Ask a room whether a KPI at 58% of target in week eight is on track and you will get opinions. Ask whether a KPI that needs 42 points in five weeks, from a team that has produced 7 points a week all quarter, is on track, and the room goes quiet. Same KPI. The second framing contains the answer, and it took one division to get there.
Pace to plan is that framing, applied routinely. It compares the rate a KPI must move to reach its target with the rate it is actually moving. It is the cheapest piece of execution analytics that exists, it catches most misses a month or more before they land, and almost nobody automates it.
The math
Four inputs: the target T, the current value V, time elapsed, time remaining. Two rates fall out.
Required run-rate: what has to happen from here. (T minus V) divided by periods remaining.
Actual run-rate: what has been happening. Progress over the last few periods, divided by those periods. Use a trailing window (the last three or four weeks) rather than the whole quarter, because the recent rate is the honest one; the early rate is contaminated by whatever was easy.
The pace ratio is actual divided by required. Above 1.0, the KPI reaches the target at current pace. Around 0.9, it is close enough that normal variance decides it. Below about 0.8, the KPI does not hit without something changing, and the something needs to be real: more capacity, cut scope, a different approach. Below 0.8, hope is not a plan input.
A worked example
Say the KPI is qualified pipeline: target 2.4M for the quarter, and the quarter is 13 weeks.
End of week 8: pipeline stands at 1.35M. The dashboard shows 56% of target with 62% of the quarter gone. That reads as "slightly behind", the kind of gap a status meeting rounds to yellow-but-fine.
Now the rates. Required: 2.4M minus 1.35M is 1.05M across 5 remaining weeks, so 210k per week. Actual, trailing four weeks: pipeline grew from 1.05M to 1.35M, so 300k over 4 weeks, or 75k per week. Pace ratio: 75 over 210, which is 0.36.
That KPI is not slightly behind. At current pace it finishes near 1.72M, about 72% of target. Reaching 2.4M requires the team to produce pipeline at nearly three times its demonstrated recent rate, in the last five weeks of a quarter. Anyone who has run a pipeline knows what week 13 adds: not triple velocity. The dashboard said 56%, yellow-but-fine. The arithmetic says the miss is already here, five weeks early, while there is still time to change the offer, add a channel, or reset the number honestly.
The gap between those two readings is the entire value of the method. Nothing was forecast. No model was built. Division was applied to numbers everyone already had.
Why linear extrapolation is enough
The obvious objection is that progress is not linear. Deals close at quarter end; adoption compounds; some metrics move in steps. All true, and mostly not a reason to skip the math.
If a KPI has a known seasonal shape, encode it: replace "periods remaining" with an expected progress curve (by week 8 we expect 55% of the quarter's pipeline) and compare position against the curve instead of the straight line. That is one refinement, not a modeling project. But be suspicious of curve arguments that only ever appear when the straight line looks bad. "It'll hockey-stick" is occasionally a fact about the metric. More often it is a fact about the meeting. If the hockey stick is real, it happened last quarter too, and the curve can be drawn from data rather than from optimism.
Linear extrapolation is not a forecast. It is a burden-of-proof device. It establishes what happens if the recent past continues, so that anyone claiming a different future has to name the thing that will be different. That conversation, held in week 8, is execution risk management working exactly as it should: the gap between goal and effort made visible while it is still cheap to close. Held in week 13, the same conversation is a post-mortem; the green-dashboard version of it is rarely held at all.
Making it routine
Running this once, in a crisis, is diagnosis. The value is in running it every week, on every KPI that matters, which is where manual practice dies: someone has to pull four numbers per KPI, maintain the trailing windows, remember the seasonal shapes, and chase teams whose metrics live in five different tools. It survives about six weeks as a spreadsheet ritual.
The durable version is structural, and it rests on the same mapping discipline that underpins strategy execution in general. Map each KPI to its target and deadline once, keep current values flowing in from the systems that produce them, and let the comparison run continuously. In Vindaris this is what the Work Graph does with every measured goal: it holds target, pace and the work underneath the KPI in one place, and flags the owner when the pace ratio says the target is no longer reachable at the current rate, with the arithmetic attached. The flag arrives in week 8, phrased as a number rather than an accusation, and week 8 is early enough to matter.
A KPI that is moving is not the same as a KPI that is arriving. The difference is one division. Run it before the quarter does.