Recency/frequency/monetary value modeling, or RFM, is a tried-and-true segmentation tactic for catalogers and other mail order marketers. But e-mail practitioners have been more reluctant to use it to segment their campaigns.
“I think the low cost of e-mail has made some e-mail marketers lazy,” says Ross Kramer, CEO of e-mail marketing firm Listrak. “It’s been so darn effective and so darn cheap that we haven’t had to apply a lot of critical thought to it.”
As e-mail response rates decline, however, amid increased competition and overmailing by some brands, more marketers may start analyzing their files by RFM so that they can target their most-recent, most-frequent, highest-spending customers differently from those who haven’t made a purchase in quite some time and who spent very little when doing so.
Online retailer Discount Beauty Center recently worked with Listrak to create a modified RFM model to reactivate customers who had not made a purchase in at least 30 days. After analyzing the records on its e-mail file by recency and frequency, it broke out past purchasers into three segments: those who had not made a purchase within 30-60 days, those who had not bought in 61-360 days, and those who had not made a purchase in more than 360 days.
It then applied what Kramer calls a discount ladder. Those who had ordered most recently—the 30-60 day group—were sent an e-mail offering them $5 off a $25 purchase. The next group, those who hadn’t purchased in at least 61 days but within 360 days, received an e-mail entitling them to 10% off their order. The coldest group, those who hadn’t bought from Discount Beauty Center in at least 361 days, were e-mailed a code good for 15% off their next purchase.
While the company didn’t make monetary value a factor in its model, it did include type of product, so that it included in the e-mails references to the type of items the customers had last bought. For instance, if a customer’s previous purchase had been a Ralph Lauren fragrance, the e-mail’s messaging included links to Ralph Lauren products.
Discount Beauty Center sent a follow-up e-mail to those who did not respond to the initial message. Although the subject line was tweaked, the message creative and the offer remained the same, which Kramer says kept the cost of the redeployment to a minimum.
The average click-through rate for Discount Beauty Center’s e-mails is about 1.5%, according to Kramer. The click-through rate for the initial e-mails sent to the coldest segment was 2.8%, with 12.0% of those who clicked making a purchase. For those who hadn’t bought in 61-360 days, the click-through was 4.4%, with 21.3% of those making a purchase. And among the most recent of the nonbuyers, those who hadn’t bought in 30-60 days, the click-through was 8.9%, with 13.6% of those responders making a purchase.
“What still amazes me,” says Kramer, was the response to the follow-up e-mails sent to those nonbuyers who had not responded to the initial offer. Among the coldest buyers, the click-through was 2.4%, with a conversion rate of 14.4%—a higher conversion rate than for the initial e-mail to this group. Among the middle sector, those who hadn’t bought in 61-360 days, the click-through rate was 3.3%, with a 25.5% conversion rate—again, a higher conversion figure than for the first e-mail. The redeployment was least effective among the most recent of the nonbuyers, with a click-through rate of 4.8% and a conversion rate of 10.0%.
Discount Beauty Center is continuing with its RFP (recency/frequency/product) model on an ongoing, automated basis. “As soon as someone falls into one of these buckets, they will get an e-mail autodeployed,” Kramer explains. “That’s the real ROI power from RFM automation.”
The other contributor to return on investment is that Discount Beauty Center doesn’t lose margin by offering discounts, or larger than necessary discounts, to customers who would have bought without the monetary incentive.
Discount Beauty Center is also using recency and frequency modeling to power an automated replenishment program. Having analyzed past behavior to determine what Kramer describes as “the average latency between purchases for each product,” the company will send a trigger reminder e-mail once the customer exceeds that average time. “Once they fall outside the standard deviation,” Kramer adds, “we start sending discounts.”
RFM, RFP, and similar segmentation strategies are hardly new. “Back in the ‘60s you wouldn’t have thought of running a direct mail campaign without some sort of segmentation or modeling like this,” Kramer says.
But as he sees it, RFM isn’t just a tried-and-true tactic; it’s also ties in to one of the more recent e-mail marketing buzzwords, relevancy. “A lot of the industry has been talking about relevancy, but it hasn’t given people a lot of practical insight into how to do it,” Kramer says. “We haven’t done a lot of models for it either.” Segmenting customers by how recently they made a purchase and how big a spender they are or what they bought is one way to start.