A fundamental part of subscription-based retail is the personalization aspect – where per-purchase retailers have the opportunity to use browsing data to personalize product recommendations (though according to Marketingland, only 13% of retailers take advantage of this information!), subscription-based retailers need to learn what they can about their customers before the first product delivery, in order to strengthen the brand-user relationship.
To dig into this more deeply, we created accounts with six different subscription boxes to see how they were handling early-stage data-collection. In the interest of exploring both ends of the spectrum, we looked at brands with data-collection processes that were excessively time-consuming, as well as some that were so short as to be almost incomplete, before landing on a process that was “just right” – here’s what we found:
Too long = “Survey Fatigue”
We created an account with a beauty subscription that delivers a monthly box of cosmetics based on the customer’s hair/skin type and makeup preferences. The brand asked us to fill out a survey of 14 questions – some of which had multiple parts to them and were unnecessarily granular, like “Is your hair dyed?” alongside “What is your current hair type?”. It’s unlikely that most customers will be willing to do this much work in order to receive a box of products that they haven’t even picked out themselves, and providing customers with such a lengthy survey right out of the gate is likely to lead to abandonment at sign-up.
(Unfortunately, it’s also worth mentioning that in this particular instance, the retailer still managed to suffer on the personalization front – despite having almost too much information about our customer preferences, the box we received wasn’t at all relevant to the products we’d expressed interest in and the beauty problems we reported wanting solutions to.)
Too short = Missed signals and missed opportunities
A popular subscription for dog-owners only asked us how big our dog was, then took us directly to the billing and payment section of the website. While a dog’s size tells you a decent amount of information about what kinds of products might be most suitable, we were surprised that the brand didn’t ask any additional questions about what kinds of treats, toys, or apparel the consumer (whether that’s the shopper or their pup!) would most like to receive. The box we received not only contained two products that weren’t suitable for large dogs (a small hat and a very flimsy toy that wouldn’t stand up to a larger breed), but the brand missed an opportunity to learn just a little bit more about us in order to optimize their personalization efforts.
While it’s certainly worth thinking about how much internal bandwidth each brand has in terms of personalization, each question that a customer is expected to answer – or, to put it differently, each time you ask a customer to do extra work to get their product – should be put to maximum use, to deliver more memorable shopping experiences.
What’s “just right”?
A different beauty subscription asked us to fill out a profile that, in addition to basic demographics (and our birthday!), asked 10 questions regarding our beauty lifestyle and general cosmetics preferences. These questions hit a happy medium of being specific but not unnecessarily so – questions were geared towards making sure the box’s contents fit our current lifestyle, such as “What’s your level of beauty knowledge?” and “How do you style your hair?”
This brand struck the balance in terms of survey length as well – enough questions for them to form an accurate picture of who I was and what kinds of beauty problems I needed help with, but not so many that I became exhausted and left without purchasing (as I badly wanted to do with the other beauty box).
The most popular solution has been to offer customers the option to answer several survey questions, in order to learn as much as possible about their interests and preferences. But there’s a delicate balance retailers need to strike – a survey that’s too long might be a turnoff, and a survey that’s too short might not be enough for you to deliver relevant recommendations. Understanding what kinds of methods are most conducive to learning about your customers while still helping them move along the purchasing path is critical to seeing more of your customers completing purchases.