The case against MER as a primary KPI for acquisition.

Blended MER is the most popular paid media KPI on the market and the most misleading one in the toolkit. It is fine for the finance layer. It is wrong for the question most operators are actually trying to answer.

The question is whether paid media is buying new customers profitably. MER cannot answer that. Not because the metric is broken, but because returning customer revenue is sitting inside the numerator and the operator is reading it as acquisition health.

This is the mistake most brands make. They watch MER lift, take it as a green light to scale, and end up funding their own retention with new customer dollars. The cohort comes back regardless. The acquisition curve quietly flattens. By the time anyone notices, the spend ladder has already broken something.

What MER is actually measuring.

MER is total revenue divided by total ad spend. It includes first-time buyers, returning buyers, subscription renewals, post-purchase upsells, the email-driven repeats, and everything that closes inside the attribution window. All of it counts toward the numerator. All of it inflates the read.

That is not a flaw. That is what the ratio is for. MER is a finance metric. It tells the business how efficiently total revenue is being produced relative to total media investment. It is useful for cashflow planning, for setting spend capacity, and for executive-level conversations about marketing's contribution to the P&L.

None of those are the same as judging whether paid media is doing the job it was hired to do.

The thing returning customer revenue does to the picture.

Take a brand running at a 3.0 MER with 60% returning customer revenue. The acquisition layer of that business is doing something close to a 1.2 MER on new customers. That is a profoundly different number. If the operating margin is 50%, the contribution math on those two reads is not even in the same conversation.

The brand is not failing. But the read is. And the read decides where the next dollar goes.

What ends up happening is the team scales spend against a flattering blended number, the returning customer base does not scale at the same pace, and three months later the MER drift looks like a platform problem. It is not a platform problem. It is the blended number doing what it was always going to do once new customer revenue had to actually pull its weight.

If your acquisition layer is underperforming and your blended MER is healthy, you are funding retention with acquisition dollars and calling it growth.

The replacement: acquisition MER.

The fix is structural. Split the ratio. New customer revenue divided by paid media spend. That is acquisition MER, or aMER for short. It does not flatter the read. It tells you whether the spend is actually buying new buyers.

Two things follow from that. First, the number gets harder. A healthy aMER is typically lower than the blended MER the team is used to looking at, because the entire returning customer numerator just left. Second, the number becomes commercially defensible. You can hand it to a CFO and they can reconcile it against the first-order P&L. They cannot do that with blended MER without doing the same separation themselves.

Where MER still earns its place.

Killing MER off entirely is the wrong overcorrection. It still belongs in the operating cadence, just not at the top of it. Three places it remains useful:

• Cashflow capacity. If MER is dropping while aMER is stable, the business has a returning customer problem, not an acquisition problem. That is a fundable insight.
• Spend ceiling planning. Blended MER is the right ratio when modelling how much total spend the business can sustain at a target contribution margin. It is a budget conversation, not a performance one.
• Executive reporting. Founders and CFOs think in blended terms because that is how their P&L is structured. Reporting MER alongside aMER lets both audiences read the account.

The mistake is asking MER to answer a more specific acquisition question than it was built for. That is where it breaks. Keep it in its lane and it earns its place. Promote it above aMER and it will quietly mislead the team for a quarter at a time.

The thing even aMER cannot do for you.

Here is the bit most people will not say. Even aMER is still credited revenue. It is a cleaner read than MER, but it is not causal. A holdout test, a geo-lift, or an in-platform incrementality experiment is what tells you whether the spend actually produced the revenue the platform is crediting itself for.

The point of aMER is not that it ends the conversation. The point is that it starts a better one. If the operator wants to make capital allocation decisions on paid media performance, the chain looks like this:

1. Use aMER as the primary KPI for whether acquisition is healthy in a given period.
2. Reconcile aMER against first-order contribution margin to confirm it is producing profit, not just revenue.
3. Validate the read with incrementality testing on a meaningful cadence to confirm causality, not just credit.

That is the order. Skip steps and the math gets prettier the wrong direction.

So if you take one thing away from this

It is this. MER is not the wrong metric. It is the right metric for the wrong question. The question paid media should be held to is whether it is buying new customers profitably, and blended MER cannot answer that without unwinding itself.

Stop optimising acquisition against blended MER. Start optimising against aMER. Reconcile both against contribution margin. Validate with incrementality on a cadence. That is the hierarchy. Full stop.