Q&A: Boxed co-founder talks about machine learning
Driven by the need “to solve problems for its shoppers,” Boxed is about to make it easier than ever for customers to reorder merchandise — without even having to think about it.
Launched in 2013, Boxed sells everyday essentials online — in bulk form only and at a discount. The e-retailer carries some 1,500 SKUs across such categories as beauty, grocery, household, and more. With warehouses in Union (New Jersey), Dallas, Las Vegas, and Atlanta, Boxed fulfills and ships orders to customers’ doorsteps within two days or less. And unlike the warehouse club retailers it has been compared to, Boxed does not require a membership fee. And most of its orders qualify for free shipping.
The model is paying off: The company has surpassed $100 million in sales. To keep revenues growing — and keep on top of customers' needs — Boxed is now launching new machine learning-based tools that automatically replenish goods before customers run out.
Boxed’s co-founder and CTO William Fong shared his thoughts about the initiative — and expectations — with Chain Store Age.
CSA: How can machine learning benefit the retail industry?
WF: We think machine learning is a way to solve problems. It can help retailers understand customer behavior and shopping patterns at a deeper level, and then help them make more accurate recommendations based on clicking patterns. We took this to the next level, and used machine learning to drive a new level of personalization for our shoppers.
CSA: How did you make this decision?
WF: We were trying to think of solutions that would help us in the long-term. We tried to imagine how our customers might want to engage with Boxed five or 10 years from now. The more we thought about it, we said, “what if the best solution supported a way for shoppers not to engage with us at all?”
We thought along the lines of how we engage with utilities. Other than reporting an outage or paying a bill, those customer relationships are autonomous. This lead us to use machine learning to put the ordering process on auto-pilot. The end goal was to create a solution that enables customers to automatically replenish their most-used merchandise — snacks, paper goods, etc. — before the run out.
CSA: Tell us more about your specific solution?
WF: The solution we created is called Smart Stock Up, and it will be available to shoppers on Friday, August 4. While we are not quite at the automated replenishment level, Smart Stock Up is a good first step. We use machine learning to understand shoppers’ most used items — and when they will run out — and push out an email that reveals the merchandise that we think they are running low on.
On average, the list will contain 10 items, giving them a good jump start to building their basket. If they agree, customers click on a link that directs them to our site. Here, they can add the merchandise to their cart, and either continue shopping or proceed directly to checkout.
CSA: Are you offering something similar for business customers?
WF: We are piloting a second service among select business partners, called Concierge. It differs from Smart Stock Up in a couple of ways, including that it is an opt-in program, and is fully automated. It is also a two-pronged approach that focuses on replenishment and discovery.
Using machine learning, we find the merchandise they purchase most often, and the items we think these clients might like, such as snacks like chips or popcorn. When we ship packages of their restocked merchandise, we also add a full-sized “discovery” product for them to try.
So far, only a few customers are participating, but we are gradually expanding the test to more companies. So far, we haven’t gotten any returns on the discovery merchandise we have shipped out. If we got wrong however, they simply return it to us. We close the loop with a survey to get feedback and data to improve experience.
We hope to roll out Concierge to our full business customer base in 2018.
CSA: Is this your first foray into machine learning?
WF: Machine learning is all over place at Boxed because we use it to achieve our goal of “problem solving.” While these two new services customer facing, we also use it on our back end.
For example, we use it to forecast our inventory levels and needs in our warehouse. It also powers robotics in our distribution network. Specifically, our automated vehicles that pick merchandise use prompts from our computer system, and results are based on machine learning output.
What do customers want from virtual shopping?
Virtual reality shopping is here and consumers — or at least the ones that are tech-savvy — are ready to use it.
Seventy to eighty percent of "early tech adopter" consumers are eager to use virtual commerce technology to design rooms, customize products and shops with friends from across the globe, according to a study from L.E.K. Consulting.
The report said that retailers are ramping up investments in two types of virtual commerce technology: virtual reality (VR), where consumers use headsets to enter a completely digital world; and the more-accessible augmented reality (AR), where they use their smartphones to get information (such as prices and color selections) overlaid on a picture of the physical showroom or shopping space.
“For retailers, the appeal is obvious,” said L.E.K. managing director Rob Haslehurst. "These technologies are a new way for retailers to do what customers want them to — create compelling shopping experiences and have rich communications with them."
Eighty percent (80%) of shoppers want to use AR or VR to design a room or physical space by browsing virtual or physical showrooms, getting information about furniture and décor, and "seeing" what it looks like. (The survey respondents were made up of consumers who had already experienced VR and AR technology.)
Savvy retailers are already in tune with the trend. Wayfair, for example, already features VR showrooms where customers can see a room come together as they fill their basket with products, and Lowe’s "Holoroom" lets customers design a virtual room and then tour the space. Alibaba's "Buy+" VR app allows consumers to browse and buy from the aisles of a virtual store, no matter where they are in physical space, the study reported.
In other key findings:
• Seventy percent of survey respondents want to use v-commerce to try on clothes and accessories, and to customize the items. Consumers can start with an image of themselves on their smartphones, then search for the perfect shade of makeup or an eyeglass frame that perfectly suits them. The Gap and Sephora are already offering these AR applications.
• Seventy percent of respondents are strongly interested in virtual shop-ping. Here, they can use VR headsets to shop in a virtual store with a friend who isn't physically present, or be guided by an artificial intelli-gence (AI) "virtual shopper" similar to Alexa or Siri.
