Balanced Testing: The Secret Weapon to Boosting Marketing ROI in 2024

11/8/2023 Timothy Pietrini

Balanced Testing: The Secret Weapon to Boosting Marketing ROI in 2024

Testing is to direct marketing what location is to real estate. It's the key to success. But not all testing is created equal. And if you're not testing the right things or in a balanced way (or testing at all), you’re most likely leaving higher response rates on the table.

If you are serious about improving response rates, adapting to cost increases, and boosting your marketing ROI, then the following five time-tested tips are something well worth your consideration — especially as you evaluate your testing approach heading into 2024.

Don’t get too “down”

As of late, we’re seeing a greater emphasis on running tests that reduce costs. We call this “testing down.” Typically we'll recommend our clients to have at least one test that adds something to lift that response (testing up) and one test that takes something out to reduce costs and try to maintain performance (this is in addition to running the control). 

Ninety percent of the time, you'll be more successful testing up than testing down. So with this increase in testing down, we’re seeing a disproportionate amount of losing tests. Our clients who are dedicated to maintaining a balanced testing approach are the ones experiencing lifts in response. Those who are trying to save themselves into prosperity are experiencing the opposite.  

Avoid over-testing

Typically 10% (and no more than 20%) of your direct mailing universe should be dedicated to testing. For example, if you’re mailing a million pieces, we would recommend no more than 200,000 of those million pieces dedicated to testing.

Why 10%? The reality is most tests will not beat your control, so over testing can actually expose a greater portion of your list to a lower performing mail piece. Heavy testing can also break up your control universe, which will most likely translate into raising your postage costs due to the need to create separate mail streams.

Leverage variable imaging to run free tests

In most cases, a physical change (e.g., format size, paper stock, special coatings or textures) to a mail piece will require a separate mail stream — and increased postage costs. However, variable imaging allows us to create multiple test cells within your control’s mail stream without increasing postage costs or compromising discounts. 

By using variable imaging to test different elements of your direct mail campaigns, you can fine-tune your messaging and design to increase your response rates and achieve better ROI. Here are some other examples of how variable imaging can be used to test direct mail for free:

  • Test different headlines
  • Test different offers
  • Test different colors and fonts
  • Test different layouts
  • Test different personalization elements

What works for one channel …    

What works well for one channel may not find the same success on another. Here’s a quick case study to better illustrate that point:

Our client, a large health insurance company, worked with us to build a series of direct mail tests that ultimately doubled their response rates. Meanwhile, the client was also finding tremendous success through television advertising. So much success, in fact, they wanted other channels in their omnichannel approach to mirror the look and feel (e.g., photography and messaging) of their television campaign. 

We tested a “television version” of the control — this involved approximately 30 different tests — and none of them came close to beating the control. This was a good reminder to the client that each channel has its own unique audience and engagement patterns, so it is important to tailor and test accordingly.

Spend more to combat a shrinking budget

When budgets come into question, direct marketers have a tendency to mail (or want to mail) their least expensive package. While this seems like it might make sense from a dollars and cents perspective, it really doesn’t. 

Similar to the “testing down” tactic I covered earlier, our testing experience has shown that the least expensive package is often synonymous with the poorest performing package. Instead, we recommend mailing your strongest (and more expensive) package to as many people as your budget will afford. To take this a step further, mail it to your top declies, because those are the people who are more likely to respond at an even higher rate to the more expensive package.  

For 2024, test like your marketing ROI depends on it

Direct mail testing is essential for any direct marketer who wants to improve their response rates, achieve better ROI, and ensure continued success. By following the five tips outlined above — and summarized below — your testing approach in 2024 will become more balanced, efficient, and effective.

  • Don't get too "down." Focus on adding to lift response and testing down to reduce costs. An emphasis on the latter rarely moves the needle.
  • Avoid over-testing. Typically, no more than 20% of your direct mailing universe should be dedicated to testing. Over-testing can actually expose a greater portion of your list to a lower performing mail piece.
  • Leverage variable imaging to run free tests. Variable imaging allows you to create multiple test cells within your control's mail stream without increasing postage costs or compromising discounts. Use it to test different headlines, offers, colors, fonts, layouts, and personalization elements.
  • What works for one channel doesn't necessarily work for another. Each channel has its own unique audience and engagement patterns, so tailor and test accordingly.
  • Spend more to combat a shrinking budget. Instead of mailing your least expensive package, mail your strongest (and more expensive) package to as many people as your budget will afford, especially your top deciles.


Tim Pietrini is the Vice President of Strategy for RRD’s Direct Marketing Group.

Contact Us