Digit Geek
Digit Geek > Recent Articles > Technology > Build your own chatbot without coding

Build your own chatbot without coding

No coding? No problem.

Right now, there’s a lot of buzz around chatbots and how they’re going to take over from apps. While we’re reserving judgement on that until we have more data, it doesn’t hurt to create a chatbot yourself anyway. Thanks to a rising interest in them, a lot of platforms have been created where you can build your own chatbot without knowing how to write a single line of code.


There are just too many ‘make a bot without coding’ platforms out there to analyse in one workshop – each with their own pros and cons. While standard platforms from big names like Facebook and Google are available, there are other popular alternatives like Botsify as well in the fray. Here’s what they are good for in brief:

  • Dialogflow: Formerly Api.ai, Google’s chatbot development platform plugs into Google’s machine learning expertise. Divided into intents and entities, you only need to provide a few possible queries for each intent and Dialogflow handles the rest.
Dialogflow is Google’s chatbot building
  • Messenger: This is essentially Facebook’s platform for developing bots for Facebook Messenger, which is the platform of choice if you want to reach a big audience.
  • Botsify: With Botsify, you get plugins that are easy to setup for stuff like your RSS Feed and more. We’re going with Botsify on this workshop because we intend to use it to develop an RSS feed bot.
  • Slack: If you want to build a bot for enterprise usage, or have some team-specific use cases in mind, then developing a bot for Slack using the Slack API is your best bet.

Bot Building with Botsify

After logging in with your Facebook account (which is recommended, especially if you’re trying to make a Messenger bot), you start off by selecting the purpose for your chatbot – whether it is a generic bot that you intend to link to your Facebook page or Slack account, or whether it will be deployed on your website – which is a premium template.

Botsify Dashboard: This is where you can start off building your chatbot with some basic steps

We’ve modelled this workshop on building a chatbot for the Digit Geek Facebook page which would help you find articles that are published on the website. You can choose your own objective and modify specific steps according to your requirements accordingly. The process is easy: select a name, followed by the Facebook page where you want this to run and then go through the following steps.

Step 1: Setup basic messages

This interface is to decide how your bot behaves at the very beginning of an interaction. We chose Generic bot for a Facebook page as our target, so we need to set up three types of messages:

  • Greeting message: This is what appears as a message before the user has clicked the Get Started button on Messenger. You can use user parameters like {first name} from a dropdown to personalise the message.
  • Get Started button: This is what determines how your bot responds to a Hi, Hello or other initial messages. You can choose simple text responses, templates, stories or plugins. Our bot is intended to browse the Digit Geek website, so we set up a template to prompt the user with a Category menu.
  • Default Message: This is what the bot responds with if it doesn’t understand something. It is good to keep a couple of options here so that the failure messages aren’t repetitive.

Step 2: Set up templates

Mouse over the main menu on the left and click on template designer. We’ll be creating two templates here:

  • Menu:
    • Select the Generic Template
    • Click on the Image button to provide a link to the image for the first card, and provide Title and Subtitle from respective buttons.
    • Click on Plugin button and select RSS plugin. Provide the URL to the RSS feed and other attributes, hit Edit on the floating window to save it, then provide a title for the Button and hit Edit again.
    • Add as many cards needed on the right. These cards can also point to web links that can be added by the Weblink button.
Enter the URL of the RSS Feed
  • FirstResponse:
    • Select the Button Template.
    • Similar to Menu, provide Text for the message box.
    • Click on Template button, add a title for the Button and select Menu from the drop-down list, which is essentially the template from the previous step.
Creating a template to be returned to the user

That’s it for click-based interactions. Now for the chat-based interactions.

Step 3: Set up stories

  • Go to the main menu and click on New Story.
  • You can choose the criteria for a response to be a keyword, a phrase or an entity (more on that later). Go with Keyword.
  • For this chatbot, specify the categories from the website with as many keywords as possible. For instance, for Tech, you would use variations like Technology, Gadgets, Techs, etc.
A story determines how your user will interact with your bot
  • Specify a Bot says message followed by a Plugin response, where you will have to select RSS feed and proceed similarly as before. Do this for as many categories as possible.
  • Depending on how many RSS feeds are available, you can (and should) create as many stories as possible to account for various user interactions.

Step 4: Optimise

All the steps so far have been to set up the chatbot and it should be operational by now. You can optimise it by doing one or many of the following:

  • Go to Learning and Understanding to specify phrases that are similar to existing stories.
  • Go to Live chat to monitor chats in real time and link missed phrases/words to similar stories

There’s a lot more you can do if you play around with the settings. Do hit us up with the chatbots you create, we would love to try them out.

Arnab Mukherjee

Arnab Mukherjee

A former tech-support desk jockey, you can find this individual delving deep into all things tech, fiction and food. Calling his sense of humour merely terrible would be a much better joke than what he usually makes.