Field notes

When will this customer run out of commit? Three ways to forecast burn-down

Six months in, a customer has used $52,000 of a $120,000 annual commit. Depending on how you project it, they finish the year $16,000 short, $4,000 short, or out of credit by about 12 November.

TrueUpp Research · Published 19 October 2026 · 5 min read

About these field notes. They draw on public discussions with RevOps, billing and finance practitioners in October 2026 about how their teams run usage true-ups. We paraphrase and don't name anyone. The worked example is an illustration we built to show the mechanics, not any company's real figures.

Why pacing is a finance question too

In one public discussion, a practitioner running prepaid commits on Stripe described a small service their team built: every billing cycle it checks each customer's usage against commit pace, and switches to the right overage tier when someone is trending over. Their verdict: not pretty, but it works. The reason to build it is that pacing decides three things that cost money when they're late:

• whether customer success should start a shortfall conversation, or sales should offer a bigger commit before overage starts;
• whether rolled-over credits will expire unused, which the customer will remember at renewal;
• what finance expects to bill as overage or shortfall at year end.

Three projections, three answers

Straight line divides $52,000 by six months and carries that average forward. It treats January's $6,000 as just as informative as June's $12,500, so for a customer who is still ramping up it is pessimistic.

Trailing three months uses April to June only: $32,000 ÷ 3. It reacts faster and ignores the slow start, but it still assumes the customer stops growing today.

Linear trend fits a straight line through the six monthly figures. The slope is about $1,314 a month, so July comes out at roughly $13,300 and December at roughly $19,800. On that path cumulative usage reaches $112,952 by the end of October, and the remaining $7,048 runs out about 38% of the way through November.

One customer, one set of numbers: a $16,000 shortfall, a $4,000 shortfall, or an upsell conversation that needs to happen before November.

One number worth adding: the required run rate

The required run rate is what the customer must use each remaining month to land exactly on the commit: ($120,000 − $52,000) ÷ 6 = $11,333 a month. Put it next to the recent figures. June was $12,500, above it. The three-month average is $10,667, below it. This customer is on the edge, and the next two months decide which way it goes. A percentage-consumed alert at 50%, 75% or 90% would have said nothing useful here.

Which forecast to use for what

• Shortfall warnings: the trailing three-month rate compared with the required run rate. For a customer whose usage is flat or rising, it errs toward warning early. If usage is falling, check the latest month too.
• Exhaustion dates and upsell timing: the trend, but only with at least six months of history, and with a sanity cap on growth. A customer that doubled once doesn't double every quarter.
• Anything shown to the customer or the account team: the range across all three methods. A wide range means usage is changing, so call that account.

Details that move the forecast

• Rolled-over credits expire on their own dates. A rolled-over lot that expires on 31 March must be used by then, whatever the annual pace says. Forecast each lot against usage up to its own expiry, earliest first. A single "remaining balance" can't show this. See rollover caps and expiring lots.
• Amendments change the target. After an upgrade, the required run rate is measured against the new segment's commit and end date, from the amendment date. See amendments as new segments.
• Rate changes distort dollar forecasts. If the price per unit changed mid-term, forecast in units and convert at the current rate.
• Late usage makes the latest month look low. Forecast on usage by the date it happened, and expect the most recent month to fill in. See what counts as usage.
• Mid-month checks need days, not months. On 15 June, compare consumption with 166 of 365 days elapsed, not with "five and a half months".

Turning the forecast into alerts

1. Shortfall risk: the trailing three-month rate sits below the required run rate for two months in a row. Tell the account owner and finance.
2. Exhaustion within 60 days: the trend has the commit running out within 60 days. Have the expansion conversation, and confirm the overage rate the customer is about to see.
3. Credits about to expire unused: a rolled-over lot is projected to lapse. Telling the customer first costs nothing and removes a renewal argument.
4. A forecast that jumps: the projection moves by more than about 20% in a month. Check the usage data for duplicates or gaps before acting on it.

The true-up calculator works out one contract's balance, rollover and overage from monthly usage, which is the base any forecast starts from.