The agent reads each client's accounting data, together with known commitments such as upcoming bills, payroll runs and expected receipts, and keeps a rolling 90-day cash view up to date. When you ask a what-if question, it reruns the picture with the new assumption and shows what changes.
The month-end commentary is a draft, written in plain language for your accountant or CFO to edit. It states what moved and why, using only the data it can see. Typical what-if questions are practical ones: can we hire in March, what if a big customer pays thirty days late, what does a new loan do to the buffer? The agent shows the result against the buffer your client has chosen, and lists the assumptions it used so your adviser can challenge them.
A forecast is only as good as the conversation it starts. The agent prepares, and the adviser decides what to tell the client. It also helps to agree, early, how often each client should receive a pack. Your clients will notice the consistency, and so will you.
The forecast is a projection built from your inputs, not a promise of what will happen, and it is not financial advice. Clients with messy or late books will get weaker forecasts until the bookkeeping underneath is tidy, so it often pairs with the bookkeeping agent. To measure it, compare forecast to actual cash each month, note which assumptions drifted, and count how many clients now get commentary that previously got none.
This work is specific to your field, so we build it around your rules, templates and examples instead of a generic model.