Paid media planning

Find the ad budget that makes the most profit

More spend keeps adding revenue long after it stops adding profit. Enter your order costs and current ad results to see where monthly contribution peaks, and whether the next move is to scale or cut back.

Pre-filled with an example brand. Replace the numbers with your client's; they're kept in this browser only.

Step 1 · Unit economics

What one order leaves you

Costs as a share of order value. What remains is the most an ad can cost per order before you start losing money.

MYR
% of AOV
% of AOV
% of AOV
% of AOV
% of AOV

Platform & other fees covers marketplace commission, packaging or anything else that grows with each order. Leave it at 0 if none.

Contribution margin–
Break-even ROAS–
Break-even cost per order–

Step 2 · Current ads

What this month's budget creates

Enter what the client actually spent and the ROAS they got for it. To test other budgets, use Ad Spend Scenarios in Step 3. Blended or incremental ROAS is truer than the ads manager figure, which tends to claim sales that would have happened anyway.

MYR
×
Monthly contribution from ads
Contribution after ad spend, per month –
Orders from ads–
Ad cost per order–

Step 3 · Profit curve

What happens when you spend more

Each extra ringgit reaches people a little less ready to buy, so ROAS slips as the budget grows. The curve shows where profit peaks; the scenarios below test specific budgets.

Would a different budget make more profit?

Estimate, based on typical ROAS

Profit curve: contribution after ad spend

MYR per month at each budget, if ROAS follows the typical path

The curve is the typical path: what happens if only the budget changes and ROAS slides as it grows. Hover or tap it to read any budget. Ad Spend Scenarios below test what better-than-typical ROAS would be worth. Change Step 2 only when the client’s actual results change.

Typical: ROAS falls about 20% each time spend doubles. Change
20%

Calibrate the decay from your own months

Enter ad spend and ROAS for 3 to 12 past months. The fit measures how fast ROAS fell as spend rose, and you can apply it to the curve above.

Use months with similar promotions and creative. Putting an 11.11, 12.12 or Raya peak next to ordinary months distorts the slope. Use the same ROAS definition throughout.

MonthAd spendROASRemove
Underspending
Efficient, but small

Only the warmest buyers are being reached, so ROAS looks great. Each extra ringgit still returns more margin than it costs, so profit is being left on the table.

Sweet spot
The most profit in total

Within 5% of the curve's peak. ROAS is lower than at small budgets, but the contribution left after ads is as high as it gets.

Overspending
Revenue up, profit down

Spending beyond the best range, each extra ringgit brings back less margin than it costs. Revenue still grows, but profit falls.

Ad Spend Scenarios

Each row starts with what typically happens to ROAS at that budget. Change the ROAS to show what holding or improving it would be worth.

Change in monthly contribution vs today

Other budgets at typical ROASSame margins, with ROAS following the typical path.
Monthly spendROASRevenueCM after adsvs now
How the numbers are calculated
Contribution margin = 1 − (cost of goods + shipping + payment + returns + other fees, as % of AOV)
Break-even ROAS = 1 ÷ contribution margin
Contribution after ad spend = spend × ROAS × margin − spend
Typical ROAS above current spend S₀ = current ROAS × (S ÷ S₀)−k,   k = −log₂(1 − drop per doubling); below S₀, current ROAS × (1 + k/3 × (1 − S ÷ S₀)), easing off below about a tenth of S₀ where budgets are too small to optimise
Next-unit ROAS = (1 − k) × typical ROAS above S₀  ·  Peak: where next-unit ROAS × margin = 1, or today’s spend if neither adding nor cutting pays
  • Contribution after ad spend comes before fixed costs such as salaries, rent, software, agency fees and tax. It is not net profit.
  • The model treats every sale counted in your ROAS as caused by the ads. If that ROAS includes sales that would have happened anyway, every figure here is optimistic.
  • The optimal zone covers every budget within 5% of the curve's peak contribution.
  • This is a planning model, not a forecast. Change budgets in steps, measure for 2 to 4 weeks, then recalibrate.