During Holiday 2016, we took a look at E.L.F, the chatbot at the Mall of America in Minneapolis and another AI offering out of the IBM Watson stables. The Mall has over 520 stores and restaurants, an indoor theme park, a 1.3 million gallon aquarium, and its own wedding chapel. Forty million visitors a year walk the 4.2 million square feet each year. And they now have a chatbot. visitors a year walk the 4.2 million square feet each year. And they now have a chatbot. 1080bots keeps up with the latest ecommerce chatbots to help our clients create brilliant omnichannel experiences.
E.L.F Chatbot at your Service
E.L.F. or Experimental List Formulator is a chatbot created by agency Satis.fi to showcase IBM Watson’s Artificial Intelligence Technology. The bot begins by asking a few simple questions and provides buttons for a quick and easy response.
As a good chatbot should, E.L.F introduces itself as helping with activity ideas for visitors to the Mall of America. It asks a few questions using buttons, such as children or adult activities, and how long we’d be spending at the mall. It then asked what we like shopping for and offered two options plus a More button.
Can E.L.F Help with Chelsea Boots?
We use the same simple search term on most of the retail chatbots we look at, we call it the “Chelsea Boots Challenge”. Since we knew E.L.F. was an IBM Watson chatbot, we also know it’s powered by some pretty solid AI, so we decided to skip the buttons and give it our ‘Chelsea Boots’ challenge. Would simply typing in a style name in natural language quickly get us to a shoe store? E.L.F wasn’t able to help us with our search and politely asked us to restart the session.
We restarted and went back to pressing buttons — “Shopping,” followed by “Fashion for Him” — but the chatbot returned one suggestion for coffee and another to try the mobile app. Neither had active links, so we pressed the “More Ideas” button only to receive a couple more dead-end bullets. This seems like a lost opportunity, as we were clearly looking to shop.
One of the suggestions was for the store Askov Finlayson so we typed that in. E.L.F could go no further and let us know we could chat with a live human.
We returned to the Facebook Messenger interface and asked to speak with a concierge. Chris came on the line and we let him know we were looking for Chelsea Boots. He got back to us within minutes with a number of shoe shops. We became more specific asking for high-end stores, and he was good in getting back to us again. Meanwhile, we were looking at the online store directory ourselves and getting similar results. Were we all using the same app?
What We Learned
Our experience with E.L.F provided some excellent takeaways on how a real holiday shopper might interact with a chatbot, and how it can ideally respond:
- Start with an introduction: E.L.F. greeted us and set clear expectations as to how it could help us
- Make it easy to get help: there were persistent stopwords that took us back to the start or to get help
- Let your customers find a person, too: handoff to a human concierge assistant was readily available and seamlessly handled
- Remove dead-ends: even if your bot can’t follow the conversation, try to first direct the customer to something useful before just restarting.
If your bot is rooted in AI, you have the opportunity to take the conversation to the next level. For example, if the user types instead of pressing buttons, and if the text is relevant, then the chatbot should be able to interpret and respond appropriately, even matching a brand or product query to a relevant store or product page.
At 1080bots we’re reviewing ecommerce chatbots to help us build brilliant conversational commerce experiences for retailers and brands to engage their customers. If you want to figure out what functions you should hand off to a chatbot, drop us a line at firstname.lastname@example.org or fill out the form below and we’ll get right back to you.