insights. vision. productivity. knowledge. fun.

Taking a Fresh Look at ROAS

ROAS

Return on Ad Spend, or ROAS, is a common metric to measure campaign performance, especially in direct mail campaigns. To calculate ROAS, divide the revenue generated from a campaign by the cost to execute it. This metric is easy to calculate, easy to explain, and resonates across stakeholder. However, we previously discussed 3 reasons brands should move away from ROAS. Optimizing ROAS has little to no impact on top line sales, and excludes fringe and lapsed customers. It is also not an ideal metric to compare performance across channels.

The first step in moving away from such a heavy reliance on ROAS is being able to trust data. All stakeholders need to be able to understand and trust lift metrics, future projections, and advanced analytics. These metrics are not as explicit from a revenue standpoint, but it’s key to have internal buy-in that data and advanced metrics will be more valuable in the long term.

Move Away from ROAS

Once you have overcome that hurdle, here are 3 things to consider when moving away from ROAS:

  1. Don’t Rip Off the Band-Aid. Marketers have been using ROAS forever, so internal stakeholders are used to hearing about optimize it. When you start off, you need to continue mailing to groups that generate high ROAS. As you are moving towards an incremental metric, you will notice your ROAS drop significantly. This may be uncomfortable to see, but you need to trust that the data and your tests are driving revenue in the long run. In the meantime, you still need to prove that your campaigns are profitable in order to get the funding you need.
  2. Test and Control. While you are still mailing groups that generate high ROAS, you should begin to use test and control groups. Test small groups of lower ROAS customers so you don’t completely dilute the existing metric that people are comfortable with while you learn about the relationship between ROAS and incrementality. Find a small group of people you aren’t currently sending to, and measure both the ROAS and lift on this group. Compare those metrics to the group of high ROAS customers, and start to identify opportunities to optimize true incrementality.
  3. Frequency Testing. Define long-term hold out groups to establish frequency testing and the long-term impact of receiving mail beyond the matchback period. You should try to find the optimal frequency for all your different customer segments. (Hint: Customers probably don’t need a direct mail piece every month.) For example, try sending best customers 6 catalogs a year, and sending low-end customers 3 catalogs a year. It will take more time to collect the data, but you will be able to find out what works best for each segment for their long-term customer journey.

Putting it into Action

At the end of the day, you need a marketing database system that allows you to pull test and control groups, manage their journeys and measure the results over a period of time. In addition, you need the skill set and marketing analysts to closely monitor and report on these new metrics.

The most important thing to remember is to be patient. This change in mindset, and the results, will not happen overnight. It may not even happen in 3 months, or 6 months. To see incremental change, especially in direct mail, you may need to wait 12-18 months. While incremental results may be less intuitive, they help paint a more accurate picture that effects top line revenue across the entire organization.