measuring whether staff are developing the judgement to handle situations outside the process, or just following protocol?
30How often do experienced staff members push back against your AI's recommendations, and have you actually investigated why?
Brand Distinctiveness and Competitive Position
31If your Canva AI and your three nearest competitor hotels all use the same AI design tools, what makes your brand visually distinct?
32When you use the same ChatGPT prompts as other hotels for guest communication, where is your competitive advantage?
33Does your AI adoption plan account for the fact that your brand strength comes from what makes you different, not what makes you efficient?
34Are you using AI to automate what every hotel does, or to amplify what only your hotel does well?
35When you chose your Revinate and Duetto tools, did you choose based on what your guests value or based on industry benchmarks?
36What part of your guest experience cannot be optimised by AI without disappearing?
37If your AI systems were adopted by all your competitors tomorrow, would your hotel still be the one guests prefer?
38Are you using AI to commoditise your service or to personalise it in ways your competitors cannot match?
39What would happen to your booking rate if guests knew all your communications were AI-generated?
40Is your AI investment helping you compete on what you do best, or making you compete on cost and efficiency where you cannot win?
How to use these questions
Before you accept an AI recommendation, ask who benefits from that recommendation being right and who bears the cost if it is wrong. If your guests bear the cost, that is your signal to scrutinise harder.
Your satisfaction scores can go up while loyalty goes down. Measure the intangibles that AI cannot optimise: whether guests mention you to friends, whether they book with you again without comparing prices, whether staff feel empowered to break the process for a guest.
Every AI tool your hotel uses is also used by other hotels. Your competitive advantage is not in using the same tools but in the human judgement you apply after the AI recommendation.
When AI flags a pattern in guest behaviour or operations, always ask whether the pattern matters because it is a real problem or because it is easy to measure. Ease of measurement is not the same as business importance.
Set a rule that at least one person in your leadership team must understand why the AI made each major recommendation before you act on it. If no one can explain it, you are outsourcing your judgement.