How to Test the Efficacy of a Direct Mail Marketing Campaign
One of the key tenets of creating a truly effective piece of direct mail is testing. Marketers need to pay close attention to the results of their campaigns and to revise their strategies based on what they find. Operating blindly can cost your business both money and time. Marketers should be strategic about how they investigate the efficacy of various campaigns and create tests that can provide them with valuable, objective data about how to shape campaigns in the future. Testing is particularly valuable when small adjustments have been made to a piece of mail. Determining whether or not the change was worthwhile is extremely important.
What is the first step in testing a direct mail campaign?
Before marketers can begin to strategize about analysis, they need to develop a control sample. The ideal control piece depends on the specific position of a company. If no direct mail campaigns have previously been launched, then marketers need to focus on creating a piece that has a clean design, a great offer, and a compelling call to action. Once this piece is sent, the company needs to gather some baseline metrics related to your response rate and sales. These metrics provide a baseline against which future data can be compared. True testing cannot begin until you have developed a control piece and baseline metrics against which to evaluate the results of future campaigns.
In some cases, companies will already have some data related to several different mailings that they have sent. In this case, the best control is a piece that has performed well in the past. Again, companies should have robust baseline data for this piece. If not, it is important to collect some before proceeding. Sometimes, it makes sense to use a mailing that subjectively seemed less successful if there is data available, rather than developing a whole new campaign to create a baseline. Over time, marketers should generally always choose their top-performing piece as the new control to push themselves continually. However, it is also wise to test a piece in several different scenarios before adopting it as the new control.
What are the different strategies used to test campaigns?
Once a control has been established, marketers can decide on the specific type of test that they would like to run. In general, marketers use an A/B (split) test or a multivariate test. An A/B test may also be called a split test or a bucket test. This strategy involves sending out two direct mail pieces, a control and a variation. The two pieces should be fairly identical except for one key difference that the marketer wants to test, such as a shift in the offer, some reformatting, a different mailing list, or a rephrased call to action. By collecting data on each of the new pieces, marketers can develop an objective sense of whether or not the change made a real difference. This is especially useful when testing out a new list or determining whether a simple change alters response.
Sometimes, marketers have rather limited time between campaigns and a number of different ideas about how to improve upon each of them. In this situation, making one change may seem like an ineffective strategy. Instead, marketers can use a multivariate test, which involves changing several different components of a direct mail piece at the same time and then comparing them with the control. While this strategy can be an effective way to see large differences between different mailers, it makes it difficult to figure out exactly what caused one mailer to be more successful than another. Ideally, marketers should keep track of changes and create multiple variations for each one. For example, if a company wants to change the main image and call to action, they should have four variations: one with the new image alone, one with the new copy alone, one with both changes, and the original.
Developing so many different pieces limits the sample size of each version, so the results are less valid and further testing may be needed to confirm them. That means multivariate testing can sometimes involve quite an investment of resources. Still, multivariate testing can be useful for companies that are just starting out with direct mail and that need to get a quick sense of what works best with their particular customer base. The feedback provided by this technique can help to more quickly refine a strategy.
How should marketers keep track of the results?
The utility of testing depends on keeping track of the results of each run. The results that a company tracks largely depend on their specific goals for a given campaign. Sometimes it makes the most sense to look at the conversion rate. Other times, companies may want to focus on the response rate. Ideally, marketers should identify their endpoints prior to mailing out the pieces so that they can start tracking them right away and gather reliable data. Often, it makes the most sense to keep track of all endpoints and compare them. Marketers may find that a campaign designed to increase awareness actually ends up with a high rate of conversions, and recognizing this fact provides a lot of good data for how future campaigns should be designed.
