Campaign budget pacing: planned vs actual, explained

What is budget pacing?

Pacing answers one question continuously: is this campaign spending and delivering in line with the plan? Not at the end of the month, when the reconciliation produces a number nobody can act on, but today, while a drifting line can still be corrected. A campaign that finishes 15% underspent did not fail in week four. It failed quietly across weeks one to three, one unnoticed day at a time.

The calculation, and the mistake inside it

Pacing % = actual spend to date ÷ planned spend to date. A $310K flight that should have consumed $150K by day 15 and has spent $120K is pacing at 80% - under-pacing by a fifth.

The mistake lives in 'planned spend to date'. Divide the flight budget by days and you get a linear target, which is only correct if the plan was flat. If the flighting front-loads a launch week or builds toward a retail moment, the pacing target must follow that shape, or the alerts will fire on drift that is actually the plan working. Weighted pacing targets - planned-to-date built from the flighting curve - are the difference between alerts you trust and alerts you mute.

Why spreadsheet pacing fails

Not because the formulas are hard. Because the inputs are manual. Spreadsheet pacing depends on someone exporting spend from each platform and pasting it into the tracker, which happens when someone remembers, which is weekly at best and 'before the client call' in practice. The data is stale on arrival, the ritual consumes hours, and one mispasted column poisons the sheet. The structural fix is pulling spend, impressions and conversions from the platforms automatically - which is what connected pacing does, refreshing actual versus planned daily across Google, Meta, TikTok and LinkedIn without an export in sight.

Reading drift: direction is not diagnosis

The pacing number tells you a line has diverged. It does not tell you why, and the why decides the fix. Under-pacing might be an audience exhausted, a bid too conservative, a creative rejected in review, or inventory that simply is not there at your price. Over-pacing might be efficient scale worth keeping - or a broad-match term eating budget on irrelevant queries. Ahead of plan with CPA holding is an opportunity. Ahead of plan with CPA climbing is a leak. The spend column cannot tell those apart; the delivery and outcome metrics beside it can.

Fixing under-pacing and over-pacing

The reallocation cases are where pre-agreed rules earn their keep. A pacing alert that arrives with a suggested, quantified fix - move this amount from here to there to recover by end of month - turns a firefight into a decision. That is the design logic behind drift alerts with one-click reallocation suggestions, and the wider decision framework is covered in in-flight media optimisation.

Thresholds and cadence that actually work

• Threshold: flag drift beyond plus or minus 10% of the weighted planned-to-date, sustained for three consecutive days. Single-day wobbles are noise; three days is a trend.
• Short flights: tighten to 5% and check daily. A two-week burst has no recovery runway.
• Cadence: daily reads for large budgets and launch periods, twice-weekly as the floor for everything else. Monthly pacing is not pacing, it is accounting.
• Ownership: every alert needs a named owner and a response window. An alert nobody owns is a notification.

The end-of-flight catch-up, and why it costs double

The classic failure: under-pacing goes unnoticed until week three of four, then the remaining budget gets forced through a compressed window. Spend concentrates, auction pressure rises, frequency spikes, and the marginal cost of every unit climbs precisely when there is no time left to react. The campaign hits its spend number and misses its outcome number, and the post-mortem calls it a delivery problem when it was a detection problem. Catching the drift in week one is not a nicety. It is the difference between reallocating on your terms and buying badly on the platform's.