Marketing efficiency ratio vs ROAS, POAS, ROI and nCAC

Marketing efficiency ratio (MER), ROAS, POAS, ROI and nCAC all divide something by what you spent, so they get mixed up. They answer different questions, and using one for another’s job is how a shop ends up pushing a number up while profit stays flat.

The ROAS and POAS parts build on break-even ROAS per product, and every example uses round numbers you can check by hand.

Five metrics, five questions

Each answers its question at one or more levels, and only MER cannot go below the account.

MER can’t be read per product because its top line is every order the shop took, including orders from email, organic search and returning customers that no product’s ads caused. Dividing that by one product’s spend would credit the product with revenue it didn’t earn. The other four need only the spend and the orders linked to it, so they work at any level.

What is MER?

Short for marketing efficiency ratio, MER divides total revenue by total marketing spend. It answers “how much revenue does each euro of marketing support?” with one division:

Illustrative data.

The two channels are credited with €30,000 between them, a combined ROAS of 3 on the €10,000 of ad spend. The shop took €50,000 in total, because €20,000 of orders came from email, organic search, direct visits and returning customers that neither channel is credited with. Divide the full €50,000 by the €10,000 spent and you get a MER of 5.0. The MER calculator does the same division with your own totals and adds nCAC and marketing ROI.

MER is useful because it uses the shop’s own revenue, so it doesn’t change when two platforms argue over who deserves the credit for an order. It is also the number to watch when you change total spend. What it hides is the split: here one channel returns €4.00 per euro and the other €1.50, and a MER of 5.0 looks healthy either way. To decide which channel or product to fund, go down a level.

What is a good MER?

Your margin sets the bar, and another shop’s average says little about it. MER counts revenue before costs, so it hides margin in the same way ROAS does, and the line to beat is your break-even MER.

Break-even MER is 1 ÷ your average contribution margin. At a 25% margin, every €1.00 of revenue leaves €0.25 after the product and the order, so €1.00 of marketing needs €4.00 of revenue just to break even. Below that MER the marketing costs more than the profit it supports. Fixed costs and the profit you want come on top.

This uses a single average margin across all revenue, and that average moves when your product mix changes. A MER that holds steady while the mix drifts towards thin-margin products is getting worse, and the MER itself will not tell you.

ROI is POAS minus 1

POAS is profit counted before ad spend, divided by ad spend, and ROI is what is left after the ad spend, divided by the same ad spend. So with profit counted before ad spend, ROI = (profit − ad spend) ÷ ad spend = POAS − 1. A POAS of 1.0 is therefore an ROI of 0%, and both mean break-even.

Put €1,000 of ad spend against €4,000 of revenue, a ROAS of 4, at three margins:

Illustrative data.

All three rows have the same ROAS. The margin turns it from a loss into a doubling of the ad spend, which is why what is a good ROAS says a good ROAS depends on margin, and why POAS vs ROAS shows POAS as ROAS × contribution margin.

Plotted across ROAS, each margin crosses zero at a different point (every point is ROAS × margin − 1):

If you need the break-even per product, the break-even ROAS calculator gives it from a margin. The glossary has short definitions of ROAS and marketing ROI, including the common mistake of calling revenue ÷ ad spend ROI.

nCAC prices a new customer

New customer acquisition cost (nCAC) is ad spend divided by the new customers it brought. It differs from ROAS because it counts people instead of revenue, and from cost per order because returning customers are left out. Google Ads Help defines customer acquisition cost the same way: the ad spend allocated to new customers divided by the unique new customers acquired through a campaign.

In the example, €10,000 of spend with 200 orders looks like €50.00 an order, but 80 of those orders came from returning customers, so each of the 120 new customers cost €83.33. Customer acquisition cost for ecommerce shows how to work out the most you can pay for that first order, and the glossary defines CAC.

LTV:CAC compares what a customer has brought in over a chosen window with what the first order cost, and it needs the lifetime value counted in profit. LTV:CAC ratio for ecommerce explains how to calculate it from observed orders.

nCAC is also one of the metrics you can read per product. Product Metrics shows nCAC per product and as an average. The average is ad spend divided by new customers, who have their own conversion action in Google Ads, and it hides which products bring new customers cheaply and which don’t.

Match the metric to the decision

Start from the decision and work down to the number:

In practice, MER sets the total budget, ROAS or POAS decides how it is split between campaigns and products, and nCAC tells you whether first orders pay back. Product Metrics works at that product level: Product Segmentation segments products by their ad clicks and their return, where return is ROAS or POAS, so the POAS-above-1.0 check runs for every product. It works on ROAS straight away until margins are connected.

Every metric depends on recorded orders

Each of them counts only the orders that the ad platform and the shop recorded. If an order never reaches the ad platform, ROAS, POAS, ROI and nCAC all look worse than the ads really did. Shop revenue still holds the order, so MER is less exposed, but the split between channels and products is not.

nCAC has an extra requirement: every order must be marked new or returning. Google Ads can detect it automatically or take it from the conversion tag, according to Google Ads Help, and a missing flag pushes new customers into the returning count.

To see how much your own setup loses, run the ad blocker conversion tracking test in your browser. Full Signal Tracking runs on a unique endpoint on your own domain and gets through every adblocker, so more of your orders reach Google Ads.

None of these metrics decides which click earned an order. Whether the ads caused the orders at all is a separate question, which an incrementality test answers with a holdout or geo split.

Two checks on your own numbers

The break-even ROAS calculator gives the starting number for each product, and Product Segmentation runs the same check across every product.

First, work out your break-even MER from your average margin and compare it with your current MER. Then, for the products that take most of your spend, multiply ROAS by margin and check that POAS is above 1.0.