“V-commerce can create new, special experiences that would otherwise not be possible, and that leads to greater consumer engagement,” said L.E.K. managing director Maria Steingoltz. "It enables retailers to unify physical and digital channels — brick-and-mortar retailers can bring digital capabilities into the store experience, and online-only retailers can create virtual ‘stores'. And the rich experience can generate more sales — a customer can 'see' a sofa in his or her own living room, and then be shown the cushions, lamps and side tables that go with it.”
L.E.K. offered the following tips for retailers who want to take advantage of the v-commerce opportunity:
• Act immediately to make AR and VR a part of their digital strategy.
• Establish a compelling value proposition and define the business model. "Make sure customers understand from the first encounter how the technology solves their pain," said Steingoltz. "And make sure to define the resources, concrete goals, and metrics for the pro-ject."
• Consider making alliances with technology leaders. "Retailers don't need to be technology experts," said Haslehurst. "Look for alliances that provide access to world-class technology and give technology makers a good story to tell.
Amazon Opens Old Wounds for Retailers (Again)
Before we completely move on from last month’s Amazon Prime Day, it’s important to understand why retail brands once again braced themselves to lose customers and online sales when they had a whole year to do things differently.
Amazon’s ingredients for the promotion were extensive and strategic, but were also simple extensions of its carefully built ecosystem: They offered sneak previews of the sale a week in advance, gave Alexa shoppers a several-day head start on Alexa-specific product deals, gave Prime Now members a two-day head start on deals for products with a 2-hour delivery, pushed voice shopping by giving first-time voice shoppers first dibs on deals the day of the promotion, and extended access to 100 Alexa-related deals beyond the close of the promotion.
The most glaring takeaway from all of this is that Amazon didn’t do anything that other retailers can’t do themselves. When you break it down, they put together a promotion that rewarded loyal shoppers, drew attention to the benefits of membership, pushed valuable behaviors they want to see more of, and created deal segments according to user and product type.
While there are admittedly few retailers that have the infrastructure and product variety that Amazon does, what most retailers do have is lots and lots of data. And as of very recently, they also have access to advanced technologies like artificial intelligence that can put this data to work so they’re less vulnerable to retailers like Amazon.
What continues to stand between many retailers and Amazon is their inability to put technology and data to work for them, so they can think less about the day-to-day minutiae and more about the customer experience. Here’s how Amazon did that and what retailers can learn.
Get out of the weeds and put the customer first.
A major inefficiency for retailers is the overwhelming amount of customer data they must wade through to come up with the type of tangible insights — recurring visitor trends, buying patterns, and product preferences — that Amazon bases most of its marketing and promotions on. The same data that allows retailers to hyper target their audiences with personalized messages has simultaneously become way too much for them to analyze and act on creatively in other ways that would enhance the customer experience.
Amazon got ahead of this problem years ago by introducing a proprietary technology infrastructure that relies on artificial intelligence and machine learning to maximize customers’ experiences on the site. Because Amazon has automated many of the day-to-day processes that fuel their understanding of their customers, while other retailers still rely on teams to manually crunch and interpret data, Amazon has freed up its teams to work on more creative programs.
Retailer agnostic AI platforms are now emerging to let retailers bridge the gap between themselves and major retailers like Amazon. Amazon Prime Day is a perfect example of what’s possible when retailers have the luxury of time
Stop telling customers who they are and what they want.
If the groups that Amazon prioritized for Prime Day taught us anything, it’s that Amazon isn’t overthinking things. In fact, quite the opposite. What Amazon continues to do so well is use its data to identify and understand it customers, both individually and as parts of the larger behavior-driven groups they fall into. Amazon doesn’t make the mistake of pre-determining user segments and buyer personas, and then retroactively fitting customers into them.
Instead, Amazon lets its customers’ demographic and behavioral data tell them what they like and don’t like, what products are trending among which types of people, and where there are both predictable and unexpected patterns that inform larger user trends. Amazon additionally offers its customers several opportunities to self-identify themselves, their commitment level, and their priorities through structured programs like Amazon Prime and Prime Now. All of this gives Amazon digestible groups-based insights that they can use to seamlessly cater to these groups.
Supplement retargeting with rewarding.
Retailers can dedicate a lot of time and resources to retargeting disinterested customers while inadvertently ignoring highly valuable customers they don’t know about. Amazon has solved for this with laser-precision targeting, based on the type of acute insights discussed above.
What they do next is also worth noting. During Amazon Prime Day, Amazon isolated a very specific group of customers to activate: Alexa customers who have not yet used the voice shopping feature. Presumably, Amazon has seen increased order frequency or order values from voice shoppers, and sees an opportunity to grow this segment. So, in retargeting them, Amazon didn’t just put Alexa-related products in front of these shoppers; it rewarded everyone in this very specific segment with two hours advanced access to Prime Day deals and a 6-day extension on voice deals across segments.
Not surprisingly, Amazon reported that "Prime members’ most popular purchase was the Echo Dot, which was not only the best-selling Amazon device this Prime Day, but also the best-selling product from any manufacturer in any category across Amazon globally."
There are several other takeaways and lessons to be learned from the success of Prime Day, but at the core of all of it, what Amazon continues to do better than other retailers is use technology to free up time and maximize reach, data to point them in the right direction, and good old-fashioned creativity to tailor promotions that get customers to do and buy what they want them to.
For retail brands who have never understood why they couldn’t compete with Amazon, emerging AI solutions will illuminate the advantages Amazon has enjoyed, while simultaneously eliminating them.