How to Make a Chatbot People will Love In 3 Steps
How to go from idea to MVP to live chatbot over the weekend.
The goal of any product should be love…your customers should absolutely love the solution you have created for them. Using these 3 steps, you can quickly go from idea to a chatbot people love over the course of a weekend! Let it begin…
Step 1: Chatbot Only Use Cases
For thousands of years, we solved problems directly through conversation, chatbots are a throwback to this simpler time.
Chatbots, in their current form, don’t provide as rich a GUI experience as apps. As a result, the value presented by a conversational interface needs to be utterly simple in order to be truly leveraged.
There are Two Main Reasons Why Anyone Would Use a Chatbot:
- Conversational: When an App can’t do it because multiple variable inputs are needed to solve the problem.
- Simplicity: When a bot offers the most immediate and direct solution to a person’s problem.
This should lead you towards ‘Chatbot Only Use cases’ which require multiple conversation inputs and can solve problems in a more direct way than apps or offer superior simplicity.
You can’t download an app for all your problems, however one day you might use a bot.
Discover Latest Jobs in AI, ML, Big Data & Computer Vision:
- Conversational Designer
- NLP / NLU Engineer
- AI Jobs & ML Jobs
- Big Data, Data Engineering & Data Science Jobs
- Product Manager
Examples:
DoNotPay: chatbot Lawyer that Can Fight Parking Tickets for You
How to Properly Leverage Machine Learning
Artificial Intelligence is still in its infancy and there is a long way to go.
When it comes to machine learning, it excels in areas where the epistemology of the field is clear and well defined. For example, machine learning can be used to a great effect in fields like medicine and law but will fail in a areas such as sales.
Great Use Cases for Ai & Machine Learning:
- Ross.ai : ROSS is an A.I. lawyer that helps human lawyers research faster and focus on advising clients.
- Predicting Cancer: Computers trounce pathologists in predicting lung cancer type, severity.
- Gradberry: No more applying. Gradberry uses AI to evaluate developers by checking their code and matches them with ideal company.
Discover Latest Jobs in Bots, NLP, AI, ML, NLU & More
- Conversational Designer
- NLP / NLU Engineer
- AI Jobs & ML Jobs
- Big Data, Data Engineering & Data Science Jobs
- Product Manager
Step 2: Seeing the Future… WeChat
For a number of months now, Facebook has been quite transparent and obvious in their goals for Messenger chatbots. Ultimately, they hope that Messenger platform will resemble WeChat and Facebook has even urged developers to look at WeChat as the model.
Featured CBM: China, Wechat, and the Origins of Chatbots
So what does this mean? Well let’s take a quick look at some of the things you can do on WeChat:
On-Demand Service via Text: Via WeChat you can get just about everything on-demand with a simple text message. Need to get you place cleaned? Hire a plumber? etc… You can do it instantly via text.
No-Lines, No Servers, No Cashiers: WeChat has revolutionized the restaurant experience. You can now order and pay for you meal via WeChat. When you enter the restaurant, you simple pick up your meal and enjoy! (Currently, Allset is doing this, more below)
Virality: Grouping so many functions together in one app allows everyday things to go viral.
The Coming Restaurant Revolution
According to Stas Matviyenko, Founder and CEO at Allset, “The restaurant revolution is here. New technologies are changing the traditional restaurant model and the ways restaurants interact with their customers. We see more and more restaurants who want to bring ordering and reservations directly to where people spending a lot of time chatting with friends. Facebook has created a great opportunity for businesses to offer a whole new experience to their customers.”
Step 3: Rapid Prototyping using Chatfuel & Releasing Bots in Hours using GupShup
Chatfuel allows you to build bots in 10–15 minutes without coding! Once you have your bot up, you can quickly test your idea.
I strongly recommend using Chatfuel to:
1. Test Your Concept: Is your bot solving a painful problem that people care about?
2. Usability Testing: How usable is your bot? Does the flow make sense? Where do users drop off?
3. Advance your Copywriting: Can you get to the ‘Aha Moment’ right away and provide value as quickly as possible?
Necessary CBM: Usability Heuristics for Bots
Once you know what to build, I recommend using GupShup:
- Very Easy to Use and Minimal Programing: You can create and go live in a few short hours.
- Instantly Deploys to Multiple Platforms: FB Messenger, Twitter, Telegram, Slack, SMS, etc…
- NLP Provided: GupShup seamlessly connects to Wit.ai
Goal is to release as fast as possible, gather insights and feedback and iterate quickly towards a product people love.
To Give your Bot Personality & the Right Viral Hooks Check out: 10 Tips on Creating an Addictive Chatbot
Let’s Hack Chatbots Together
Creator of 10+ bots, including Smart Notes Bot. Founder of Chatbot’s Life, where we help companies create great chatbots and share our insights along the way.
Want to Talk Bots? Best way to chat directly and see my latest projects is via my Personal Bot: Stefan’s Bot.
Chatbot Projects
Currently, I’m consulting a number of companies on their chatbot projects. To get feedback on your Chatbot project or to Start a Chatbot Project, contact me.
