Geek Culture
Published in

Geek Culture

How We Transformed Our Idea From an MVP to an Attractive No-code Tool to Build Chatbots

Building a chatbot is a hard thing to pull off. A couple of years ago we started a fresh idea to provide the easiest way for both the marketers and developers to create engaging bots for messaging services. While the idea was great, our product — WotNot was not up to the level it should be. We had to almost start from scratch and build everything from the ground up. So how did we do it?

MVP to Nocode tool

From Problem to Idea

One of our long-term clients approached us with a problem almost every business struggles with these days. Due to their limited customer support staff, they were unable to answer all calls promptly, which made customers upset and they missed out on this potential revenue.

We proposed to build a chatbot that would address this worrying issue. Immediately after we deployed this bot, the client started seeing exceptional results And thus we began ideating on building a chatbot platform that businesses could use to gain similar results for their business across verticals. That’s how we conceived the idea to build a chatbot platform called WotNot

The Journey

While building a platform to build chatbots, we wanted to keep things simple so we didn’t go overboard and decided to start with an MVP. To reduce the need for creating UI, we thought of using google sheets to create and manage the chatbot flow. Using Google Sheets saved us a lot of time as it had in-built features like version control, auto-save, multi-language support, etc. which we otherwise would have to build.

Our idea was to get started with a simple bot that could respond to basic questions and collecting user details, but quickly we realized how powerful google sheets really are. So things became pretty straightforward. Google Sheets were fast and simple.

This is a story about hardship, hard work, and perseverance. It is the story of how despite many obstacles and many mistakes, we were finally able to launch and bring WotNot to life. What I learned was that even the most difficult task could be overcome if we work hard collectively and consistently. The launch of WotNot was a whirlwind event for us.

This was how v1 of the chatbot looked like:

First version

As the chatbot was managed Google Sheets, this paved the way for us to automate the development process. The workflow became more streamlined and efficient as we could create our automation scripts directly in Google Sheets by creating custom plugins. Many people know Google Sheets as a way to structure their table or data by grouping elements together or making a few changes that will update the whole sheet. Here is how defining the conversation flow in Google Sheets looked like,

Google Sheet showing the flow

For us and our team, this seemed to be the easiest and fastest way to design a flow for the chatbot. Pleased with the product we had our first paying customer happily deploying the chatbot on their website to capture more leads for their business. As time passed by, we started onboarding many customers and as a result, our company boomed. Our growth rate was extraordinary.

The Problem

This idea to control the chatbot flow via Google Sheets looked nice at first glance as it was easy to manage but as you work with updating the chatbot flow it became difficult to keep track of the flow. If I wanted to make changes to an already developed flow, I needed to figure out how? As the entire sheet needed to be updated.

Onboarding new customers was a challenge as we had to learn how to structure the conversation flow on Google Sheets, which made them scratch their heads. Over time, we noticed that many of our customers did not want the headache of managing the flow using Google Sheets and they passed this headache on to our team.

This was the time when we thought to build a tool that could help our customers easily control the flow and update it whenever needed through an easier visual diagram.

MVP to Nocode tool

Google sheets was a killer idea but it had a lot of pain with onboarding new customers as customers could not build a bot on their own. customers would give us a flow on a flowchart tool like Whimsical or we would do it for them. Then they would approve. And finally, we would begin making the bot on Google Sheets. We analyzed and found that customers were okay with creating a flowchart because they could visually see the conversation. This is how we got the idea to build a visual builder where customers could build the bot on their own.

We decided to build an easy-to-use tool that will help users easily design chatbots. Taking control over the tool and doing things yourself is something everyone loves. To build an effective tool from scratch, we had to think about our audience and what they want. Designing a tool that people find easy to use requires attention to detail and execution.

We spent a lot of time on the research and eventually built a nocode chatbot tool that helps users deploy chatbot in minutes without any hassle. Our chatbot builder solution allowed users to build, train, and deploy the bot on their own without having to learn any code. How amazing is that?

Nocode chatbot tool
Chatbot deployed on website

Bot Deployed on the website Building this nocode tool also gave us some of the additional benefits like

  • A/B testing of the conversation
  • Support to multiple channels like Facebook, WhatsApp, SMS
  • Third-party integration to different tools like Airtable, Salesforce, Hubspot, Slack, Twilio, and a lot more.

Since then, We have been helping businesses cut costs and improve their customers’ experience. We take pride in our simplistic chatbot platform, which stands out in the crowd of complicated and technical chatbot services.

Conclusion

The transformation from MVP to a nocode tool was the best decision ever made. We had a lot of experiences (Good and Bad) in this journey and these experiences help you eliminate mistakes and get things done faster.

Diversification vs. concentration. (@visualizevalue)
  • We focused on one thing at a time, even though we were getting a lot of other feature requests from our customers.
  • Test with real people, your product might work, but it can get better. You need to know if your product idea is viable and customers will actually spend money on it.
  • Don’t just make and release an MVP, keep building and learning with your users
  • By making the building feedback loop shorter, you get better output faster, while getting critical user feedback to improve your offering.

Just like other journeys, long journeys are often punctuated by many short rests. So here’s the rest. Thanks for your stopping by and reading this article.

Feel free to signup at WotNot and try building a bot on your own. I would be happy to chat with you — Link

I often write about technical blogs and our product journey, feel free to connect with me on Twitter (@MakadiaHarsh) for more updates.

Happy Learning!

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store