Skip to main content
Pacing is a weekly check with a monthly consequence; with an agent filling the flat file daily, it’s computed every day for free and surfaced in the daily brief whenever it leaves the on-plan band. A campaign that under-delivers by 8% a week finishes 30% under budget, and nobody notices until the flight is over and the money is gone unspent. This runs against the daily rate recorded on the brief in budget planning. Persist each week’s pace value (or derive it from the daily spend rows), so “holds two weeks” below is checkable rather than remembered.

The one calculation

Pace = spend to date ÷ (daily rate × days elapsed)
  • 1.0 is exactly on plan.
  • Above 1.0 is running hot and will exhaust the budget early.
  • Below 1.0 is underspending and will leave money unspent at the end.
E.g. Day 20 of a 50-day flight at $500 a day. Expected spend is $10,000. Actual is $8,400. Pace is 0.84, meaning the campaign is tracking to finish around $21,000 of a $25,000 budget.

When to act

Not every variance is a problem. Acting on noise is how a campaign gets reset out of its learning phase every week.
Week 1 is not comparable to any other week. Platforms spend aggressively while learning, and most campaigns pace above 1.2 for the first few days. Reacting to that by cutting budgets is the most common self-inflicted pacing wound.

Underspending is a delivery problem first

The instinct is to raise budgets. Check the cause first, because raising the budget on a campaign that can’t spend what it already has does nothing. Every check in the table is readable from the platform (or its API), so an agent can run all five and attach the findings to a pacing alert; the intervention that follows is the owner’s call.

What to change, in order of preference

1

1. Fix delivery

Widen the audience, resolve rejections, consolidate ad sets. This is the only intervention that adds volume rather than moving it.
2

2. Shift between channels

Move budget from a channel underdelivering on cost per result to one that isn’t. Record the shift and the reason, because it will explain a movement in next week’s report.
3

3. Adjust the daily rate

Recalculate against the remaining days, not the original flight. Raising the daily rate late in a flight buys less than the arithmetic suggests, since the platform still needs time to spend it.
4

4. Release the reserve

Only if the release condition set in budget planning is met. A reserve released because a campaign is underspending, rather than because it is working, is just a larger loss.
5

5. Accept the underspend

A legitimate outcome. A campaign that cannot spend efficiently should not spend. Record it so the retrospective can ask why the plan assumed it could.

Record every change

Any budget change explains a movement in a later report. Undocumented changes are why a report says performance improved when what actually happened is that someone moved $4,000. Every change that moves money is a person’s decision; an agent recommends with its findings attached and never moves budget on its own. Log the date, what changed, by how much, who decided, and why, in the data dictionary’s decisions and changes section. A change an agent executes through a platform API is logged there at execution time by construction; a manual change still needs the entry. The campaign QA checklist checks this weekly, and budget reporting reads from it.

Exceptions

Always-on budgets pace against a monthly rate rather than a flight. The calculation is the same with the month as the denominator, and the month-end cliff is the thing to watch: a campaign that paces at 1.15 all month stops delivering in the last four days.
Type: process
Domains: budget, reporting
Provides: process:budget-pacing, process:budget-pacing
Uses: budget-plan, campaign-structure-best-practices, flat-file, glossary, library:change-log
Hands off to: process:budget-reporting, checklist-qa
Triggers: “Are we on pace?” · “Campaign X is underspending” · “Why is spend so high this week?” · “Check pacing” · “Should I raise the budget?”Asks, in order:
  1. Which campaign? Options: library:campaign. Skip when: one live campaign.
  2. Here’s pace and the likely cause. Which change do you want to make? Options: pacing-interventions. Skip when: pace on plan.
Output: Pace value, band verdict (week-1 aware), delivery diagnostics, recommended intervention; on owner decision, change-log entry (date, change, amount, decider, reason). Never moves budget itself.

Get the complete Marketing OS

Subscribe to Snow Doughnut, our free weekly newsletter, and get the Notion workspace where you do the work: tasks, a marketing calendar, name generators, briefs, and more.

Have us run it for you

We implement this system, use it to help you grow and scale, then hand it all off to you when you’re ready.
Last modified on November 2, 2026