Start from the contracted schedule
Most of next quarter's billings are already determined. Pull every active contract and lay out its invoice dates and amounts. Annual prepay contracts invoice on their anniversary. Quarterly-in-advance contracts invoice on a fixed cycle. Monthly subscriptions invoice continuously. Multi-year deals invoice on whatever payment schedule was negotiated. That base schedule is a calculation rather than a prediction, and at a company with a mature installed base it covers the majority of the period.
Then subtract what will not invoice. Contracts flagged as non-renewing, accounts in collections dispute, and customers on a negotiated pause all sit in the schedule and will produce no cash.
Layer the new bookings that bill in-period
New business becomes billings only when it closes early enough for its first invoice to fall inside the period. A deal that closes on the last day of the quarter with net-30 annual prepay terms produces billings next quarter.
So the new-business layer needs two inputs: the sales forecast broken out by expected close date, and the billing terms attached to each opportunity. Deals on annual prepay convert to billings at full ACV. Deals on monthly terms convert at one twelfth. Deals with a delayed start date convert whenever the start date says they do.
Slippage moves billings harder than it moves bookings
A deal that slips two weeks inside a quarter barely changes the bookings number. The same slip can move an entire annual prepay invoice into the following period. That makes deal slippage signals the highest-value input in a billings model.
ORM treats a rep changing the close date as the single best indicator of slippage, and a deal that moves from one quarter into the next is less likely to close even when it sits in commit. Feed those flags into the billings schedule directly rather than assuming the CRM close date is accurate. The quieter signal matters too: an opportunity with no stage change, no amount change, and no close date change is not progressing toward an invoice.
Run all three forecasts off one data set
When billings, bookings, and recognized revenue are modeled independently, they disagree, and the disagreement surfaces in the board meeting instead of in the model. Building all three from the same opportunity and contract records means a single close date change updates each of them at once. That is the practical difference between a revenue forecast that holds through the quarter and one that gets rebuilt every month.
Frequently Asked Questions
What is the difference between a billings forecast and a bookings forecast?
Bookings record contract value on the signature date. Billings record invoices on the dates the contract specifies. A deal booked in March on quarterly-in-advance terms produces one bookings event and four separate billings events across the year.
How much of a billings forecast is already determined?
At a company with a mature installed base, most of it. Active contracts carry fixed invoice dates and amounts, so the base schedule is a calculation. The forecasting work sits in new bookings that bill in-period and in contracts that will not invoice.
Why do billings and recognized revenue diverge?
Billings follow the payment schedule and recognized revenue follows delivery. An annual prepay contract bills the full year in one invoice and recognizes one twelfth of it each month, and the gap sits on the balance sheet as deferred revenue.
What causes the most error in a billings forecast?
Close date movement. A deal that slips two weeks barely changes bookings, and the same slip can push an entire annual prepay invoice into the next period. Billings forecasts are more sensitive to close date accuracy than bookings forecasts are.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like how do you forecast billings? into prescriptive action for your team.
Schedule a Demo