The Campaign That Was Supposed to Deliver
You committed to 5 million impressions in October. By October 15th, you've delivered 1.8 million. At this pace, you'll hit 3.6 million — 72% of the commitment. The campaign will under-deliver by 1.4 million impressions, and nobody will notice until November.
This is the pacing problem. It's not that under-delivery happens — it's that it's detected too late to act. By the time the campaign ends and the delivery report comes in, the vendor has moved on, the make-good window has closed, and the shortfall is permanent.
What Pacing Actually Means
Pacing is the comparison of expected spend against actual spend at any point during a flight. It answers a simple question: are we on track to deliver what we committed to?
A flight that's 50% through its schedule should have delivered roughly 50% of its committed volume. If it's delivered 30%, there's a pacing gap. If it's delivered 70%, the flight is over-pacing (which has its own issues — inventory might be exhausted early).
Pacing isn't a single check at the end. It's a continuous comparison throughout the flight. The earlier you detect a gap, the more time you have to act.
The Three Pacing States
Every flight is in one of three pacing states:
On pace. Actual delivery is within an acceptable range of expected delivery. No action needed.
Under-pacing. Actual delivery is below expected. Action needed: contact the vendor to increase delivery, or document the shortfall for a make-good claim.
Over-pacing. Actual delivery is above expected. Action needed: contact the vendor to slow delivery, or accept that the inventory will be exhausted before the flight ends.
Why Pacing Checks Fail
Most pacing checks fail for one of two reasons:
1. No baseline. You can't compare actual delivery to expected delivery if you don't have a committed volume. If the IO says "up to 5 million impressions" instead of "5 million impressions," there's no baseline to pace against.
2. No frequency. A pacing check at the end of the campaign is not pacing — it's a post-mortem. Pacing needs to happen regularly, ideally daily or at least weekly, throughout the flight.
Building a Pacing Process That Works
A pacing process that works has three components:
Committed volume. Every IO line has a guaranteed volume — the number the vendor committed to deliver. This is the baseline.
Actual delivery. Delivery data is imported or recorded regularly — daily for digital, weekly for traditional media. This is the comparison.
Pacing snapshots. At regular intervals, the system compares actual delivery to expected delivery and flags flights that are off-pace. This is the alert.
When a pacing alert fires, someone acts. They contact the vendor, they adjust the buy, or they document the shortfall. The point isn't the alert — it's the action that follows.
The Cost of Late Detection
The difference between detecting under-delivery on day 15 and detecting it on day 30 is the difference between a fixable problem and a permanent loss. On day 15, the vendor can increase delivery. On day 30, the flight is over and the shortfall is locked in.
Pacing checks aren't about catching vendors who under-deliver. They're about catching under-delivery while there's still time to do something about it.