An ad spend alert fired: a 15-minute triage runbook for overspend and underspend
When a pacing alert fires, do five things in order: confirm the data is fresh, read the gap in dollars, find the cause (shared budget, recent change, schedule, audience limit or tracking), set the daily budget to what is left divided by the days left, and log the decision. Most alerts are resolved in 15 minutes if you check the cause before touching the budget. Changing the budget first hides the cause and repeats the alert next week.
Why a runbook beats a gut reaction
An alert tells you a campaign is outside its band. It does not tell you why, and the right action depends on why. A cut that fixes an overspend caused by a shared budget does nothing for an overspend caused by a bid change. A raise that fixes an underspend caused by a budget cap makes no difference to one caused by a small audience.
A fixed order of checks keeps the response short and the same every time, which is what you need when several people can receive the alert. If you are still choosing thresholds, start with our article on which budget pacing alert thresholds work, and for the delivery channels see how to set up alerts in Slack, Teams and email.
The five steps
- Confirm the data. Check the time of the last sync for the platform and account. If the data is hours or days old, or a platform has stopped sending data, the alert is about the feed, not the campaign. Fix the connection first.
- Read the gap. Note spend to date, expected spend to date, pace, forecast month-end and the gap to plan in dollars. A 20% gap on a small campaign can matter less than a 6% gap on the largest one.
- Find the cause using the table below. Spend five minutes here and you save the repeat alert.
- Act. Set the daily budget to the budget left divided by active days left. Move in two steps if the change is more than about 20%. Or decide to wait, move the money or tell the client or finance.
- Log it. Write the date, the campaign, the cause, the change and who approved it. Re-check in two days.
Finding the cause
Use the pattern of the miss to narrow the cause. The table lists the usual suspects for B2B paid media, in the order we check them.
| Symptom | Likely cause | What to do |
|---|---|---|
| Over plan for several days, one campaign | Recent bid strategy, target or budget change | Compare spend before and after the change date. Revert or lower the budget to the amount that lands the month. |
| Over plan, several campaigns, same account | A shared budget feeding all of them | Check which campaign took the share. Split the budget or set a per-campaign limit. |
| Over plan after a mid-month budget raise | Plan not updated, or daily budget not recalculated | Update the plan record and recalculate the daily budget from what is left. |
| Under plan from day 1 | Late start, approval delay or a tracking or disapproval problem | Check status, start dates and ad approvals before touching the budget. |
| Under plan, small audience, budget not limiting | Audience or bid cap too low | Widen the audience, raise the bid or move the money. More budget will not help. |
| Weekday-only campaign looks behind | Pacing against calendar days | Pace against active days and check the daily budget formula for the schedule. |
| Everything under plan on one platform | Data feed stopped or lagging | Check the last sync before changing any budget. |
For Google Ads specifically, remember that a day above the average daily budget is allowed, up to twice that amount, and that the monthly limit for most campaigns is 30.4 times the average daily budget. A single high day is not an incident. Several days in a row is.
Who acts, and by how much
Decide before the first alert who owns each campaign and what they may change without asking. A workable rule: the owner can move a daily budget up to 15% from plan on their own. Larger moves, or any move that takes the month above the approved total, need a named approver. Write the rule in the same place as the alert so nobody has to look it up under pressure.
On teams that use automated budget changes, set a floor, a cap and a maximum step per run, and make sure software never pauses a campaign on its own. Our article on whether software should change your ad budgets covers the guardrails.
A worked example with hypothetical numbers
Hypothetical: an alert shows a Google Ads campaign at 118% of expected spend on day 14 of a 30-day month. Plan is US$15,000. Spend to date is US$8,260 against expected US$7,000. The sync is two hours old, so the data is fine.
The campaign’s forecast is US$17,700, US$2,700 over plan. A check of recent changes shows the target was loosened on day 9. The campaign now spends about US$590 a day. Budget left is US$6,740 over 16 days, so the daily budget to hit is about US$421. Lowering straight to US$421 is a 29% cut, so we go to US$500 today and US$421 in two days after re-forecasting. The note reads: target loosened on day 9, budget lowered in two steps, owner approved, re-check on day 16.
After the incident: stop the repeat
- If the cause was a plan that was never updated, fix the process that lets a budget change skip the plan.
- If it was a shared budget, decide whether the sharing is worth the lack of control.
- If it was a band that was too tight, adjust it, and write down why.
- Review last month’s log for repeat causes. Two repeats of the same cause is a process problem, not an incident.
How Pacing Console fits
Pacing Console sends a morning email with only the campaigns that are over budget, overpacing or underpacing, the daily budget that fixes each one, and a warning when a platform has stopped sending data, which is step 1 of this runbook. On Pro, alerts can also go to Slack or Microsoft Teams, and recommendations let you apply a Google Ads or Microsoft Ads budget change, logged with who asked.
The first month is free with no credit card. Solo is $19 and Pro is $49 USD per workspace a month after that. The same runbook applies to the accounts our paid media team manages.
Key takeaways
Keep going
Frequently asked questions
What should I do when a budget pacing alert fires?
Confirm the data is fresh, read the gap in dollars, find the cause, set the daily budget to what is left divided by the days left, and log the decision. Check the cause before changing the budget.
Why does a campaign look underspent when it is not?
Often the data is stale or a platform stopped syncing, or a weekday-only campaign is being paced against calendar days. Check the last sync and the schedule before changing any budget.
Should I pause a campaign when an overspend alert fires?
Usually not. Find the cause, then lower the daily budget to the amount that lands the month, in steps if the change is large. Pause only if finance has set a hard cap.
How do I stop the same pacing alert from repeating every week?
Treat repeat causes as process problems: update the plan whenever a budget changes, review shared budgets, and adjust bands that are too tight. Review the log monthly for repeats.
Who should be allowed to change a budget after an alert?
The campaign owner, within a written limit such as 15% from plan. Larger changes, or any change that takes the month above the approved total, should need a named approver.
Sources
Related service: Paid media. See also pricing and client results.
We run enterprise B2B demand generation for teams in Montreal, Toronto, Ottawa and the United States.