It is a Saturday night in December, and a family of four reaches your desk with a MakeMyTrip confirmation. Your last family room went to a Booking.com guest at six. Now someone has to ring the hotel down the road, pay for a room there, find a taxi and say sorry.
This post shows why OTAs reward hotels that keep every promise, and how to keep your availability right on every channel, a year ahead.
An OTA earns only when the guest stays
An online travel agency makes its money from commission on bookings that turn into stays. A booking the hotel can't honour earns it nothing, and it may cost it a customer who books through another app next time. So every OTA has a reason to show first the hotels that turn lookers into bookers and keep their word.
No OTA publishes its full formula, so be careful with anyone who promises you a position. What is widely known is the kind of signal that matters.
| Signal | What the OTA sees | What you control |
|---|---|---|
| Availability | Whether your rooms can be booked for the dates searched | Keep a year open; close dates only on purpose |
| Conversion | How many people who open your listing go on to book | Photos, room names, clear policies, a fair price |
| Reviews | Your score and how recent your reviews are | Service, and asking every guest to review |
| Cancellations by the hotel | Bookings you cancelled or could not honour | One count of rooms that every channel sells from |
| Rate competitiveness | Whether your price looks fair next to other sites | Steady rates from one base rate, sent everywhere at once |
Two of those rows, availability and cancellations, are about reliability, and reliability is the quickest to fix. Reviews take months to move, but your availability can be right by Friday. For the rate row, read why your website must never cost more. For reviews, read how your rating decides your price.
A room sold twice costs more than an empty one
Walking a guest means sending someone with a confirmed booking to another hotel because you have no room. You pay for their night elsewhere, and usually for the taxi too. Then you spend the evening apologising, and the review arrives a few days later.
Worked example
One family room, sold twice
In this example, the family room sells for ₹4,000 a night. Two OTAs sell it for the same Saturday, and the second family has to be walked.
| Line | Amount |
|---|---|
| Paid by the guest who got the room | +₹4,000 |
| A room at the hotel down the road, booked at 9 pm | −₹5,500 |
| A taxi for the family and their bags | −₹500 |
| What that room earned you that night | −₹2,000 |
Left empty, the room would have earned nothing. Sold twice, it cost you ₹2,000, and that is before the review.
A room sold twice costs you more than a room left empty.
The OTA has its own view of that night. If the family writes in, the OTA learns you took a booking you could not keep. Some hotels make it worse by cancelling the booking themselves to tidy things up. OTAs keep track of cancellations made by the hotel, and they don't read them like a guest changing plans.
Double-selling happens in the gap between systems
Rooms rarely get sold twice through one big mistake. They get sold twice, or not at all, through small gaps. A walk-in goes into the register but not the extranet. A room is blocked for a broken geyser on the board but still shows free online. A guest checks out two days early and the room stays closed. Someone types a number into one extranet and forgets the other three.
Each of those is a count that changed in one place and not in the others. The fix is to keep one count and send it everywhere.
ZimmerStack works this way: every booking, cancellation, early check-out and room block changes one count. That count goes out to every channel you sell on, from one channel manager connected to 170+ channels. A push that fails is retried, not dropped. If pushes keep failing, the desk sees an alert and the owner gets an email, so a stale count doesn't sit unnoticed.
No system can stop an OTA from sending a booking in the moment before the new count arrives. When that happens, ZimmerStack still accepts the booking, marks it oversold and alerts the desk. You find out at noon, while there is time to move a guest or upgrade one, not at nine at night.
Keep a full year of dates on sale
Guests book weddings, festival trips and winter holidays months ahead. A date that isn't loaded on a channel can't be sold there, however good your listing is. A hotel that loads rooms three months at a time loses those early bookings without ever seeing them.
ZimmerStack keeps 365 nights of availability on the channel manager, adding the next night every evening. It sends the whole year again every Sunday, and a nightly check compares what the channel manager holds with your own numbers. If channel managers are new to you, start with what a channel manager does.
Close dates on purpose, never by accident
Stop-sells, closed-to-arrival and minimum stays are useful tools when you use them on purpose. Close a date for a buyout, a renovation or a wedding block, or set a two-night minimum over a festival weekend. The trouble starts when a closure outlives its reason and nobody remembers making it.
- You can say in one line why this date is closed.
- You know the day it should open again.
- Only the room type that needs closing is closed, not the whole hotel.
- The closure was made in your own system, not typed into one extranet.
- Someone besides you knows about it.
In the ZimmerStack PMS, the availability sheet asks for a reason whenever you set a count by hand. You also choose how long it holds: 7, 30 or 90 days, or until someone clears it. The grid marks each hand-set number with who set it, why and until when. When it ends, that room type goes back to the live count.
See one inventory at workSee how one inventory feeds 170+ channels, and what happens the moment a booking lands.See the channel managerWhat to do this week
- Search your own hotel. Look yourself up as a guest on your two biggest OTAs for three dates: this Saturday, three months out and ten months out.
- Compare tonight's count. Put your own count for each room type next to what each OTA shows. Any gap is a leak.
- Audit every closure. List every closed date and hand-set count, then give each one a reason and an end date, or remove it.
- Read your cancellations. Pull the last 90 days and mark the ones the hotel made. In ZimmerStack, the Cancellations & wash report shows what fell off the books and why.
- Make one rule. Nobody changes availability on an extranet directly, because every change goes through your own system.