Big Moose Is Watching You
When it comes to smart use of customer data, L. L. Bean makes other retailers look like rubes.
--
A century ago, in Freeport, Maine, a shrewd outdoorsman named Leon Leonwood Bean decided to dox the guts out of some out-of-state hunters in hopes of taking their money.
Dox, that is, in 1912 terms. In that steampunk framework, doxxing meant that Bean got his hands on a mailing list of men who held nonresident Maine hunting licenses. These men were sitting ducks — if you will—to his pitch for rugged field-and-stream equipment that allegedly worked wicked well in the wet-dry terrain Way Down East. A lifelong salesman who as a kid sold steel traps and dead deer to anyone he could find, Bean had acquired the salesman’s holy grail: a list of near-perfect leads.
Data Asset Number One.
Bean then devised a four-page piece of direct-mail marketing and used it to cajole the listed hunters into buying, sight unseen, some ugly-duckling lace-up boots that were part rubber and part leather. You couldn’t be a true deerslaying Mainer without them. Bean’s cobblers ginned up these would-be waterproof “hunting shoes” — almost as an afterthought. Like salesmen immemorial, Bean put his faith not in his goods, but in his leads and his rhetoric, which made the Maine Hunting Shoes sound revolutionary before they even existed.
Bean’s one truly inspired touch that first time out: If his customers were, for any reason, disappointed with the boots, they could get their money back — or a brand-new pair. Disappointed they were. Appalled. Fully 90 percent of the buyers of the first Bean boots wrote to say they hated their boots — which crumbled after a few jaunts — and they wanted a refund.
Unfazed, Bean scored some decent rubber, went back to the cobbling board to produce the iconically homely-and-somehow-also-dashing hunting boots, the Bean boots, that didn’t fall apart. He shipped them. The customers were grateful for Bean’s alacrity, accessibility and willingness to accommodate their criticism. More than grateful: as customers, they married the brand. The boots too were a hit, and L.L. Bean was born.
L.L. Bean is nominally an outfitter, trading in (these days) well-made white-flecked navy wool sweaters, a staple of my New England adolescence; canvas tote bags, many with monograms; and of course the unmistakeable leather-rubber boots that have made a style comeback among hipsters and in a Red Sox-themed limited 2013 edition—as well as heaps of other stuff, including recessively beautiful cashmere crewnecks and Primaloft compressible outerwear in California colors that Leon Leonwood Bean probably never beheld.
But from the beginning L.L. Bean, as a catalog business, distinguished itself as a customer-service juggernaut and — above all —as a stealth data enterprise, claiming as its core asset massive amounts of information akin to that first priceless list of nonresident Maine hunters. The boots did or didn’t work, remember: Bean was initially unconcerned about the durability of his 3D product. Once he had buyers, and knew what they needed, he could always fix the boots. But his data and his marketing — that was where his treasure was. And it worked like a charm. Superb data, superb marketing, boots in beta: That was enough to found a company on.
So it should have been no surprise — and still it was — to see mild-mannered Chris Wilson, Senior Vice President of something called Direct Channel at L.L. Bean, at the Javits Center in Manhattan this month, discussing data. Many in the audience — programmers from around the world — had never heard of the Bean boots. But they knew about the company’s ravenous appetite for data. Specifically, they packed the standing-room-only Bean session at the Strata + Hadoop World “Make Data Work” conference to hear about L.L. Bean’s 10+ TB on-premise enterprise data warehouse and its newer deployment of (still more extensive) cloud data, fully 100 TB, which can be collected and used in realtime by customer-service reps on the phone, online and in stores.
Also on stage was Doug Bryan, a data scientist from RichRelevance, which has partnered with Bean to a create a data-centric, single view of each online Bean consumer, as lavish and lifelike as a portrait by John Singer Sargent. These data profiles are in turn used to create “relevant experiences” for each customer, i.e. dead-on marketing targeted as if by a sharp-shooter.
You’re the kind of person who likes multi-ply cashmere and you ship dozens of Christmas gifts to Europe for senior citizens every year? Wherever such data surfaces — through whichever of the many channels in the Bean sales apparatus — it can be incorporated into the digital portrait of you as an L.L. Bean customer.
Evidently the company has been doing “consumer modeling” since the 1960s: accumulating, accessing and analyzing customer data. And all that stuff, especially as the tools evolve, piles up.
The company now hosts Big Data Boot Camps in Maine, teaching marketing, I.T. and analytics teams how to use the sophisticated, nonlinear language for data storage and retrieval called NoSQL — Not Only Structured Query Language. In partnering with RichRelevance, L.L. Bean has also taken on a separate, non-retail mission: data for “omnichannel” presentation to customers.
