Breaking down the ultimate ad measurement tool: Return on ad spend.
There is no shortage of ways to measure whether an advertising campaign was effective.
Some campaigns, like cost per acquisition campaigns, measure whether consumers took a specific action, such as clicking through to the brand’s website, signing up for an email newsletter or redeeming a coupon. Gross rating point, a measurement standard typically used for TV, captures how effectively and how often an ad campaign reached its intended target market. Brands will sometimes conduct brand lift studies to gauge whether a campaign positively influenced how consumers view their company. And there are a slew of ad tech companies that specialize in measurement and peddle their proprietary sales effect metrics.
Arguably no metric is as important as return on advertising spend, though. ROAS, for short, is the granddaddy of ad metrics, in that it measures how a brand’s advertising affects its revenue. Advertising is, at its heart, about increasing sales, so connecting ad spend to sales has incredible value. But while ROAS might be easy to understand conceptually, determining it and its implications can be complex. Here’s how it works.
First, how is ROAS calculated?
Calculating ROAS is straightforward. You take the revenue generated by an advertising campaign and divide it by the cost of that campaign. Voila, you’ve determined the ROAS for the campaign.
ROAS = (revenue) / (ad spend)
Though ROAS is technically a percentage or a ratio, it’s often expressed as a dollar amount. For example, if a campaign cost $100,000, and it generated $300,000 in sales. You could say it had a 3-to-1 or 300% ROAS, and both are technically correct. But it’d be more common to say the campaign had an ROAS of $3.
Okay, but how do we know the sales were caused by the advertising?
An astute question. You’ve touched on the fundamental issue with determining ROAS.
If you’ve taken a 100-level statistics course, you’re familiar with the phrase “correlation does not imply causation,” meaning that just because two variables seem to fluctuate in tandem with each other doesn't necessarily mean they have a direct, causal relationship. The change can be caused by some unaccounted outside force.
The same applies to advertising — it’s extremely difficult to determine whether an increase in sales was a direct result of an advertising campaign. There can be other factors at play, such as seasonality. Beer brands tend to step up their advertising during the summer, for instance, and beer sales increase in the warm weather months. But it’s hard to tease out how much of the sales lift is caused by advertising and how much is due to people spending more time outdoors, hanging out at barbecues, cooling off with a frosty brew.
What’s the solution?
There isn’t a simple one, especially considering that advertising campaigns often exist across so many different channels.
Using the beer example again, let’s say a big, iconic beer brand runs an equally big, multi-channel advertising campaign built around the 4th of July holiday in the U.S. The campaign includes commercials during MLB broadcasts, sponsored posts on Instagram and banner ads across a large number of websites. Sales increase dramatically around the holiday, and the marketing team at the beer brand considers the campaign a huge success.
The challenge is determining how much of the sales lift should be attributed to the TV commercials vs. the Instagram posts vs. the banners (and that’s before you take seasonality and other outside factors into consideration).
That is complicated…
Indeed. That’s why there is an entire cottage industry built on measurement and determining ROAS. It’s hard stuff, and it often involves highly scientific studies that include lots of different data points. Everything from coupon renewal codes, to click-through rates for banner ads, to brand affinity studies to Nielsen TV ratings are used to help calculate ROAS.
It brings up one of the most pressing and interesting topics in advertising: attribution. Attribution, or more accurately, multi-touch attribution, attempts to parse out how influential the different components of an advertising campaign are and how they complement each other. If a consumer sees an Instagram ad for a pair of shoes, and a day later sees a banner ad for the same pair, and makes a purchase, attribution aims to answer how much of the sale should be attributed to each ad.
I’ve heard the term ROI before. Are ROI and ROAS similar?
ROI, short for return on investment, is a more general business term designed to measure how all kinds of expenses affect a company’s bottom line.
ROI = (profits – costs) / (costs)
Some advertising experts say that, given the multi-channel nature of modern advertising, ROI is a better way to measure advertising’s impact. Sometimes people use the terms interchangeably, which only adds to the confusion (yay).
As you can see, ROAS is more complicated than it initially seems. But just concentrate on delivering an ROAS of more than $2 a campaign, and you’ll be fine. And just as the data gets more complicated, there are always ad tech solutions that will help you make sense of it.