In the past few weeks, we’ve started exploring how strategic-level choices around loyalty programs impact redemption rates, average order values and repurchase rates.
Hopefully, it’s helped you think about your program’s structure.
One of the pieces I wanted to explore more, though, was how redemption rates and average order value lifts combine to deliver incremental revenue.
As a reminder, we’ve found:
Orders where loyalty program customers redeemed percentage-based discounts have a 47% higher AOV than orders where returning customers don’t redeem a loyalty reward.
Participation rates are highest among variable discount programs, which allow for a lower barrier to redemption.
The follow-on question, then, is figuring out what balance is better for your loyalty transactions: Does a higher participation rate offset the lower lift in AOV? Or, might it be better to keep participation lower in exchange for significantly higher AOV?
I built a very simple model (available for you to use here) to parse this out:
Though percentage-based discounts have the lowest redemption rates of any program type we analyzed, their impact on average order value so far outpaces the other core reward types that it’s still a clear winner—if you’re optimizing for average order value.
There are, I think, three takeaways here:
First, benchmark your loyalty program by pulling performance metrics from your loyalty program dashboard. Understanding how your program is delivering value—relative to others—can help you with the next two takeaways.
They are:
Your program structure matters.
Merchandising your program might matter more.
While changing the structure of your program may not be at the top of your list, finding the right structure for your program can be easier if you view it through the lens of what you can expect (based on other programs).
If you’re greatly trailing benchmarks, it might be worth adjusting at the strategy level before you adjust at the execution level (why spend more time telling customers about your program if it’s not set up to deliver at least average returns?).
If, however, you’re relatively in line with benchmarks, the execution level matters a lot more.
As an example: say you’re sitting around a 2% redemption rate, but doing little to promote your loyalty program aside from getting customers into the program.
Redemption rates are probably that low because there’s little merchandising around the program. And though we haven’t proven this with any data—yet—it feels like that’s probably the metric that’s easiest to move.
As I mentioned a few weeks ago, when we started talking about loyalty in this newsletter:
It’s no secret that many in our industry have grown skeptical of loyalty programs…
The reason for that, I think, is that the loyalty program space began to promise something that wasn’t defensible. Selection bias is a real hurdle in proving a loyalty program’s impact, and many players in the space haven’t wanted to expend the effort to clear it. And I don’t think they’re alone. Plenty of brands don’t want to spend the time necessary to clear that hurdle, either.
Merchandising is the solution.
Take this example from Mexicali Blues, which serves almost as a second confirmation email and encourages account creation after purchase:
Because the email is coming after a purchase, it has a 44% open rate and a 9% click-through rate, creating a strong win in the merchandising category.
By bringing loyalty program reminders (like points totals, discount availability, and points expirations) into SMS and email, you can begin to readily measure the impact of your program on behavior—be that conversion rates, average order value, basket size/mix.
We will be continuing to test this execution-level work with our customers, and sharing results here. If you agree with my takeaways from the data above, and want to partner together on those tests, let me know.
We’d love to talk.