How to Measure ROAS for Your Video Campaigns in 2025

For years, we’ve relied on one metric to measure advertising success on social media or Google. It’s a metric fed to us by the ad platforms, and that is – Return on Ad Spend – or ‘ROAS’.

In this blog, I will talk about the background to ROAS, its lack of subtlety, and — importantly — some newer metrics and a much more nuanced approach you should think seriously about.

ROAS, a metric that has been around longer than the platforms themselves, was always calculated by the advertiser. It was a simple equation: How much did I spend on that ad, and how much did I make in return? However, it was often difficult to measure accurately, especially regarding individual ads.

The buyer’s journey could be complex. They might have heard a radio ad, driven past a billboard, or been recommended by a friend. ROAS was generally measured by lumping everything together. This historical context is essential to understanding the evolution of ROAS.

Sure, you could always ask the buyer, but it was always going to be just an estimate.

Then along came social media ads and Google ads, and it was easy to work out your return if you had your conversions set up correctly. ROAS was the North Star — a number for agencies to sell you on and for them to point at, to tell you they had been doing their job. Of course, measuring in the app meant adding conversion values and telling the app how much you made from each sale.

Then we could opt out of cookies. Here, our iPhones stopped tracking some clicks, and the buyers became more sophisticated and wary of sponsored posts or Google ads. They often wouldn’t click on ads, but leave the platform and do a brand search for the company that was doing the advertising. I know that I don’t click on ads on Facebook or LinkedIn because I know that they are going to want my details straight away, without me knowing anything about them. Why, then, do I expect other people to click on my ads when I won’t do it myself?

As a business owner, there is only one metric that I need to be looking at regularly, and that is the company bank balance. Is it going up? Or is it going down?

When I relied on in-app ROAS, and when the ROAS was good, I was never able to scale up reliably. Clicks became more expensive and infrequent the more I tried to scale up. I just wanted to be able to double my ad spend, for example, to double my clicks, but it never worked that way. It felt like the law of diminishing returns was at play.

Think of two lines on a graph: ROAS and ad spend. The more money you had sunk into the platform, these lines would get.

I wanted my two lines to be perpendicular and my clicks to stay the same price no matter how much I spent.

Google ads, for example, worked well for us … until they didn’t. We would get to a certain level of success, I would throw more money at it to increase that success, but it would then become untenable because the cost per click on the ad would start to blow out.

I started to look at the data, though, and started experimenting. The first thing I did was leave my conversions in place, but I removed the value attributed to a conversion, and my per-click costs instantly went down, often to about a third of the suggested bottom-of-page cost per click. Then, if I added more money to the campaign, the cost per click was the same. Weird, right? It’s an auction, and I know there are several factors at play, but I was winning the auctions with a much lower price.

Google, of course, was trying to keep me interested, trying to keep my ads ticking over, and without a conversion value it didn’t know how valuable those clicks were to me, so it got me more clicks.

My in-app ROAS started to go up again, but I had to measure it outside the app.

The real question was, however, ‘Was its Google ads that were getting me the conversions?’

Yes, in this instance, Google ads were the last touch in what was potentially a very long chain of events. Google ads chalked up a win, but it can’t take all the credit.

For people to purchase our service, they need to trust us. They need to know who we are and what we stand for. In a world of fake news, AI, and inauthenticity, potential customers will need to spend time with us (virtually, of course, to start with) before they will even contemplate spending thousands — or tens of thousands of dollars with us.

No one is going to impulse buy what we must sell, or even fill out a form based on one Google ad, or one Facebook ad.

Advertising and Marketing for us is no longer a single platform approach. It’s a multi-ad platform approach every time we launch a new campaign.

If we are advertising on LinkedIn, we are also advertising on Google; if on Facebook, we will also use TikTok. We cannot look at ads in isolation anymore.

And we are not just multiplatform; we are also multi-step.

We might have three ads running in one campaign on any one of the channels: one for prospecting, one for engagement, and one for remarketing. If you watch the first 25% of our video in the first part, you will get served the second. Watch 25% of the second, and you will get the third. Depending on the campaign, you might then see the same ads (obviously corrected for the different platforms) on other platforms that you visit.

We want you to consume our video content on every platform that you are on and then once you have seen enough to get us over your own internal “slight trust” metric, you might search for the name of our company through Google and clicking organically, or through our Google ads brand campaign.

We also want you to consume as much content as you possibly can, and you can’t do that with photos. For us, the answer is obviously video. A few years ago, we might have said, “Make your video content as short as possible because people have short attention spans”, but now if you have a long video that is engaging to the right audience on the right platform, there is every chance they will watch a lot more of it.

Think about the videos you consume. You don’t just watch 15-second clips on Facebook. If it’s the right topic, delivered engagingly, then you will watch a video that is a lot longer (often without sound, lying in bed at night next to your partner, which is why captions are so important … but that’s a topic for another blog).

So, why do we conduct all these multi-platform, multi-ad, often long-form video ad campaigns?

In our example earlier in the blog, we talked about a click on a Google ad leading to a conversion on the website. So, do we just look at Google ad spend to work out the ROAS?

There is every chance that that lead came from someone watching three of our videos on Facebook, seeing a clip-on Instagram, reading our Google reviews from the website, having a think and then Googling us again, which eventually led to the conversion. So, sure, it was Google ads that got us the lead, but it wasn’t the only thing.

ROAS, therefore, must be calculated at a wider level.

Gross Revenue (all Sales) / All Sales/Marketing expenses (including salaries of the marketing team, SAAS products, Ad Spend) = Return on ad spend.

Some people are calling this new ROAS, MER or Marketing efficiency ratio.

There are also some other interesting acronyms/equations that you can perform to help your business grow.

nCAC – New Customer Acquisition Cost

Sales/Marketing Expenses / Number of New Customers (over a given period) = nCAC

This shows the average cost to acquire a brand-new customer. (It can help you work out a marketing budget per customer, or the amount you are willing to spend).

nAOV – New Customer Average Order Value

Total Gross Revenue from brand new customers/number of brand-new customers = nAOV.

Total gross revenue from new customers over a year – $100,000 with 50 new customers

nAOV = $2000

This, again, helps with a marketing plan and marketing budget. Is your nCAC more than your nAOV? Maybe that’s OK because your LTV (the next metric we will look at) is much higher.

AOV is the average order value and looks at all orders instead of new customers.

LTV – Customer Lifetime Value

(Average Sale Value x Average Number of sales (in a given time)) x Average Number of time periods = Customer Lifetime Value

For example, the average Sale Value might be $2000.

The average number of sales in a year (per customer) – 6

Number of years that a client stays with you on average – 2.5

(2000 x 6) x 2.5 = 30000

Is it worth spending $3000 to acquire a new customer who might spend only $2000 at first when the average lifetime value is $30,000? Every business is different; speak to your accountant or bookkeeper.

There are many more equations you can use to see if your marketing is working. Let me know if you would like to learn more.

In today’s complex digital landscape, relying solely on in-app ROAS is short-sighted. It’s time to look at the bigger picture. Adopt a multi-platform, multi-step approach that recognises the intricate buyer’s journey. Long-form video content, strategic ad placement across platforms, and a holistic view of your marketing efforts are crucial. Remember, your company’s bank balance is the ultimate metric.

By considering nCAC, nAOV, and LTV alongside a more comprehensive ROAS or MER calculation, you’ll gain a clearer understanding of your marketing’s true impact. It’s time to move beyond outdated metrics and embrace a more nuanced approach to measuring marketing success.