Your net-30 clients pay at day 40. Your net-7 clients pay at day 12. One client pays the morning you send the invoice and another will never pay without a statutory letter. Every freelance book is a mosaic of payment behaviours, and until today no invoicing tool did the obvious thing with that information — forecast when the money actually arrives.
QuickBooks and Xero will show you aging buckets. FreshBooks will send a reminder. None of them will tell you what's coming in next week with a probability attached. That gap isn't technical — the data has always been there. It's that their customer is the accountant, not the operator. InvoiceUp is built for the operator.
Every invoice you've ever marked paid has two timestamps: the due date, and the paid date. The difference is a single number — days late or days early. Three or more paid invoices from the same client gives us a strong signal for their payment behaviour. One or two gives us a medium signal. Zero falls back to the global average across your book, which gives us a low signal.
For every unpaid invoice, we apply the appropriate signal to project an expected paid date. Aggregated into 14-day, 30-day, and 60-day buckets, that becomes a forecast. The dashboard card now shows how much you can expect, how many invoices that represents, and the confidence level — because a forecast without a confidence level is just a guess in better clothing.
Every late payment is data. Every early payment is data. The forecast is just what happens when you don't throw that data away.
Invoices our model expects to clear more than 60 days from now get a separate red row: "at risk past 60 days." That's the money that should already have a statutory letter attached — and because the EU late-fee integration shipped last week, you can send one with a button in the invoice detail. Forecast first, act second.
It's deterministic — same data in, same forecast out. A simple per-client mean with a confidence tier. This is intentional: a forecast you can't explain is one you don't trust, and one you don't trust is one you don't use. We'd rather ship a simple method that's right most of the time than a complex one that's right in aggregate but nonsensical on any given invoice.
Pair this with the runway calculator and the concentration card, and InvoiceUp is starting to look less like an invoicing app and more like the financial cockpit nobody built for solo businesses. Which is exactly the point.