How often should a forecast final account change?

The right rhythm is a monthly review, and the right amount of movement is exactly as much as the account genuinely justifies. A forecast final account is your running estimate of where the account will finally land: the contract sum, plus variations both agreed and pending priced at what you honestly expect to recover, plus claim exposure carried openly, less contra risk, with retention timed to when it truly releases. Each month you review it against what actually happened, and move a number only when a reason moves with it: a variation agreed at a figure, a contra accepted, a claim conceded or firmed up, a provisional sum expended, a rate proven on site. That reason keeps the forecast trustworthy, because a figure that changes without an explanation is one nobody believes and everybody quietly reworks in their own head. Two failure modes give it away. If it never moves, it is being copied forward rather than reviewed, and will lurch violently the month reality can no longer be ignored. If it drifts every month with no reason attached, it has become guesswork dressed as a report. And if it keeps landing lower than forecast, the cause almost always lives upstream: variations going unpriced, applications going in thin, changes not captured as they happen. A good forecast says so and points at the leak, rather than absorbing the error silently and looking accurate while the real account bleeds elsewhere.

Take a drylining subcontractor whose commercial director maintains a forecast final account that, for four straight months, reads exactly the same as the contract sum. It looks stable and is fiction: eleven variations instructed and none priced in, a stack of dayworks unassessed. In month five, when the variations can no longer be ignored, the forecast jumps by a large sum overnight, and because it moved all at once with no running history, the client's surveyor treats the increase with suspicion and the negotiation starts badly. On the next job it is maintained honestly: moving most months, but only a little, each movement carrying its reason, a variation agreed here, a contra conceded there, a claim marked up as evidence firms. When it keeps drifting downward, the note beside it says plainly that variations are being instructed faster than they are priced and applied for, sending the fix upstream rather than leaving it to surface as a final-account shock.