Why a seasonal peak changes the calculator's inputs
The calculator's default use treats cost per acquisition as stable across the period. That holds for most of the year. During a peak, however, more advertisers compete for the same audiences, and prices in the auction rise — the CPMCost per 1,000 impressions. and cost per click both climb. Shoppers convert more readily too, which offsets some of the rise, but the net effect on cost per acquisition is usually upward. A budget built on the annual average therefore buys fewer sales than planned at exactly the moment they matter most.
The other inputs shift as well. The window is short, often a month or less, so the monthly output equals or nearly equals the total. The reserve that funds experiments on an ordinary campaign is less useful at the peak, because every test that fails does so at the year's highest prices. Teams that plan well move their testing into the run-up and spend the peak on what has already proved itself. The budget allocation guide treats this as allocation across time: deciding how much of the year's money belongs in the peak before deciding which channels get it. The pages that receive the peak's traffic need planning too, and the guide to seasonal and sale event pages keeps one permanent page per event so it carries its links and history from one peak to the next.
The inputs for a seasonal peak
| Calculator input | On the default page | For a seasonal peak | Where to find it |
|---|---|---|---|
| Conversions needed | Sales for the planning period | Sales the peak window must deliver, from last peak and this year's plan | Sales history for the same weeks |
| Target CPA | Average or target CPA | The CPA observed in the same window last year, adjusted for known changes | Platform reports for last year's peak weeks |
| Testing reserve | A standard share for tests | A small reserve for reacting during the peak; tests run in the weeks before | Plan |
| Months | A quarter or similar | The peak window, usually one month or less | Trading calendar |
For the CPA, pull the platform reports for the same calendar weeks last year rather than the same month, since peak dates often fall differently. If the product range, pricing or landing pages changed materially since then, adjust the figure, but start from the observed peak cost rather than nudging the average upwards by a guess. For the goal, combine last year's peak sales with this year's growth plan; check that the stock and fulfilment plan can actually supply that many orders in the window.
Entering the seasonal inputs step by step
Fix the peak window on the trading calendar
Mark the first and last day of concentrated demand, using last year's daily sales to find where the rise begins and ends. Enter the window as the months input, rounding up to the nearest whole month.
Pull last peak's acquisition cost
Export platform data for the same calendar weeks last year and calculate cost per acquisition for those weeks alone. Enter that figure, not the annual figure, in the CPA field.
Set the goal and a small reserve
Enter the sales the peak must deliver. Set the reserve low and record which tests ran in the run-up, so the peak budget spends on creative and audiences already proven.
Build a daily pacing plan from the total
Take the calculator's total and divide it across the window according to last year's daily demand. Load the pacing into the platforms and the spend tracker before the first peak day.
A worked seasonal peak with illustrative inputs
An online toy shop plans its four busiest weeks
Puddlejump Toys, an invented online toy retailer, wants 1,200 sales from paid media in the four peak weeks before the holidays. Its annual average CPA is $21, but in the same four weeks last year its CPA was $28. It ran its creative and audience tests in the six weeks before the peak, so it holds only a 10% reserve to react to stock changes. Entered with a one-month window, the calculator returns a total of $36,960, the same as the monthly figure, with $3,360 held in reserve. Planned on the annual average instead, the total would have been $27,720 — roughly a quarter short, enough to run dry in the final week, which last year carried the highest daily sales.
Marketing Budget Calculator
Open the Marketing Budget Calculator →Prefilled with the example's inputs: Conversions needed 1,200 · Target CPA $28 · Testing reserve 10% · Months 1 → Total budget $36,960 · Monthly budget $36,960 · Testing reserve amount $3,360.
All figures above belong to the invented retailer. Swap the CPA in the embedded calculator for the annual average to see the size of the gap a peak plan built on averages would leave.
Interpreting a peak-season budget
Read the total budget as the money the peak itself needs. With a one-month window, the monthly output simply repeats it, so the useful comparison is not between the two outputs but between the total and the amount the annual plan set aside for those weeks. If the annual plan assumed the average cost, the calculator's answer will be higher, and that difference is the conversation to have with finance before the peak, not during it.
Next, consider the CPA you entered against the annual figure. Even when conversion rates rise in the peak, the higher price of each auction usually means each sale costs more. Treat the peak CPA as the planning CPACost of each conversion. for those weeks and keep the everyday target for the rest of the year; mixing them in a single report makes the peak look worse than planned and the off-peak look better.
Then read the testing reserve amount. A small figure here is deliberate. The reserve is for reacting — shifting money to a product line that is selling faster than expected, or answering a competitor's offer — not for discovering new audiences at peak prices. If the team still has untested ideas when the peak opens, the right move is usually to wait until the following quarter.
Finally, convert the total into daily pacing. The calculator spreads money evenly, but peak demand is uneven: it builds, spikes on particular days and tails off. Allocate the total to each day by its share of last year's sales or search interest, and track spend against that curve. The budgets and pacing guide explains how daily budgets behave in the auction, and the guide to tracking spend against budget covers the daily variance report that keeps the peak on plan. Use daily cost per sale as the leading KPIA metric tied directly to an objective. during the window. Before the peak opens, the sale promotion readiness checklist confirms the budgets and pacing are set alongside the offer, stock and site checks. Where much of the peak budget runs on Meta, the guide to planning Meta Ads around a sale or peak explains how daily and lifetime budgets pace through the window and why the ad sets should launch before it.
The calculator cannot see what happens after the peak. Many categories see a dip in the following weeks as purchases were pulled forward. Plan for that dip so the post-peak quarter is not judged against a trend inflated by the peak.
Pitfalls when peak-season numbers go into the calculator
- Planning the peak on the annual-average CPA. The total comes out too low, and the shortfall lands on the busiest days, when it costs the most sales.
- Testing new creative during the peak days. Every losing variant spends at the highest prices of the year. Run tests in the run-up and carry the winners into the peak.
- Spreading the budget evenly across an uneven window. An even split overspends the quiet opening days and underspends the days of greatest demand. Pace by the demand curve.
- Ignoring the pull-back in the weeks after. A sharp dip after the peak is often the cost of demand brought forward. Budget and report the post-peak weeks with that dip expected, rather than treating it as a fault to fix with extra spend.
