The only way to know if your loyalty or other guest-engagement campaigns are succeeding is to track them and analyze your results. Unfortunately, even numbers that appear straightforward can contain hidden implications.
Depending on how you look at your data, a campaign can seem like a roaring success or a misfire. But sometimes it requires further analysis to determine if your initial assessments are accurate.
In worst-case scenarios, you may discover that a campaign you decide to run again based on previous positive data had been a misfire the entire time. Not only are you producing unfavorable results, but you are also wasting time, manpower, and money.
For example, let’s say you set up a “We Miss You” campaign. These are fairly common loyalty program campaigns in which you send out an offer to people who have not visited your stores in a while to encourage them to return. After deciding that you’re going to set up a campaign for people who have not visited in 60 days, you send them a coupon for 15 percent off an entrée.
Once you’ve run the campaign and examined your results, the numbers look great. Many guests who had not visited in 60 days made visits during the campaign period and used the coupon you sent them. It seems like the campaign was a resounding success!
But was it truly successful? Sometimes results in campaigns like this are deceiving.
Within that group of guests who hadn’t visited in 60 days, there are bound to be people who consistently go more than 60 days between visits. Maybe they visit every 61 days or every 70 or 75. This means they were almost ready to come in anyway, and if they used your coupon as a result of the “We Miss You” campaign, you may be cannibalizing sales and losing money on this segment.
Also, remember that the period you’re using for measurement before that visit – those 60 days – had zero visits. Relatively speaking, the number of visits during the campaign can only go up. Some people in that group are sure to come in, and since you’re comparing results against a period with zero visits, your campaign will look like a success.
So, how do you prevent a problem like this? The best method is using test and control groups.
Instead of mailing to your entire lapsed list, mail to a portion of it, and at the end of the campaign, compare the behavior of your test (coupon) group against that of your control (non-coupon) group. If your test group came in more frequently than your control group, you may have a winning campaign on your hands. But if the test and control groups visited at about the same frequency and spent about the same amount, your campaign didn’t make much of a difference.
Testing helps you validate your results, and testing with a smaller segment first helps keep you from wasting (or losing) money on a bigger campaign.
Of course, this is just one example. There are inherent data flaws in the way many companies run visit challenges and other popular guest-engagement campaigns. The problem is that these flaws are hard to detect and impossible to fix once a program has launched.
If you’d like to make sure your programs are built correctly and you’re interpreting your data in the right way, we encourage you to watch our free on-demand webinar called “How Your Guest Data Can Trick You.”
Lee Barnes, the head of our Data Insights team, takes you through some of the biggest mistakes that companies make when setting up campaigns and analyzing data, and he shows you how to avoid having the same flaws in your own campaigns.
Check out our free streaming webinar before you even begin brainstorming your next promotion to ensure that your data and your revenue are in line for success. Click here to access the presentation.