Cancellation risk, while there's time to act
Signals scores each reservation for cancellation risk from lead time, rate plan, channel and history. High-risk weekends get a reconfirmation message or a waitlist offer — drafted for you to approve.
Solutions · Hotels, resorts & lodges
See the empty room coming before it's empty
Independent hotels and lodges run on small teams and thin margins. Nova reads your booking patterns for cancellation risk and soft weeks, builds packages around what guests actually do nearby, and gives the front desk a concierge that knows today's road conditions.
Timeline
A week in late October at a 28-room lodge
High desert, two hours from the nearest airport, a team of nine. Fictional, and probably familiar.
- Monday
Two weekend cancellations, one rate plan
Both were booked months ago on a free-cancellation rate; both cancel just inside the window. Nobody saw it coming, though the same pattern is sitting in last year's data.
- Tuesday
The same question, fourteen times
Is the road to the canyon overlook paved? Is the hot spring open in November? The desk answers by phone, by email and through the booking channel's message box.
- Wednesday
A guest wants a three-day plan
Your best staff member writes one by hand on lodge notepaper, from memory. It's excellent. It leaves with the guest and is never seen again.
- Thursday
November looks soft. Again.
The revenue meeting stares at a booking curve running behind last year. Someone suggests a discount; someone else points out the discount didn't work last year either.
- Saturday
A storm closes the canyon trail
Eleven guests had planned the hike. That's eleven conversations at the desk, each starting from scratch.
- Sunday
An assistant gets you wrong
A prospective guest says an AI assistant told her you don't allow dogs. You do. She's already booked somewhere else.
What changes with Nova
The same team, the same rooms. Fewer surprises, and fewer hours spent answering what a well-informed system could answer.
Packages built from real behavior
When Signals spots a soft stretch, Itinerary Studio assembles packages around what past guests in that season actually booked: stargazing, a pottery workshop, a cool-morning guided walk.
A concierge that knows today
Concierge answers guests on web chat, SMS or WhatsApp from your own property facts plus live weather and road data, and passes anything sensitive to the desk.
Represented correctly by AI assistants
Accurate, structured property facts and policies, so assistants stop guessing about dogs and spas — and, through Agent Gateway, can book you directly.
Recommended modules and services
Single properties usually start with Signals or Concierge. Collections tend to start with the data work, because every property has recorded things slightly differently.
- ModuleSignalsCancellation risk, demand forecasts and price-sensitivity estimates — suggestions, never automatic rate changes.
- ModuleConciergeGuest questions answered in their own channel, grounded in your facts, with hand-off to staff.
- ModuleItinerary StudioPackages and stay plans built from local partners and what guests actually do.
- ServiceAI Search Visibility for TravelMake your property's facts, policies and character citable when travelers ask an assistant where to stay.
- ServiceAgentic Commerce ReadinessRates, policies and checkout prepared so AI agents can book rooms and packages safely, on your terms.
- ServiceTravel Data Engineering & IntegrationYour PMS, channel manager and CRM joined into one reliable history the forecasts can learn from.
I used to find out a weekend was in trouble on the Thursday before. Now I find out three weeks ahead, with a short list of who to call.
How it works
A sample workflow: a weekend that's starting to wobble
- 01
Connect the booking history
Weeks 1–3Two or more seasons of reservations from your PMS and channel manager, joined with rate plans, channels and lead times.
- 02
Every reservation gets a risk score
Updated daily, and explained in plain terms: long lead time, flexible rate, a pattern seen in previous seasons.
- 03
A wobbly weekend surfaces
Three weeks outThree weeks out, Signals flags that a holiday weekend has several high-risk bookings on the same flexible rate.
- 04
Actions drafted, not taken
Friendly reconfirmation messages for the riskiest bookings; a waitlist offer prepared for past guests who asked about those dates. Your team approves, edits or ignores each one.
- 05
Count what actually happened
Cancellations that occurred, rooms refilled and messages that changed nothing are all tracked, so the next forecast is better than this one.
Field report: a collection of desert lodges
A small group of high-desert lodges asked us two questions: which reservations are likely to cancel, and what would bring guests in during the shoulder months without simply cutting rates. The case study walks through the data we joined, how the team reviewed predictions before acting on them, and how the first shoulder-season packages were built.
Questions
Frequently asked questions
Do we have to change our PMS or channel manager?
No. Nova reads from them — through APIs where they exist, exports where they don't — and writes back only what you've approved. See Integrations.
Will Signals change our rates automatically?
No. It estimates risk, demand and price sensitivity and suggests actions. Your team sets prices; rate changes stay a human decision.
How much booking history do we need?
Two seasons is a comfortable start. With less, we lean more on market and calendar signals, show wider forecast ranges, and say so plainly.
Can Concierge handle what the front desk normally does?
The factual parts — hours, directions, what's open in the rain. Anything involving money, complaints or a guest's wellbeing goes to a person, with the conversation attached.
We're a single property. Is this for us?
Yes. Many independent properties start with Signals alone, or with Concierge for the questions your desk answers fourteen times a week. For quiet months specifically, see shoulder-season demand shaping.
Related
Keep exploring
- PlatformSignalsDemand forecasts, cancellation risk, price sensitivity and shoulder-season shaping, each with a confidence range, its drivers and a next step a person approves.
- PlatformConciergeAn in-trip assistant on web chat, WhatsApp, SMS and email that answers from your live data, re-plans when the day breaks, and hands off to your staff in time.
- Use casesShoulder-Season Demand ShapingShoulder-season demand shaping: predictive signals spot soft weeks early and suggest packages worth building, long before a discount looks like the only option.
- WorkDesert lodge collectionHow a small collection of desert lodges flagged 7 in 10 late cancellations two weeks early and filled quiet shoulder-season midweeks with packages.
- Field Notes · Data & AnalyticsPredicting Cancellations Before They HappenA practical walk-through of cancellation prediction for lodges and tour operators: defining the label, choosing features, calibrating scores and acting on them.
Bring us last year's quietest month
We'll look at your booking curve together and show you what Signals would have flagged, and when.