Velocity to date
Spend in the category divided by days elapsed, carried forward to the end of the cycle. Deliberately simple, and for most categories the single most honest indicator you have on day 12.
Tracking tells you where a category stands. Forecasting tells you where it ends up. The engine takes spend velocity to date, adds the obligations it already knows about, and projects the close of the cycle, so a category can be flagged as trending over while it is still comfortably inside its limit.
A forecast that cannot be reconstructed by hand is a forecast nobody will act on. These are the only three things feeding it.
Spend in the category divided by days elapsed, carried forward to the end of the cycle. Deliberately simple, and for most categories the single most honest indicator you have on day 12.
Recurring charges with a billing date still ahead, pending authorizations, and scheduled obligations inside the cycle. These are added at face value rather than smoothed into a daily average.
Some categories are back-loaded by nature. Where you have enough history, the projection is weighted by how that category has historically distributed across a cycle rather than assuming a flat line.
Knowing Travel will finish at 187 percent is useful. Knowing it crosses its limit on day 16 is what gets the decision made, because there are fourteen days left to make it in.
| Category | Current pace | Projected close | Limit | Lands at | Crosses limit |
|---|---|---|---|---|---|
| Marketing | 5,200 / day | 1,56,000 | 1,20,000 | 130% | Day 23 |
| Payroll | Fixed | 4,80,000 | 4,80,000 | 100% | Day 30 |
| Software | 1,920 / day | 57,600 | 85,000 | 68% | Not projected |
| Travel | 9,333 / day | 2,80,000 | 1,50,000 | 187% | Day 16 |
| Supplies | 1,220 / day | 36,600 | 60,000 | 61% | Not projected |
A threshold flag waits for the category to actually reach its near-limit percentage. A pace flag fires the moment the projection crosses the limit, which is usually days or weeks earlier. Both are available, and you decide which categories warrant the earlier warning.
A reallocation source is only genuinely free if it is also projected to finish under. The engine will not propose moving room out of a category that currently reads 68 percent but is on pace to close at 104, because that trades one overage for two.
Budget is a plan. Cash is a constraint. The same velocity and commitment data rolls up into a near-term view of what is scheduled to leave the connected accounts over the coming weeks, so a month that balances on paper does not catch you short in the third week.
This is a projection of outflow from spend data. It is not a treasury forecast and it does not model your receivables or financing.
A projection is arithmetic on the spend you have already made plus the obligations you already carry. It does not know about the campaign you are about to launch or the trip you just cancelled.
So treat it as what it is: an early reading that buys you time to decide, not a prediction that decides for you.
Where the decision actually happensLive spend against every category limit, with status the moment it changes.
OpenTransactions sorted into your categories as they settle, with override control.
OpenRoute spend and reallocation decisions to the person who owns the budget.
OpenVariance to budget by category, period over period, export ready.
OpenOne of them still lets you do something about it.
The category closed at 112 percent. The money left the account three weeks ago. The only decision available now is which line to explain it against.
At 74 percent on day 11 the category is flagged as trending over, with an under-spent line named as the source. You approve the shift, or you tighten the spend. Either way you chose.