You have a forecast, but nobody believes it
I rebuild the forecast so everyone quotes the same number: a rolling 13-week cash view, updated every week, that lets you make the right call in time.
What it looks like
- Two people quote two different numbers for the same position.
- The forecast is updated when someone has time, so nobody knows how old it is.
- Vendors are paid late because nobody can say when customers will pay.
- A bank deadline is coming and the cash picture is a guess.
Who it is for
Owners, CEOs and boards of growing companies that have a forecast nobody trusts. A company with no forecast at all fits too, and starts from a cleaner page.
What I do
- Find where the number goes wrong. I sort the drivers into three groups: the P&L (when revenue and cost land), balance sheet items (when customers actually pay and vendors are actually paid) and tax and financing (VAT, loan repayments, interest).
- Build it up from the P&L to real cash. Revenue, cost of goods as a share of revenue, operating costs as a steadier line and financial items give a monthly result. I then adjust it for actual payment timing, VAT included, to get cash.
- Automate it into a rolling 13 weeks. The weekly refresh runs from the bank and the ledger, with a short note on what moved. A late customer, a new hire or an investment becomes a number the board can discuss.
- Lock each week and measure it. Each week's forecast is locked once it is made, and its accuracy is measured afterwards, so you can see how far ahead it can be trusted.
How it works
I build the first version with you, in the tools you already use. Your team then runs the weekly update and I check it, until the forecast no longer needs me.
What I am seeing
A company I am working with has cash close to zero, a bank that has set a quarterly deadline and vendors growing impatient. I am helping it move toward the same live weekly forecast I run on Alfen itself. The work is in progress, so there is no outcome to report yet.
In another company, cash visibility was used to time a sales hire against its negative payback period. The forecast was a tool for growth timing, not only for avoiding a crisis.
Questions I often get
What is a rolling forecast?
A forecast that always looks the same number of months ahead. Each month you drop the month that has passed and add a new one at the end, so the view never runs out the way an annual budget does.
How far ahead should a company look at cash?
Thirteen weeks is the usual horizon for a weekly cash view. It is long enough to see a gap coming and short enough to stay accurate. Longer horizons belong in the rolling forecast.
Why is a forecast from 13 weeks ago wrong by now?
Because it is adjusted every week, and that is how it is meant to work. The value is the visibility you have each week, not one static prediction holding for a quarter.
Proof point
At Billetto, the team built a daily reforecast dashboard, so the forecast was refreshed every day.
The result is not a perfect prediction. It is enough sight ahead, every week, to make the right call in time.