Read the platform selection case, explore the source register and put the proposition through a practical trial.
An ambition the industry calls its “holy grail”
PwC and NYU’s 2023 report states: “Guest data is the holy grail of the hospitality industry.” It connects that knowledge with personal service and stresses governance and trust. The report draws on industry interviews and focus groups. Read the original report, page 18.
In Oracle–Skift’s 2022 report, Dan Kornick, identified as Loews Hotels’ CIO, describes knowing the customer and their data as the key to personalised service, using the same “holy grail” phrase. Read the original report, page 24.
These sources describe an industry ambition. They do not endorse GuestMemoryOS or establish one exclusive route to achieving it.
Why it matters commercially
Gallup’s 2014 hospitality research associates stronger guest engagement with greater spending and lower price sensitivity. Its analysis of return visits emphasises reliable service, responsive employees and problem resolution. These are reasons to evaluate better continuity, rather than a forecast of our product’s financial effect. Gallup’s hotel research.
What your team remembers. What your guest remembers.
The organisation’s memory is the useful knowledge available to the people delivering service: what the guest requested, what staff noticed, what changed and what worked.
The guest’s memory is the experience they carry away: feeling understood, having a concern followed through, or finding that the next colleague already has the context.
A 2018 study in the Journal of Travel Research reports relationships between memorable tourism experiences, revisit intentions and word of mouth. That supports the importance of the remembered experience; it does not prove that a software tool causes repeat bookings. Research abstract and citation.
Our proposition is that better operational memory can help staff create a better remembered experience. A trial should test whether that happens.
What hospitality already uses
Operators already have valuable foundations. PMS and CRM platforms can hold preferences, share profiles and support service workflows. AI can identify patterns and help select relevant offers. Staff briefings and recognition programmes carry human knowledge between colleagues.
Oracle OPERA documents guest preferences in its profiles. Revinate describes collecting preferences directly from guests. At sea, Princess says MedallionClass helps crew recognise preferences.
The question for a buyer is specific: how reliably does a discovery made during service reach the next appropriate employee, lead to an action, and contribute to what the team learns? GuestMemoryOS is designed around that daily workflow. Its incremental value should be evaluated alongside what the operator already has.
The missing detail has to be collected
Consider a message from someone booking for ten people: “Could we have gluten-free bread?” The request is real. The identity of the person who needs it may be absent. Reading that message again does not establish who it concerns.
During normal service, a staff member establishes who requested the bread and the stated reason. They record that context against the appropriate guest. The kitchen receives the relevant information and follows its usual dietary and allergy confirmation process.
The important difference is the evidence available. A digital interaction can contain an explicit, well-attributed request. A browsing signal can suggest an interest. A frontline interaction can supply new information that neither record contains.
GuestMemoryOS actively prompts staff to discover useful details and makes those details simple to record. AI helps organise the resulting evidence and produce recommendations. An observation retains its source and context; it does not automatically become a permanent preference.
What each source can establish
A browsing signal: an interaction with content, which may suggest interest but may not identify the intended consumer.
A booking message: what the sender actually stated. Individual attribution depends on the content.
A recorded service interaction: what the staff member learned about a particular guest in that situation. Its value depends on accurate capture, context and appropriate review.
These are information types, not claims that every competing product works in one way. How we distinguish observations, preferences and estimates.
People discover. Technology helps the whole team remember.
GuestMemoryOS combines five daily prompts for each staff role with separate, unlimited QuickNotes. Prompts reflect the setting and current stay. Staff answer through normal service; the guest does not complete a daily questionnaire.
One-button voice or text entry in the staff member’s own language keeps capture practical. AFG reports an average of six seconds per note. The system helps structure that input and asks for useful missing detail. The source and guest context matter more than the volume of notes.
A staff member can contribute without access to the full guest profile. The manager sees the combined answers and has separate guest-experience questions. Relevant departments receive service guidance.
The team uses that guidance during the current stay, records the response and passes useful knowledge to the next shift. Appropriate memories also support a return welcome. The service history provides examples for recognition and coaching.
GuestMemoryOS also makes inferences and estimates, including an estimated rebooking percentage. Those outputs are distinct from a guest’s own statements or an actual booking. People remain responsible for service decisions.
Standalone operation is available. PMS and CRM connections are scoped with the operator. GuestMemoryOS does not replace booking, billing or customer-relationship systems. Explore the operating fit.
The next property starts with a relationship.
At another participating hotel or ship, the team can prepare from the guest’s previous preferences before arrival. The new stay adds fresh context, while approved memories help the marketing team make relevant return invitations. Group sharing, identity and permissions are agreed with the operator.
Live adoption evidence. A measurable next step.
GuestMemoryOS operates at Mandarin Beach Villa, Koh Samui. AFG reports three days of training for its six-person Thai team, followed by continued use after the trainer left on 22 July 2026. Reported daily contributions include 11–18 additional observations per employee alongside the five assigned questions. These are company-reported results from this setting.
The residence’s separate public record contained 40 five-star Airbnb reviews when checked on 21 September 2026. Reviews reflect the whole guest experience and do not isolate software impact. Read the dated evidence and qualifications.
AFG has completed internal cruise simulations and a simulation of a 600-room hotel in Dubai. A live trial with a cruise line or larger hotel is the next goal. With the operator, we propose measuring useful staff contributions, handover continuity, completed service actions, repeated requests and guest feedback. Actual rebookings need follow-up; a predicted rebooking percentage remains an estimate.
Let your team show what they learn, what the next shift does with it, and whether guests notice the difference.
Read the longer category white paper · Explore the service behind the score