Ad budget forecasting: how to forecast quarterly and annual ad spend and when to re-forecast
To forecast quarterly and annual ad spend, build each month from the campaign-level plan, replace past months with actuals, forecast the current month from its run rate and schedule, and keep future months at plan unless something has changed. Re-forecast on a fixed cadence: weekly for the current month, monthly for the quarter and quarterly for the year. Show a range, not one number, and explain every change from the last version in a short variance bridge.
Month-end forecasting is not annual forecasting
Forecasting this month’s spend is a pacing problem: you know the budget, the days and the run rate. Forecasting the quarter or the year is a planning problem. Budgets will change, campaigns will start and stop, and platforms will behave differently in a seasonal peak. Our guide to month-end forecasting covers the first. This guide covers the second.
The approach is the same in both cases: start from what you know, project the rest with simple, visible rules and update often. Complex models rarely beat that for paid media budgets, and they are harder to explain to finance.
The three layers of a forecast
Treat the forecast as three layers that you can read separately.
| Layer | Source | How to update it |
|---|---|---|
| Closed months | Actual platform spend, reconciled to invoices | Replace plan with actuals at each close; do not edit afterwards |
| Current month | Run rate and ad schedule, per campaign | Weekly, or daily if you pace daily; show the gap to plan in dollars |
| Future months | Approved plan, adjusted for known changes | Monthly; change only for named reasons (launch, pause, budget approval, seasonality) |
The annual total is the sum of the three layers. If someone asks why the forecast moved, you can point to the layer: a closed month came in lower, the current month is trending over, or a future month was changed.
A re-forecast cadence that holds up
- Weekly: re-forecast the current month by campaign. Share the total and the gap to plan.
- Monthly, after the close: replace the month with actuals, re-forecast the quarter and write the bridge.
- Quarterly: re-forecast the year, review the plan against pipeline results, and decide whether to move budget between platforms or programs.
- On a trigger: when a budget is raised, a campaign paused or a launch moves, update the affected months the same day.
The cadence matters more than the method. A simple forecast updated every week beats a clever one updated every quarter.
Show a range, not a single number
Spend is more predictable than conversions, but not perfectly: ad schedules, audience size, seasonality and bid changes all move it. Give three numbers for the quarter: a low case (current run rate, no catch-up), an expected case (plan adjusted for known changes) and a high case (plan plus any approved increases and a catch-up for current underspend).
A range lets finance plan for the worst case and lets marketing say honestly where the money is going to land. It also makes the conversation about the assumptions, not about whether the number is right.
A worked example with hypothetical numbers
Hypothetical: the annual paid media plan is US$600,000, US$50,000 a month. After four months, actual spend is US$188,000, US$12,000 below plan, because a LinkedIn Ads campaign launched late and one Google Ads campaign was capped by audience. The current month is forecast at US$47,000. Future months stay at plan, US$50,000 times seven.
Annual forecast: US$188,000 + US$47,000 + US$350,000 = US$585,000, US$15,000 below plan. The bridge to the last forecast of US$592,000 is a US$5,000 drop in the current month and a US$2,000 rise in a future month for a confirmed launch. The expected case is US$585,000. The high case, if the audience is widened and the underspend is caught up in the second half, is US$600,000. The low case keeps the current month’s rate and lands near US$575,000.
The story for finance is now short: we are 2.5% below plan, here is why, here is the range, and here is the decision we need if you want the money spent.
Commitments and the floor
Not all budget is flexible. Insertion orders, purchase orders and fixed-flight campaigns commit spend, and finance expects those amounts to land. Mark them on the forecast as a floor. Spend above the floor is the flexible part you can move between programs. Our article on tracking spend against insertion orders and purchase orders shows the layout.
Do not forget accruals. Finance needs the month’s spend recorded in the month it happened, before the invoice arrives. A forecast built from campaign-level data makes the accrual a by-product instead of a monthly scramble.
From spend to pipeline
For B2B, a spend forecast without a pipeline view answers only half the question. Pair the forecast with cost per opportunity and the pipeline each program is expected to create, using the lag your sales cycle implies. A quarterly forecast that says spend is on plan while opportunities are behind is the signal to reallocate, not to celebrate. Our paid media management team builds this view for the accounts we run, with the CRM connected so that spend and pipeline sit in one report.
Tooling
A spreadsheet handles a quarterly forecast for a modest number of campaigns. Beyond that, the weekly refresh is the bottleneck. Pacing Console forecasts month-end spend and conversions per campaign from the run rate, and on Pro adds forecast ranges, quarter-end views and CRM lifecycle stages by campaign. Enterprise adds scenarios and budget groups for teams that plan by brand and region.
The first month is free for up to 3 users, no credit card. After that Pro is $49 USD per workspace a month. To compare options, see our ranked list of ad spend forecasting tools.
Key takeaways
Keep going
Frequently asked questions
How do you forecast annual ad spend?
Use actuals for closed months, a run-rate and schedule-based forecast for the current month, and the approved plan for future months, adjusted only for named changes. Sum the three layers, show a low, expected and high case, and re-forecast on a fixed cadence.
How often should I re-forecast my ad budget?
Weekly for the current month, monthly for the quarter after each close, and quarterly for the year. Also update the same day when a budget is raised, a campaign is paused or a launch moves.
What is a variance bridge for ad spend?
A short walk from the last forecast to the new one: each change (a budget raise, a paused campaign, a launch date, seasonality) with its dollar effect, then the new total. It explains why the number moved without a long conversation.
How accurate can an ad spend forecast be?
Spend is more predictable than conversions, but ad schedules, audience limits, seasonality and bid changes all move it. Give a range, and measure your own accuracy by comparing each month’s forecast with its actual.
Should the forecast include pipeline?
For B2B, yes. Pair spend with cost per opportunity and expected pipeline, using the lag of your sales cycle, so that on-plan spend with behind-plan opportunities triggers a reallocation decision.
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.