MER vs ROAS: which number should you run the business on?

ROAS is the revenue an ad platform attributes to itself, divided by what you spent on that platform. MER, the marketing efficiency ratio, also called blended ROAS or media efficiency ratio, is total sales from your booking system or store backend, divided by total ad spend for the same period. In short, ROAS is the better tool for comparing campaigns inside one ad account, and MER is the better check on whether total sales are keeping pace with total ad spend.

Most of the arguments I see about whether ads are working come from people looking at different numbers without saying so. The ad account shows one return, the finance sheet shows another, and both sides are right about what they measured. This guide explains what each number is, how to calculate MER for a tour business or an online store, and how to use the two together.

What ROAS measures, and what it cannot see

ROAS, return on ad spend, is the conversion value a platform has recorded divided by the spend on that platform. Google Ads reports its ROAS. Meta reports its own. Each one only counts what its own tag or connection managed to record.

That leaves real sales outside the count. A booking that finishes on Viator or GetYourGuide is invisible to it. Phone bookings and sales at the desk are also missed unless you import them as offline conversions, and so is a checkout on another domain that the tag was never set up to follow. It can also count things it should not, which I come back to below.

ROAS is still useful for one job. Campaigns inside one ad account usually share the same tag and the same conversion settings, so ROAS is a reasonable way to compare them and to judge a test. That holds only while the conversion setup is the same for every campaign you compare. It is a poor way to judge whether the business got its money's worth.

What MER measures: the MER formula

MER stands for marketing efficiency ratio. Blended ROAS and media efficiency ratio are two other names for the same thing, so if you are asking what blended ROAS is, it is this division.

Total sales come from the system that actually takes the money: the booking system for a tour operator, the store backend for an e-commerce brand. Total ad spend is every paid platform added together, for example Google, Microsoft and Meta.

On this site MER means ad spend only. Some teams also put agency fees and software into the bottom of the division. Either version works, as long as you pick one and never change it from month to month, because a ratio whose definition moves cannot show you a trend.

MER is not attribution. It does not tell you which click caused which sale, and it moves with seasonality. It also rises when sales grow for reasons that have nothing to do with ads, such as repeat customers, word of mouth or organic search. A better MER does not by itself prove the ads improved. What it gives you is an answer to a simpler question: for every unit of money that went into ads, how much did the whole business sell.

MER vs ROAS side by side

When ROAS is wrong in both directions

The usual assumption is that platform ROAS flatters the ads. Sometimes it does. Often, in tourism, it does the opposite.

A London operator whose ads I run showed a 0.89x return in the ad account. The booking system showed 1.88x from direct website sales alone. Once I added the bookings that landed on the OTAs during the same campaigns, the same spend was sitting at roughly 2.4x. Judged on ROAS alone, that account looked like it was losing money. The backend numbers were a strong reason not to cut it on that basis, and a reason to look much harder at what the ads were actually doing.

Another operator I work with sells around 85% of their bookings through the OTAs. There the tag claimed more direct revenue than the booking system showed. The ads were clearly feeding the OTA listings, and the tag was clearly counting things it should not. I wrote up both cases in more detail in the guide on how ads lift OTA sales.

E-commerce brands see the same two-way error through different mechanisms.

• Two platforms can each claim the same order, so adding up platform-reported revenue counts one sale twice.
• View-through conversions can credit an ad the buyer may never have noticed.
• Refunds and cancellations can arrive after the platform has already counted the sale.
• Buyers who click on a phone and buy on a laptop can fall out of the count altogether.

MER sidesteps these problems because it only uses sales the backend recorded. It cannot tell you which platform earned them, which is why you still need ROAS alongside it.

How to calculate MER for a tour operator

You need three numbers for the same period, and you can put them in one row of a spreadsheet.

• The first is total ad spend across every platform you advertise on.
• The second is direct sales from your booking system, including website, phone and walk-in bookings.
• The third is OTA sales from your booking system or the OTA dashboards.

Most booking systems record the sales channel on each booking, which is how you separate direct sales from OTA sales. Make sure an OTA booking that also appears in the booking system is only counted once. Divide direct sales by ad spend and you have a narrow MER. Divide direct plus OTA sales by ad spend and you have the full blended figure. Neither one measures what the ads caused, because some of those sales would have happened without any ads. Read them as two views of the same month, and use the OTA lift guide to test how much of the movement the ads are responsible for.

How to calculate MER for a D2C brand

Take the order total from your store backend for the period and divide it by the total paid spend across every platform for the same dates.

The division is simple. Most of the work is in agreeing the inputs once and never changing them.

• Decide whether sales are gross or net of refunds and cancellations, and apply the same rule every month.
• Exclude sales tax from revenue, so you are dividing what the business keeps rather than what it collects for the tax office.
• Convert every figure into one currency at a consistent rate.
• Date sales by order date, not payout date. For tours, pick booking date or travel date and stick with it.

If you want to go one step further and compare channels without trusting each platform's own claims, you need clean source data in your analytics. My guide to GA4 source groups and hostname filters covers the two settings that make that comparison possible.

What is a good MER?

There is no universal number. Margin, seasonality, repeat purchase and channel mix differ too much from one business to the next, and a figure that is healthy for one operator can lose money for another.

The useful target comes from your own margin. Work out the return at which the ads break even on a first sale, then decide how much contribution you want above that. I walk through that one-line calculation in what is a good ROAS for a tour operator. The same logic works for MER, with one condition. The margin has to match the sales in the numerator. If MER includes OTA sales, use the margin you keep after OTA commission on those bookings, blended with your direct margin in the right proportions. The floor also only covers the costs you put into the margin, so decide whether it includes staff, fees and fixed costs or only the cost of delivering each sale. After that, judge your own trend against your own floor, month after month.

Using both without arguing about dashboards

Once both numbers exist, each one has a clear job.

• MER is the monthly scoreboard for the business.
• ROAS is for ranking campaigns, ad groups and tests inside one account.
• The gap between them is worth tracking. In tourism it usually mixes tracking loss, OTA lift and sales that would have happened anyway. A sudden change in the gap is a reason to investigate, not an answer in itself.

One warning. If you hand the platform a target ROAS while the tag is recording the wrong things, the bidding optimises toward the wrong number. I explain why in the section on target ROAS bidding. Fix the measurement first, then automate against it.

Common questions

Is MER the same as blended ROAS?

In practice, yes. Both are total sales divided by total ad spend for the same period, across every channel. Media efficiency ratio is a third name for the same division. What matters is that you define the inputs once and keep them fixed from month to month.

Should MER include agency fees and software?

It can, as long as the definition never changes. I use ad spend only, so the number sits next to the ad account and the two can be compared directly. If you include fees and tools, the ratio answers a slightly different question about total marketing cost, which is also useful if you track it the same way every month.

Can Google Ads bid to MER?

Not directly. Google Ads bids toward the conversion data it receives, which can include tag conversions, imported offline conversions and modelled conversions. MER is not one of those inputs. Use MER to judge the month and to set or move your target, and send the bid strategy the cleanest conversion data you can, including offline imports where they are set up.

Pull last month's total sales from your backend and your total ad spend from every platform, and divide one by the other. Put that number next to what your ad account reported. The difference between the two is where the useful conversation about your ads begins.