Our ROAS Is 1.5x, Unless It’s Actually 3.1x 🤷🏻‍♂️

By George Ellis, Creative Director

If you ask Google or ChatGPT for the average Meta Ads ROAS benchmarks, the answer will be anywhere from 1x to 4x.

Which is the opposite of helpful.

That spread is so wide, it’s meaningless. Even if you try to break average ROAS numbers down by industry or category, it’s only marginally better. When it comes to toys, for example, the spread is 1.9x to 2.7x. A profitable spend at the higher level may be a losing proposition at the lower level.

But before judging whether you have a good or bad ROAS, you need to understand what your actual ROAS even is.

We currently run ads for Roasty Buds Coffee. It helps that we started the company and control all the creative. But even when the ads seem to be performing well, there are a variety of numbers we have to wade through. Over the past month, our attributed Meta ROAS, according to the platform itself, is 1.5x. That’s right in the pocket of average for food and beverage ads.

But that’s not our real ROAS. Or at least not the number we look at.

Because we sell coffee, a product people buy over and over. And also, that 1.5x is basically first-day ROAS. If you look at 7-day ROAS, some of our ad sets hit 2-4x, thanks to a strong email funnel.

Here’s how it works: First day ROAS is 1.5x. Meaning we spend $1 to bring in $1.50 in sales. Cool. Not gonna scale to the moon with that, but a nice base. Along with those sales, we also get a lot of emails of people signing up to our newsletter. Once we send those people emails, we get more sales in the first week or month. If it’s the first week, the platform will attribute those sales to the day we make the sale (not the day we captured the lead).

So let’s say an ad has a 1.5x ROAS on Tuesday. But then Thursday we get a bunch more sales via email from leads captured on Tuesday. Well, suddenly Thursday’s ads have a 3x ROAS!

Is real ROAS the platform number, the first-day number or the 7-day number? Yes. It’s all of those.

This can be confusing. If we didn’t know the impact of our email funnel, we’d assume Thursday was some massive ad ROAS day and we’d try to replicate it via ads or creative or targeting. But really, it was just the Tuesday ad bearing more fruit.

So, how can we judge the Meta performance?

Simple, take the longest view possible AND don’t just trust the platform numbers. Because Meta only really tracks 7 days after the initial ad. You can see 28 days with the right views and analytics, but even that might miss 10% of sales. And any time our funnel converts someone after 28 days? Not counted. Which isn’t ideal because as a coffee company, a customer who only buys once in the attribution window may go on to buy 10 more times that year!

To combat that noise, we look at Blended ROAS. How much did we spend in Q1 and what was our revenue? We subtract any organic sales or sales from other channels, and there we have an approximation of Meta ROAS.

Confused yet? It’s easy to get lost in the logic. We won’t even get into the fact we use Recharge to convert customers into loyal 5- and 10-time purchasers.

Which ROAS metric is the right one?
It depends!

• Meta Platform ROAS: Easy, quick reference to see how ads are performing in real-time
• 7-Day Trailing ROAS: A good option when you have a strong initial funnel set up to capture sales from leads in the first week
• 28-Day Trailing ROAS: The most complete metric for judging whether your Meta ads and overall funnel are profitable or at least capturing solid customers
• Blended ROAS: If Meta is your primary medium for advertising, a blended ROAS over the course of a month is a way to step back and see the overall impact of Meta ads on your revenue

The point is this: there is no one average ROAS. You have to really dig into the numbers to see if your ROAS is good or bad or…achieving whatever goal you set for it.