Laurie Keith has a wealth of knowledge to share about chatbots and Conversational UI, which she refers to as "marketing that scales." As VP of Media at the Ad Council, Keith is at the unique intersection of advertising, media/technology, and social good. In her role, Keith is responsible for building and managing relationships with major media, tech and entertainment companies. Keith has utilized chatbots to drive meaningful and intimate conversations with users around large scale innovative social-impact campaigns.
Rosenthal: Unlike social media, which is one to many, messaging is one to one marketing at scale. What about this shift is most profound?
Keith What makes this shift so profound and game changing is the ability to send a personalized message based on user interests and behaviors. Most brands sit on extensive amounts of customer data that they can now use this to start a dialogue. As a customer, I’d be much more likely to interact with a marketing message if it was personally directed to me and provided information that was useful or beneficial to me. Because of the 1-1 nature of messaging, it provides a space for vulnerable, intimate conversations. This was important to the Ad Council when we built our first bot with Kik a few years back for our Teen Dating Violence Prevention campaign. Through the Kik Messenger bot, we provided different dating scenarios for teens to rate as “cool” or “not cool.” Resources were given to teens that responded to the bot and needed help or more information. Because messages within Kik were already being used for personal, confidential conversations – teens felt comfortable interacting with the bot about such a sensitive topic. Chatbots make it possible to have personal interactions with users as compared to a 1-to-many advertisement.
Rosenthal: When you think about the future of conversational UI, what about it are you most excited by?
Keith For me, it’s the 1-1 personal interaction on a platform that people are using daily. Every marketer always talks about “meeting the consumer where they are,” and that’s messaging – with 2 billion messages exchanged every month. With that being said, the content needs to be helpful or entertaining if a user is going to continue to interact. I’m an avid runner and I buy the same running shoes every year from Brooks – the Ghost series. I’m in the market for new running shoes right now and noticed they sent me a catalogue in the mail. In the future, they’ll be able to send me a personal message telling me it’s time to purchase new shoes. Within the same interface, they can how me all of their new styles and allow me to purchase. This is the true definition of marketing that scales.
Rosenthal: If a brand had a perfect data set, how should they use it to personalize the chatbot experience with their customers? What type of data should brands be most interested in collecting?
Keith I think it really depends on the brand and their objectives. If you’re a retail brand and have data on every customer purchase – you have the ability to personalize a chatbot experience relating to the products they’re specifically interested in and will likely buy again. If you’re in the entertainment business, you’ll want to collect data on what topics or events are of interest to the users so you can notify them when something comes up. Before the bot is built, I think it’s important to collect data that will help personalize the bot experience and increase engagement.
Rosenthal: How do you think chatbots should best be utilized by brands? For commerce, content delivery, utility, CRM, customer service?
Keith Again, I think it comes down to the brand objectives but I don’t think a chatbot will survive if it doesn’t provide some type of utility to the consumer. Consumers are smart and they’ll turn away if they feel like they’re being sold or preached to. If the bot is able to provide a solution to a problem or answer to a question – the user will want to interact with it. Before the election last year, we wanted to help people register to vote so we built a very simple and easy FB Messenger “Go vote” chat bot and targeted newly eligible voters and people who have just moved – two groups likely to skip registration. Similar to a chat bot – we partnered with Amazon Alexa this year to build a “Save the Food” skill to help reduce food waste in our country. The new skill can tell you how to best store food so it’ll last longer and whether a vegetable can still be eaten even if it’s completely wilted. Considering most people interact with Alexa in the kitchen – we knew this would provide great utility and a skill the user will continue to go back to. In just one month, the skill had been downloaded by over 45k people resulting in 255,000 utterances. The reason the skill has been so successful is because it provides great utility to the end consumer. And this is the same with chat bots. If you want your customer to trust your bot and engage with it – you need to provide them something useful in return.