Most small houses track one number, occupancy, and feel good or bad about it. Occupancy on its own is close to useless: a house that is 90 percent full at a price that does not cover the season is in worse shape than one at 65 percent that keeps its margin.
Five numbers cover almost everything a house of five to twenty-five rooms needs to decide. None of them require an accountant, and all of them can be read once a month in ten minutes.
1. Revenue per available bed night
Take all accommodation revenue for the month and divide it by the number of bed nights you had to sell, whether or not you sold them. Twelve beds times thirty days is 360 bed nights; IDR 190 million of room revenue gives you about IDR 528,000 per available bed night.
This is the number that ends the occupancy argument, because it moves when either occupancy or price moves. Compare it against the same month last year and nothing else. Comparing August to October tells you that August is busier, which you knew.
2. The bar and services share
What percentage of total revenue comes from things that are not the bed: drinks, food, lessons, rentals, trips, transfers. In houses that run a real bar this is often a third or more, and it is usually the part with the best margin and the worst bookkeeping.
The reason to watch it monthly is that it falls silently. A month where the share drops by ten points almost never means guests drank less. It means somebody stopped putting things on tabs and started writing on paper, and the difference is money you never collected.
3. Direct share
The percentage of bookings, or better, of room nights, that came without a portal in the middle. Everything about the economics of a small house improves as this number rises: no commission, deposits earlier, the guest's real contact details, and a relationship you can use next season.
Watch it as a trend, not as a target. A house going from 30 to 45 percent over two seasons has built something durable. A house that hits 80 percent in one month has probably just had a quiet month in which only regulars came, which is a different story entirely. The levers that move it are all slow ones.
4. Repeat and referral rate
What share of this month's guests had stayed before, or came because someone who stayed before sent them? Most systems will give you the first half if guests are stored as people rather than as bookings. The second half you have to ask for at check-in, in one sentence: how did you hear about us.
This is the closest thing a surf house has to a quality score, and it is far more honest than a review average. Reviews measure the guests who wrote. Returns measure the guests who paid again.
5. Average length of stay
Total nights divided by number of bookings. It looks like a curiosity and it is actually a workload number. Every departure costs a cleaning round, a checkout, a bill and an arrival, so a house averaging three nights runs roughly twice the operational load of a house averaging six at the same occupancy.
If this number is falling while occupancy holds, your staff are getting busier for the same revenue, and the fix is usually a minimum stay in the busy weeks rather than more hands.
The two numbers underneath
Those five describe the business. Two more describe whether it is under control: the daily cash difference, expected against counted, which should be small and explained rather than zero and mysterious, and the open balance at any moment, meaning what guests currently in the house still owe you. A house that cannot answer the second question in ten seconds is carrying an amount of money it has never seen.
How to actually read them
Once a month, same day, same order, next to the same month last year. Not weekly, because weekly numbers in a twelve-room house are noise: two cancellations move everything, and you will start making decisions about weather.
And read them together. Revenue per bed night up, repeat rate down, means you raised prices and lost your regulars, which is a slow problem that looks like a good month. Occupancy up, services share down, means you are full of people who are not spending. Any single number can be improved by doing something stupid. It is the combination that tells you whether the season is working.