Field Notes: How E-tail Benefits from Digital Transformation
In two previous posts on the topic of "e-tail" ( "E-tail, Data, and the 'Smoosh'" and "API Architectures for E-tail") we discussed observations from customer visits on some of the drivers for digital transformation (the “killer app”, if you will, that conjoins ecommerce and retail), as well as the typical API architectures that companies use to get there. In this third post on e-tail, we’ll examine some of the sectoral benefits that result from these investments.
Apparel-focused retail: personalized experiences
Higher-end retail businesses that concentrate on (but aren’t limited to) apparel tend to have originated in the brick-and-mortar world, and gained strong footholds by creating a brick-and-mortar brand experience for customers (they tend to think of a store when they think of the brand). These retailers have also built significant ecommerce capabilities over the last 15 to 18 years.
To make some sweeping generalizations, these businesses generally share these characteristics:
- Continental in operation; one may be big in North America, or in Europe, but not both
At the higher end; brand and customer experience differentiates them from competitors
The typical customer in-store experience tends to be at a flagship store, or at a large, three- to four-floor satellite store
Keenly interested in the growth of the wealthy and middle class in Asia, particularly China, and looking to leverage existing capabilities to capitalize on this growth (and pay particular attention to brand in these locations, at the high end)
Facing an inflection point in the economics of in-person purchasing; to expand internationally, or even domestically, these businesses need to discern how to succeed in a larger number of smaller-footprint retail sites
Looking at multi-channel (sometimes referred to as “omni-channel”) as a critical new way to offer a consistent user experience across devices and to reach individuals with specific, targeted offers
To think about where these businesses want to go, consider what Apple did with its stores, and imagine a 21st-century version of that for selling clothes. The customer can start browsing at home, place items in cart, and, because of identity and cart portability, can walk into a small store downtown, 20 minutes or two days later, and try on a garment for size or color without having to repopulate the cart. The store has kiosks with very large displays (70 inches), and the sales associates know the customer’s name, and the color and style of the customer’s previous purchases the minute he or she walks in the door. All this is enabled by data—delivered, aggregated, and unified—via APIs.
For these businesses, API-enabled multi-channel improves the odds of keeping existing customers or snagging new ones (via a cool and differentiated customer experience), but it also changes their operational economics of scaling the business. They can expand with a large number of “small soldiers”—small-footprint, small-overhead, and less capital intensive stores, as compared to the historical norm—spread over a wider geographic area.
Malls: social and store aggregation
These trends do not just affect individual stores. They also affect collections of stores in concentrated geographic areas: shopping malls. In the United States, malls are challenged to reinvent themselves, given the demographics of aging and less-mobile populations in general, but also because of a Gen-Y population minority that is hooked on thrills and experience. They want more bargains, more sensory experiences, and more social interaction.
So malls are both trying to repurpose their square footage to allow “one-stop life” by putting in medical, dental, and vision services, for example, and trying to redefine the mall experience. And that means—you guessed it—multi-channel. They need to ensure a consistent and smooth user experience ranging from ecommerce to in-store purchasing; they need to know when the user is on the premises (sort of the opposite of “Elvis has left the building”); and they need to deliver timely, highly relevant offers.
Malls have another need, too, which is to enable store aggregation and social aggregation as part of the mall experience. Imagine a store initiating a flash-mob to 30 selected teens so they can all see at once how cool they look in the latest shirt.
Even Carnaby Street, London’s high-end retail district in Soho, is getting into the act with a recently created online site that integrates brands of stores in the district, enables promotion of retail events, and allows shoppers to interact directly with one another. Crowd buying (demand aggregation for volume discounts) is just one step away. Marry that with some of the things that were discussed at SXSW recently, and you have a match made in ... well, it depends on your perspective.
What if you could take a photo of someone else at the mall wearing some garment you fancy, have it matched in the cloud in a second, and then see a map showing where that exact item can be purchased, and which store is willing to offer the deepest discount for you? That’s heaven, if you’re on the hunt.
All this requires APIs for the control of data. Because the services the mall will provide must be driven by data from the tenants, APIs are more important than ever. These are new systems, and if you don’t want to wait three years for IT to deliver them, you can’t base it on a legacy SOA or ESB architecture.
Grocery retail: big data goes small
Historically, the grocery industry was one of the first to make use of data-driven retailing, both for real-time inventory tracking and placement optimization, and for rewards points systems that drew buyers to targeted items. People are buying food differently today, and the ways in which data is used in commerce is changing.
Families are increasingly shopping multiple times per week across many stores, making smaller purchases rather than holding back for a weekly “big bang.” Couple this with the fact that grocery discounters are trying to offer the best prices on everything, rather than doing promos on specific items, and the value of rewards systems that qualify repeat customers for a blanket discount is diminished. Some grocers instead are offering so-called small discounts: a free cup of coffee, for example, based on a customer’s purchasing profile.
Of course this doesn’t mean the store isn’t collecting and keeping the data. It just means that it’s finding “smaller” ways to use it, and “smaller,” more targeted ways to incent the customer. The best usage of this shows up in how some Apigee customers are developing the ability to customize the special offers to a customer in highly targeted ways based on the customer’s known brand preferences.
This seems simple, conceptually, but it can be challenging to implement if the purchase history is in a data warehouse designed to give aggregate reports rather than highly specific results. This will almost always require some new APIs to make it easy to get the data (for example, so that the point-of-sale or cash register knows that a customer gets a 5% discount on Brand 1 aftershave, while another customer gets a bigger discount on Brand 2 because he buys more per month).
It may also require the implementation of a special persistence layer that unites data from disparate back-end data sources, and offers a single “flat” programming model (REST APIs) to access it. The two-strata API architecture discussed in the previous post is the logical way to implement something like that, when it’s necessary.
We’ve covered the drivers for change in e-tail, discussed typical API architectures to support these transformations, and examined how particular retail sectors benefit from these investments. In the next and final post in this e-tail series, we’ll summarize some of the key dynamics in API-enabled e-tail digital transformation.