Your Customers Don't Cancel Because of the Weather. They Cancel Because of the Forecast.

Saturday turned out fine. A bit of cloud in the morning, sunshine by eleven, nobody needed a coat.

It was still dead.

Look back at what the forecast said on the Wednesday and there it is: heavy rain, most of the afternoon, seventy percent. That forecast was wrong. It did not matter in the slightest, because by the time Saturday arrived and the sun came out, the decisions had already been made.

##
The decision happens days before the day

This is the bit that gets missed. Weather affects your takings, obviously. But it does not affect them on the day it happens. It affects them on the day people looked it up.

Anything people book ahead for, and anything they make a plan around, gets decided two to four days out. That is when someone glances at their phone, sees a row of rain icons across the weekend, and quietly decides not to bother. They do not ring you to say why. They either cancel with a vague reason or they simply never made the booking in the first place, which is worse, because there is no trace of them at all.

So the number that actually moved your Saturday was not Saturday's weather. It was Wednesday's forecast for Saturday. Two completely different things, and only one of them gets remembered afterwards.

##
Some businesses are weather led and some are forecast led

Worth working out which one you are, because the answer changes what you should do.

Walk-in trade responds to the actual weather. A cafe with pavement seating fills up because it is sunny right now. A shop on a high street gets busier because people are out. Those businesses genuinely live and die on the day, and there is not much to be done about it beyond staffing sensibly.

Booked trade responds to the forecast. Anything with a diary, anything planned, anything involving getting several people to agree on a time. A salon appointment, a garden job, an event, a table for eight, a dog groom, a driving lesson. Those decisions are made in advance by people looking at an app, and the weather on the day is almost irrelevant to whether the booking survived.

Plenty of businesses are both, in different proportions, at different times of year. Knowing your split tells you whether to watch the sky or watch the forecast.

##
The forecast that matters is the one they saw

There is a further wrinkle here that is easy to overlook.

Your customers are not looking at a detailed local model. They are looking at whatever is built into their phone, usually in the evening, usually for about four seconds. What registers is the icon and the percentage, not the detail underneath it.

So a day forecast as showery, which any local would read as meaning fine for most of it with a shower at some point, shows up on a phone as a rain cloud. And the rain cloud is what the decision gets made on.

That means the forecast worth watching is the plain one everybody has, not the good one. The question is not what the weather will be, it is what your customers think it will be when they check on Wednesday evening.

##
The asymmetry nobody plans for

A wrong forecast costs you in both directions and neither one is symmetrical.

Forecast rain, actual sunshine, and you lose the day completely. The bookings went, the staffing was cut, and the walk-in trade that might have saved it is spread across the whole town because nobody planned to go anywhere. Nothing about the nice weather gives you that day back.

Forecast sunshine, actual rain, and you get the opposite problem. Everyone still turns up because the plan was made, and you are short staffed, or the outside seating is useless, or the job cannot be done.

Both are planning failures rather than weather failures, and both are avoidable with three days of notice.

##
What to do with those three days

Watch the forecast for the days you have bookings on, not for today. Look at it on the Wednesday for the weekend, because that is when your customers are looking.

When a bad forecast lands on a day you need, get ahead of the cancellations rather than waiting for them. A short message saying the work happens regardless, or that you are inside, or that there is cover, stops the quiet cancellation before it becomes one. Most people cancel because they assume, not because they were told.

Move what can be moved early. A job that genuinely cannot happen in rain is better rescheduled on the Wednesday, while your diary still has options, than abandoned on the Saturday morning.

And treat a wrong forecast as an opportunity. When the forecast is grim and the day turns out fine, the whole town is under-prepared. Everyone cut their staffing. If you did not, there is trade available that nobody is ready to take.

##
Find out how weather-sensitive you actually are

This takes about ten seconds a day and is worth a season of doing.

Write down three things daily: what the forecast said two days earlier, what the weather actually did, and what you took. After a few months you will be able to see whether your trade tracks the forecast or the day, and by how much.

Most owners have a strong opinion about this and very few have ever checked it. The answer is frequently smaller than expected, or concentrated in one part of the week, and either of those changes how much attention it deserves.

Connecting the outside world to what actually happened in your business, early enough to be useful, is what Reputcon is being built to do. The waitlist is on the site.