Explaining chatbots to 4 groups of people

Aug 08 2018

“I make chatbots.”

People gives me blank stares when I say that. It’s unlike the response I get if I say “I make websites.”

This is my script to explain what I do to 4 different groups: Child, adult, business owner and techie.

Explaining chatbot means nothing if they don’t how it can help them. So here’s what I try to do.


We’re talking right now, yes?

I can read you a story. Or I could tell you why cats can’t swim.

Just like you, many people ask me lots of questions.

I want to help everyone, but I only have two hands.

So I built a robot that can think and talk like me. The robot is called a ‘chatbot.’


Sometimes it’s easier to ask your friends than googling.

And you text them. Not call.

So why can’t you text your hairstylist for an appointment? Or text the cafe to make a reservation?

It’s **because it’s pricey to hire someone to reply thousands of messages. And we need to sleep.

Today we have chatbots - robots that are trained to understand our language and answer naturally.

You can text a company to get quick answers. You don’t need to wait long because a single chatbot can talk to thousands of people at the same time.

It could even understand if you type:

“Large pepperoni 🍕 pls. Thx 👍.”

Business owners

If your business takes reservations, you know that a missed call is a missed sale.

The slower you reply an email, that person might have gone to your competitor.

Someone called you just to ask if you’re open on weekend.

It’s in your website, but no one sees it.


Because asking you is faster to them.

Chatbot allows your customers to reach you - like texting a friend.

It automates repetitive tasks so that you can focus on the creative part of your business.


All the chat apps in your phone are the front-end part of a chatbot infrastructure. Texts are sent to a chatbot server to process and return a reply. Same goes with voice messages.

But a lot of chatbots now are like our old RPG games. A question followed by 2 boxes to select.

Yes, it’s more interesting than reading a long article. But it is lacking one piece - the natural language processing (NLP).

Each of us type differently. NLP is the ‘brain’ that tries to understand the text before triggering a reply.

That means instead of filling up a form to book a table, chatbot can understand this:

“A table for 2 next Friday 8pm.”

We spend hours to make a website mobile responsive. But what if people rather text than clicking through your site?

This is why I think chat is positioned to be the user interface of the next generation.