Clario Marketing Analytics
Request a demo

Evidence-Based Marketing

How Return on Ad Spend Kills Retail Giants and Newcomers

Posted by Matt Redlon on Jan 7, 2020

Photo by chuttersnap on Unsplash

Is high return on ad spend really getting you the business results you need?

What should the goal of your marketing be? Adidas realized they wanted to prioritize marketing effectiveness over efficiency. You may wonder, what’s the difference? Or, does it even matter? The truth is it matters more than you know. 

Adidas admitted they thought digital advertising on desktop and mobile were the keys to driving sales. This pushed them into a situation where they were over-investing in paid ads to bring up a marketing metric, but this effort wasn’t building their brand. 

Since then, Adidas has made a 180-degree change on how they measure marketing, taking a long-term brand building approach. This isn’t to say that metrics like return on ad spend (ROAS) aren’t important. Let’s talk about ROAS and how you can measure your marketing more effectively. 


ROAS is important but isn’t the golden metric

ROAS is the marketer’s version of return on investment. It’s a fundamental way to decide if the way you're spending the marketing budget is yielding fruits for your company. The simple equation is revenue / spend = return on ad spend

Companies are not wrong for wanting to use ROAS to measure their marketing effectiveness.  The problem arises when they or their partners start to manipulate the data points to get a more favorable outcome. 

One way that happens is by over-attributing revenue to ad spend. Imagine that your company makes $1 million in revenue, and you spend $100,000 on marketing. If we take this at face value, it looks like your ROAS is $10. But, what if $800,000 of that revenue was generated through wholesale activities not related to the advertising you did? Suddenly ROAS drops from $10 to $2. Attributing all of that revenue to ad spend makes marketers look good but pushes the company to invest in the wrong activities. 

Another way marketers manipulate ROAS is by double counting - or attributing the same revenue to multiple  buckets of ad spend. Let’s say that I have email, direct mail, and digital marketing. Each of these departments knows what their ad spend is and attributes last-touch  revenue to their department. 

What happens when each of these departments is spending $100,000 on ad spend and all claim a $10 ROAS? As the CEO, you see that they're collectively spending $300,000 and wonder how they can all be claiming a $10 ROAS. In that case, your revenue should be $3 million, not $1 million. If organizations aren’t careful, ROAS can deliver bad insights and, as a result, drive bad spending. 


Using ROAS incorrectly incentivizes metrics over progress

Piece of paper with a graph showing improvement over time

Photo by Isaac Smith on Unsplash

It isn’t all your marketing team’s fault. It comes down to the organization’s top-level goals. Think about your performance marketing manager or agency marketing partner. If they are incentivized or required to get a $5 ROAS, it becomes necessary to prioritize metrics over progress. 

The marketing team’s goals change and don’t necessarily include acquiring or retaining a certain amount of customers. Instead, your goals incentivize them to insert themselves between the customer and a purchase they were going to make anyway — just to grab credit for their program.  

Bad incentivization is why our company typically advises clients to exclude branded searches from their measurement of ROAS in paid ad campaigns. It's common to Google the name of a business you’re looking for to find their website. So, customers performing branded searches were probably already going to your site. 

Focusing on branded searches is especially deceiving because this group is likely returning customers with a very high purchase intent. As a marketer, this allows me to blend the cheap $0.22 click with the more expensive $4 click of trying to turn new people into customers. This causes you to pay for ads that will bump up your ROAS number but do little else to grow revenue.


More effective marketing results come from more effective measurement 

Your marketing results are tied to more than a single metric. To effectively measure marketing’s progress, you need to tie metrics to your organization’s goals and help your team view those metrics in the right way. If you’re framing them in the context of ROAS, you incentivize marketers to gamify the stats. Instead, focus your organization on metrics like customer lifetime value, customer acquisition, and customer retention rates over a period of time. These measurements relate to the lifetime health of your brand. 

The difference between these types of metrics and ROAS is that when you consistently increase these numbers, revenue is going to rise. They bring tangible benefits to the organization that can't be manipulated or incorrectly measured. 

Companies can use ROAS, but they have to be strategic with it — use it as a directional signal. One way to do this is by effectively controlling the time period for attribution. For example, if you run an Instagram campaign tomorrow, are you attributing all last-click sales made that day to the campaign? A better attribution model would be to attribute any-touch sales over a lagged period of time so that it better reflects how an awareness campaign works. 

ROAS isn’t evil, but it can do more harm than good if used incorrectly. By setting the right organizational goals and using other important metrics in conjunction with ROAS, you get insights that help you disrupt the market. If you would like help interpreting and understanding your marketing data, talk to one of our experts. They’d love to help you make marketing decisions that grow your brand and business. Set up a meeting today

Matt Redlon

Written by Matt Redlon


Subscribe to our blog

Join over 3,000 retail leaders and get access to news, ideas, and best practices.