The Ecomm Analyst

Three numbers get used interchangeably in operator conversations, and treating them as the same thing leads to some genuinely bad decisions. Platform ROAS, blended ROAS, and MER measure different things, answer different questions, and break in different ways. Here is how I keep them straight.

Platform ROAS

This is the number Meta, TikTok, or Google reports inside the ad account. It is the platform grading its own homework. It uses the platform attribution window and its own definition of a conversion, and it has no idea the other platforms exist, so when several channels all take credit for the same order the numbers overlap. The absolute value is almost always inflated. What it is good for is relative comparison inside a single account. Campaign A at a reported 3.5 against Campaign B at 1.8 tells you something real about which is working better inside Meta, because the measurement bias is roughly the same across both. Use it to steer spend within a platform. Never plan the business off it.

Blended ROAS and MER

These two are basically the same calculation. Take all the revenue the business actually made in a period and divide it by everything you spent across every channel. Some people call it blended ROAS, some call it MER for marketing efficiency ratio, and the distinction is mostly vocabulary. The value of this number is that there is no double counting hiding inside it. The denominator is total spend, the numerator is real revenue, so it cannot be inflated by platforms fighting over credit. If you did 200,000 in revenue on 50,000 of spend, your MER is 4.0, and that is a fact about the business rather than a claim by an ad platform. This is the number you plan against and run a P&L on. What it is bad at is telling you which specific campaign or creative to cut, because it averages everything together.

New customer MER, the one most people skip

Blended numbers have a blind spot. A brand with strong repeat purchase rates can post a comfortable MER while its acquisition is actually underwater, because returning customers who would have bought anyway are propping up the blend. The fix is to look at new customer revenue divided by total spend, sometimes called new customer MER. This strips out the repeat revenue and asks the harder question, whether the money you spend on marketing is actually bringing in new customers efficiently or you are paying to re-acquire people you already had. For most brands, acquisition is the thing marketing spend is supposed to buy, so this is often the truest read on whether spend is working.

Which number for which decision

• For scaling or cutting a specific ad or campaign, use platform ROAS and read it directionally.
• For planning the month, setting budgets, and checking unit economics, use MER or blended ROAS.
• For judging whether acquisition is healthy, use new customer MER.

Set targets on the number people can move

When you set a target, set it on the figure that matches the decision and the person making it. A media buyer can be held to a platform efficiency floor or a new customer MER, because those are things they can actually influence inside the account. The business as a whole gets planned to a blended MER. Handing a buyer a blended target they cannot see or move inside the ad platform just creates frustration and gaming. Match the target to the lever.

The trap

The most common error is mixing the layers. People set a blended MER target and then try to optimize individual campaigns to hit it, which is impossible because the campaign only sees its own inflated platform number. Or they read platform ROAS as if it were real revenue, plan spend off it, and then wonder why the bank account does not match the dashboard. Keep the jobs separate. Platform numbers steer within channels, MER plans the business, and new customer MER tells you if acquisition is real.

None of these is the one true number, and looking for a single metric that does everything is how people end up trusting the wrong one. Each answers a specific question. The skill is knowing which question you are actually asking before you reach for a figure.