Meta, TikTok and Google all use their own attribution models, and those models are built to look good, not to tell you the truth. Meta claims every purchase where someone bought within seven days after a click, or within one day after seeing your ad without clicking. Google does the same. What happens next is that multiple platforms claim credit for the same purchase.

Someone sees your ad on Meta, does not click, searches on Google later, clicks there, and buys. Meta claims a view-through conversion. Google claims a click-through conversion. In reality, it was one purchase.

Add this up across all your channels and you understand why platform ROAS is structurally higher than reality.

The blended ROAS problem

There is a second way ROAS misleads you, and in our experience it is at least as dangerous as the attribution problem.

Your ROAS measures all revenue, including revenue from existing customers who come back through your ads. But those returning customers would most likely have bought anyway. They already know your brand. They were already planning to come back. You did not really acquire them, you just reminded them.

If a large part of your budget goes toward retargeting existing customers and warm audiences, the ROAS you see in Meta looks fantastic. But you are not growing. You are harvesting what was already there.

The three metrics that actually tell the truth

1) Marketing Efficiency Ratio (MER) Total revenue divided by total marketing spend across all channels. No attribution models, no platform claims, no overlap. Simply: how many euros in revenue does your brand generate for every euro you spend on marketing in total? A healthy MER for a growing e-commerce brand sits between 2.5 and 4.0. If yours is structurally below 2.5 and your gross margin is not exceptionally high, you are losing money on marketing, regardless of how good the platform ROAS looks.

2) New Customer ROAS (NC-ROAS) Revenue from first-time purchases divided by the budget spent on acquisition campaigns. This number is almost always lower than the total ROAS Meta reports, and that is actually good because it is honest. An NC-ROAS of 2.0 on a brand with 60% gross margin is healthy and scalable. A blended ROAS of 4.5 where 70% of revenue comes from returning customers is a very different story.

3) Attribution tools like Triple Whale and Northbeam These platforms connect your ad data with your actual Shopify orders and give you an incremental view of what your ads are genuinely contributing. They correct for attribution overlap and show you which channels are truly driving growth and which ones are just claiming credit for revenue that would have happened anyway.

What goes wrong when you optimize on the wrong metric

These are the patterns we see in brands that steer on platform ROAS for too long:

• More budget shifts to retargeting. Logical, because those campaigns show the highest ROAS. But retargeting does not scale. Eventually your warm audience is exhausted and you have no new customers in the pipeline. The ROAS still looks fine but revenue growth flatlines.

• Prospecting campaigns get cut. Because their ROAS is lower. This is exactly the opposite of what you should do. Prospecting is the only campaign that structurally brings new customers into your funnel. Stop it and you are only harvesting what already exists.

• Channels like TikTok or YouTube get written off. Because they show a lower reported ROAS than Meta. But these channels create demand, they do not harvest it. They introduce people to your brand, who then convert later via a direct search or through Meta. That conversion gets attributed to Meta or Google, while TikTok or YouTube was the real trigger. You cannot see this with platform ROAS alone. With a MER analysis you can.

  
  

Concrete steps to improve your measurement

1. Set up a weekly MER dashboard. Pull your total revenue from Shopify and your total ad spend across all platforms. Divide the first by the second. Do this every week at the same time and you have an immediate, reliable reference point for the health of your marketing.

2. Track your NC-ROAS separately. Monitor your first-purchase revenue through Shopify customer tags or through a tool like Triple Whale, which segments this automatically. Set a target for your NC-ROAS based on your gross margin and the average time it takes a new customer to place a second order.

3. Use a cross-channel attribution tool. Triple Whale and Northbeam are the most widely used options in the DTC world and are in our experience indispensable once you are active across multiple channels simultaneously.

4. Allocate your budget based on incrementality, not reported ROAS. The question is not which channel reports the highest ROAS. The question is which channel generates the most incremental revenue that would not have existed otherwise.

The bottom line

The ROAS in your dashboard is not a deliberate lie. It is a number that measures what it measures, but what it measures is not what you think it measures.

The brands that scale profitably to €1M and €2M per month are the ones that learn to look at MER, NC-ROAS and incrementality. Not because that is easier, but because it tells the truth… and you can build on the truth.

Short Commerce is a performance advertising agency specializing in Meta and TikTok advertising for e-commerce brands. We help brands look beyond the dashboard and make decisions based on the metrics that actually matter.