The short answer

A target ROAS can be translated into an impression-cost ceiling when you have explicit assumptions for clicks per impression, orders per click, and order value. The calculation exposes what the traffic would need to achieve; it does not predict auction prices.

Calculate a CPM ceiling

Enter every input to calculate your scenario.

A click-based scenario. Assumed CTR, order rate and order value may change with delivery. Inputs stay in your browser. Displayed results are rounded.

Start with revenue per thousand impressions

Revenue per 1,000 impressions = 1,000 × CTR fraction × orders-per-click fraction × AOV
Maximum CPM at target ROAS = revenue per 1,000 impressions ÷ target ROAS

Enter percentages as fractions: an illustrative 2% rate means 0.02. Use matching definitions for clicks and attributed orders. This click-based model does not incorporate separate view-attributed revenue.

Work through a scenario

Illustrative inputs: CTR 2%, orders per click 3%, AOV $80, and target ROAS 4×. Expected clicks per thousand impressions = 1,000 × 0.02 = 20. Modeled orders = 20 × 0.03 = 0.6. Modeled revenue = 0.6 × $80 = $48. The CPM ceiling for the target is $48 ÷ 4 = $12.

The fractional order count is an average expectation over a larger volume, not a claim that a fraction of an order can occur. At 100,000 impressions with these assumptions, the model gives 60 orders and $4,800 revenue against $1,200 spend.

Distinguish a revenue target from break-even

If the example’s pre-ad contribution margin is 25%, the $48 of modeled revenue contributes $12 before ads. A $12 CPM then leaves zero contribution after ads. At a lower margin, the same 4× target would fail to cover the included costs. Build the cost-based requirement before deciding which ROAS target to use.

Check whether the assumptions can move together

A cheaper placement can have a different CTR or conversion rate. A creative that wins more clicks can also change order value. Compare the complete funnel rather than assuming that lowering CPM preserves all downstream inputs.

If either click rate or order rate is unknown, show a scenario range with stated assumptions instead of reporting a single confident ceiling. Keep platform reports and the model’s click definition aligned when reviewing actual performance.

Check the impression model

  1. Gather: State click-through rate, orders per click, AOV and the target ROAS multiple.
  2. Produce: Convert percentages to fractions and calculate revenue per thousand impressions first.

Decision boundary: The click-based model does not independently account for view-attributed sales.

Next reading

Examples are illustrative, not client results or benchmarks. The calculations use the inputs shown; real-world decisions depend on your records and assumptions. Report a correction.