The Website Prototype Bot service is a paid 1-on-1 video call over Zoom. I can help you build a prototype website chatbot using Dialogflow Messenger in 90 minutes.

What we will build

We will build a prototype (pilot) chatbot which

  1. includes webhook code
  2. includes data access
  3. includes multi-step conversation flows
  4. works end to end

As the name indicates, we will build the important pieces of the chatbot together, but you would obviously have to fill in the rest of the details.

Why build a prototype?

Building a quick prototype can help you assess three things:

Is Dialogflow the right choice for the bot you are building?

Certain types of chatbots are too hard to build in Dialogflow. My rule of thumb is: if you are trying to perform some webhook logic where you are replicating Dialogflow’s NLU capabilities, it is a sort of reinvention of the wheel. And Dialogflow is probably not the correct choice for such a bot.

For example, if you have some very unique and specific requirements in the way user utterances should be handled, you might be better off building the chatbot using RASA NLU. Remember though that RASA NLU is not very non-programmer friendly, which means your bot development and maintenance budget will be a lot higher. Interestingly, this means building a quick prototype is actually a low cost way to figure out how far you can go with low code tools such as Dialogflow before resorting to the more code heavy bot frameworks like RASA NLU.

What types of questions/actions can be handled by a chatbot?

Another good reason to build a prototype chatbot is to see the kinds of questions that can be answered by a chatbot. FAQ chatbots are very good when it comes to answering about 80% of the questions that you get regularly. Combine this with a little bit of training, and you can move a majority of your FAQs into a chatbot, and handle only the rest with live customer support.


Once you go through the prototype bot service, you will also have a very good idea of the budget you will need for building your actual chatbot.

For example, you can answer all these questions once you go through the prototype service:

  • Will the Prototype Bot be sufficient for your current needs?
  • Can you maintain the bot by yourself without hiring a programmer?
  • If you can do both of the above, your bot development budget is a known, fixed amount – the cost of the Prototype Bot service
  • What kind of programmer should you hire if you need one?
  • Is Airtable capable enough as the database? (Airtable is usually sufficient, but sometimes it is not enough for some specific requirements)

Technologies we will be using

We will be using the following technologies to build out your chatbot:

  • Dialogflow Messenger as the integration channel (i.e. the user interface)
  • Airtable as the database
  • Python for the webhook using a collaborative coding environment

Free course bundle: Core Dialogflow

Everyone who books the website Prototype Bot service will also be enrolled into the Core Dialogflow course bundle for free. There are two reasons for this:

1 You need to create a flowchart for your chatbot, and for this you need to go through my Flowcharts course

2 I might refer to various concepts during our video call which you can learn from my courses. In other words, if you already have access to the course bundle we can go faster and get more stuff done during our video call

Also, you can later use the Core Dialogflow course bundle as a systematic way of learning Dialogflow.

How it works

1 First book the website prototype bot service using the link below. Choose a date which is at least a week away (you can also change it later).

Book the Prototype Website Bot service

2 I will send you an email which will give you free access to the Core Dialogflow course bundle

3 Go through the first few sections in the Dialogflow Flowcharts course and create a conversation flowchart using either Mindomo or XMind

4 Email me your flowchart (reply back to the email you get from step 2)

5 We will build the bot based on the flowchart during our live call