An Illuminating Conversation About Conversational Marketing

Fred Schonenberg has Amazon’s Alexa -- and he loves it. But does he need it?

“I live in a one-bedroom apartment in New York City,” the founder of VentureFuel asked his panelists at MediaPost’s OMMA at SXSW last Saturday. “Is it necessary for me to ask someone to turn on the lights for me? Is that something that I’ve been craving my whole life?”

And so, “Talk To The Brand: AI, Bots And Conversational Marketing” was off to a running start.

“I think there’s absolutely a purpose and a role for a bot in all of our lives,” responded Sam Olstein, global director of innovation, GE Corp. “What I’ve learned over the course of a couple of experiments is that a good bot does one thing and does one thing really well, and that’s pretty much it. So if you can crush it at that one thing, whether it’s turning on the lights, a flash news brief, helping give flowers to people we love, those are really clear pieces of value that are also removing friction. …”  


advertisement CEO Chris McCann, who oversees a family of brands including Fannie May and Harry & David, looks at things a bit differently.

“Was there a burning need for an MP3 player when it came out?” he asked. “But technology is always going to be driving change, and therefore there is a need for us to be on the forefront of that change. Consumers will migrate that way,” McCann maintained. And if people are spending their time on messaging apps, “let me interact with them in a very conversational manner, at scale, through a bot,” as his company has been doing on Facebook’s Messenger.

Stephanie Hay, head of conversation design for Capital One, agreed with McCann. “It’s a matter of being where the customer is -- and change is inevitable, so you’re part of it.”

That’s why, in early 2016, Capital One was the first financial institution to use Alexa, and last week launched the Eno chatbot, enabling customers to check balances or pay bills on their smartphones and tablets.

“If you want to continue to be part of the change, to take some of the friction out of people’s lives and make things easier, you’re going to be where your customer is,” she said. Sometimes that’s at home with Alexa; sometimes that’s out with Eno.

Video clips showcased each company’s foray into conversational marketing. Capital One’s focused on the features and benefits of Eno for consumers on the go. I-800-FLOWERS showed how its nine brands use IBM’s AI technology to better serve its customer through Gwyn, its gift concierge. It learns as it goes along and “is brand agnostic,” as McCann puts it (as long as the brand is one of 1-800-FLOWERS offerings, of course). GE’s piece was a winsome look at a girl interacting with the “lab assistant” for one of its Alexa-based Labracadabra science kits.

Here are some other highlights from the discussion:

  • We’re in the very early stages of conversational commerce. “For us, it’s about learning,” McCann said. “Learning, then iterating, enhancing the products, then expanding the products as we go forward.” It’s all built on a foundation of big data, deep analytics, machine learning and cognitive computing, he said, and there’s much more on the way. “Before we get anywhere near to perfecting this, there’s a bit of learning time to come,” he concluded.
  • One of the advantages of Capital One developing its own proprietary chatbot is that it has access to all the raw data about what customers are saying and looking for. With Alexa, Capital One has to rely on Amazon’s analysis. “Natural language in voice, which is full sentences, is actually completely different in texting,” Hay pointed out. "Natural language in texting is emojis. It’s using ‘Y’ instead of yes, or ‘N’ instead of no.”

    Responding to a question from the audience later, GE’s Olstein said, “I think the onus is on us, on brands and advertisers, rather than the platforms themselves. Yes, it would be amazing to get all kinds of data from Amazon around the performance of our skill. But we’re not going to get them to move any quicker than they want to. And it’s so crucial, so criticalto think about data signals that we can see, that we can measure, whether they’re on that platform or related to a behavior that’s tangentially related to something else.”
  • Privacy and security are major concerns, not only for Capital One -- where it is “priority number one” for obvious reasons -- but also for, where Gwyn retains credit card information and other preferences to speed along future transactions. Because Labracadabra targets children, GE employed “a lot of lawyers.” It needs to make sure users are old enough for Alexa to act on any information it may be given in the conversational back-and-forth.

The panelists also answered questions about how to get buy-in across the organization, whether they’re hitting their usage goals, what KPIs are relevant at this stage of the game, what’s coming next -- and more. You can watch the hour-long presentation here.

Next story loading loading..