ROAS vs MER: measuring paid media when attribution is broken

Why platform ROAS can’t be your only number

• Every platform credits itself. A customer who watched a TikTok ad, clicked a Meta ad and then searched your brand on Google can be counted as a conversion by all three.
• Some conversions are modelled. Where tracking has gaps, platforms fill them with statistical estimates.
• View-through attribution is generous. An ad that was merely shown can get credit for a sale that would have happened anyway.
• Retargeting and branded search flatter themselves. They reach people who were already close to buying, so their in-platform numbers often look better than their true impact.

Platform ROAS is still useful for comparing campaigns, ad sets and creatives within a platform. Think of it as the steering wheel, not the scoreboard.

MER: the number your bank account agrees with

Marketing efficiency ratio (MER), sometimes called blended ROAS, compares all revenue with all marketing spend over the same period:

It can’t be double-counted and it lines up with your profit and loss. Its weakness is that it doesn’t say which channel drove the result, and it includes revenue from returning customers and organic demand. Two variations fix most of that:

• New-customer MER: revenue from first-time customers divided by ad spend. It shows whether paid media is actually acquiring people, not just re-selling to existing ones.
• New-customer CAC: ad spend divided by the number of new customers. Compare it with what a new customer is worth to you.

Know your break-even before you look at any ROAS

A ROAS of 2.0 can be excellent for one brand and loss-making for another. It depends entirely on your margins. Take a simple, hypothetical example:

• Average order value: $80
• Product cost: $24, shipping and fulfilment: $10, payment fees: $3
• Contribution before marketing: $43, a contribution margin of 53.75% ($43 ÷ $80)

In this example, any ROAS below roughly 1.86 loses money on the first order, and the most you can pay to acquire a customer on their first purchase is $43. If customers reorder, you can afford to pay more, but only once you know your repeat rate. Start with first-order break-even, then extend it to 90-day or 12-month customer value as your data matures.

Check incrementality, not just attribution

Attribution tells you who gets credit. Incrementality tells you what would not have happened without the ads. A few practical ways to measure it:

• Lift studies. Several ad platforms offer conversion lift tests that compare people who could see your ads with a holdout group that couldn’t. Eligibility usually depends on spend.
• Geo holdouts. Reduce or pause spend in a set of regions for a few weeks and compare sales against similar regions. In the US, states or metro areas make natural test cells.
• Post-purchase surveys. A single “How did you first hear about us?” question at checkout is cheap and surprisingly revealing.
• Simple on/off reads. When a channel pauses for any reason, watch what happens to total revenue. It is noisy, but informative.

A weekly scorecard that keeps everyone honest

1. Total revenue and new-customer revenue
2. Total ad spend, by channel
3. MER and new-customer MER
4. New-customer CAC against your break-even
5. Contribution margin after ad spend
6. Platform ROAS and CPA by campaign, as directional signals

None of this needs expensive software. A shared spreadsheet updated every Monday is enough to make better decisions than most dashboards ever will.