Building An AI Bot That Meets Six Basic, Universal Needs And Makes Us Tick (Part 1)
I came across a New York Times best selling author, Tony Robbins’ speech the other day. He talked about whatever decisions one makes or path one pursues — whether it be quitting one’s job to be an entrepreneur or traveling around the world — is driven by six basic, universal needs. They are certainty, variety, significance, love & connection, growth, contribution, which determine how we experience life. Combined, they are the force behind the crazy things (other) people do and the great things we do☺.
Beyond the hype of chatbots being the gateway to a fully integrated and automated world, bots are means to better serve people, not ends. In this sense, we still need to put the customer experience and human needs at the center of the design thinking process. This post will discuss the first two needs — Certainty and Variety.
Need 1: Certainty/Comfort
I have been working on a bot in the event space. A few weeks ago, I went to a pop-up lunch and pitched my bot to the manager. I got so excited talking about what languages this bot is being built on and how intelligent it could get with its AI capacity. After a while, I noticed she looked absolutely lost and said to me, “It sounds so cool but so high-tech, how do we use it?”
Then it hit me. Early adopters in the tech community would try a bot just for the sake of testing it out. The first human need is the need for certainty. It is our need to feel in control and know what’s coming next so we can feel secure. Unless you are building a bot specifically for the tech community (well, that means a much higher bar for customer satisfaction), everyday consumers (non-geeks) wouldn’t chat with bots for fun. The bot’s functionality that actually solves a problem has to come first before the fact the product is a “cool” bot and has some AI/NLP capabilities.
Regarding reliability or building trust between users and a bot, the CEO of Assist points out some challenges a bot faces. If there is a bug in any part of the flow, the simple interface shows up nothing or funny responses, which would fail to meet this specific human need.
“Beware though, bots have the illusion of simplicity on the front end but there are many hurdles to overcome to create a great experience. So much work to be done. Analytics, flow optimization, keeping up with ever changing platforms that have no standard. For deeper integrations and real commerce like Assist powers, you have error checking, integrations to APIs, routing and escalation to live human support, understanding NLP, no back buttons, no home button, etc etc. We have to unlearn everything we learned the past 20 years to create an amazing experience in this new browser.” — Shane Mac, CEO of Assist
Need 2: Uncertainty/Variety
People get bored if they constantly feel secure. Especially for millennials, they are constantly craving for new things/experiences/surprises, of course, within their comfort zone most of the times. The second need actually creates the dilemma between explore (gathering information) and exploit (acting upon the information collected). There have been so many recommendation engines built around this need to a point the sheer volume of events available in event-based social networks undermines the users’ ability to choose the events that best fit their interests. I can’t tell you how many times I felt too overwhelmed and had to close the browser after opening 50+ tabs for things or events I was interested in.
Finding peace in a stressed-out, digitally dependent culture may just be a matter of thinking differently — Kate Pickert
The chatbot medium, in my opinion, can actually solve this problem by either reducing the volume or shorten the decision making interval to trigger actions. To reduce volume, we can choose which datasets to include or what kind of curated content to present to users. Platforms like Facebook, Eventbrite have to be unbiased and let anyone post. However, a bot can be discerning and have certain personalities.
On the other hand, Brian Christian and Tom Griffiths, in their new book, Algorithms to Live By, point out ‘interval’ over which we plan to enjoy or utilize the results of our actions is another important factor to consider. How much time do you have to make a decision? Can the chatbots or user cases somehow shorten the ‘interval’ to push users to make decisions faster? Even though people always tend to over-explore beyond the optional stopping point, given a time frame, people can make much faster decisions, which would translate to transactions for businesses.
- Note: Next post will be on the third need — Significance.
- I’m interested in hearing your perspectives or feedback as you are building your bots.
Florent is an artificial intelligence SMS chatbot that captures event traffic and influences event attendees’ next logical moves based on each event’s unique characteristics.