A century since the founding of L.L. Bean we’re accustomed to retailers whose ace is their data. What’s devlishly cool about L.L. Bean is that — unlike known big-data leveragers like Amazon, J. Crew and Nordstrom — the company’s name (since the first bust boots) evokes durability, craftsmanship and attention to detail. That would sound like marketing palaver if it weren’t also true. In an era of floridly disposable clothes (Zara, H&M, Old Navy) that tend to burn bright and die young, L.L. Bean still trades in a modest ideal of “the satisfactory.” After the boots, its most famous offering has to be the limitless full-refund return policy it offers on all goods. You don’t even need a receipt. In some cases, you don’t even need to have the thing you bought in hand.
(This gonzo Bean return policy became the industry standard for outfitters selling rugged wear. But, last year, REI, the sporting-goods retailer based in Kent, Washington, punked out on the lifetime guarantee after feeling conned by shady customers, like the one who returned an old stroller because her kids outgrew it. They clipped their policy from lifetime to one year. EMS, an outdoor retailer based in Peterborough, NH, has a Bean-like lifetime policy, but asks that customers pay return shipping.)
This return policy — even for those who don’t exercise it — sets up a lifelong relationship with catalog and online consumers, who then oblige the company (and RichRelevance) with details about themselves, their habits and their preferences that go straight into the data mill where they can be churned into sales opportunities.
That data is, these days, now quickly retrieved by Bean’s canny heirs in marketing who use it to sell you mittens.
If this sounds ominous — and it does —there’s a big upside: the elegant uptake and deployment of Big Data (as with NoSQL) makes marketing less obnoxious. If L.L. Bean knows who you are (a thrifty huntress who likes turquoise, say) it can make more “relevant” offers, in the manner of Amazon, which bases its “recommendations” on your customer profile, your history and the history of customers like you — as well as keenly extrapolated data about what you might want. In a way that benefits both customer inboxes and marketers, the precise data means fewer off-base emailings. Our thrifty blue-green huntress doesn’t get email advertising splurge-worthy black running shoes.
Nor does she, in theory, get led to those shoes by an L.L. Bean clerk. His iPad, when L.L. Bean data is working well, tells him those pricey kicks are not her scene, and he instead shows her deals on the Easton Arrow Six-Pack (archery equipment being the rare hunting gear that comes in bright colors). This is possible because Bean and RichRelevance have set up a front-end interaction wherein salesfolk at call centers and in stores access data on their iPads through Apache Kafka, a “message broker” that handles worldwide transaction logs — data on purchases, non-purchases, returns, resoles, repairs — in real time. As Chris Wilson puts it, the meticulous use of Big Data, running at top speed, results in — of all things — customer “delight.” More relevant content, better shopping and better service.
Sure enough, last holiday season, L.L. Bean ended up tying Amazon as tops in customer satisfaction as an online retailer. And that was before the full Big Data rollout, and the engagement (this very week) of Erwin Penland, the hip, rootsy South Carolina ad agency, which will handle digital, retail, creative and some marketing strategy.
All of this, presumably, will raise the brand’s satisfaction marks still higher. This is extraordinary, considering that L.L. Bean is still perceived as a regional brand, and many in the multi-ethnic audience at the Javits Center had — by show of hands — never patronized.
Wilson seemed unbugged by the room’s partial ignorance of L.L. Bean. It may even have delighted him. Bean doesn’t spend much time or money waving its hands in sponsored links and Facebook ads, trying to create pointlessly ambient brand awareness or worm its way into the minds of people, like programmers in Lahore, who couldn’t and shouldn’t care less about New England style. Instead it cultivates the customers it has, and those a quarter-step away from that core, finding sometimes-unlikely new customers by the data only (as in Japan, where Bean is status brand), and lavishing attention on return shoppers as on members of an exclusive club.
As an occasional Bean shopper, I can testify that I get few promo emails, and online I get the feeling the site is my place. At llbean.com the company seems to carry just the well-made heritage things I as a discriminating shopper like, the same way the nonresident Maine hunters must have been flattered to get that 1912 mailing.
Finally someone understood their needs, could solve their wet-dry hunting problems and had their number.
Not until they were out in their boots, guns over shoulders, eyes peeled for deer did they discover who had really been shrewdly targeted, aimed at, hunted. But by then the boots were working just fine, and looked cool as hell, and it didn’t matter.
