Using n8n for chat bot

Hi n8n community!

I’ve spent the past little while experimenting with n8n and it’s been a really cool tool to work with! I’ve been working on a prototype chat bot workflow which has so far seemed very promising for what we’re hoping to accomplish. As I continue to learn this tool, I have a couple of questions that I am hoping to get some quick feedback on.

But first, a little more context: We are hoping to build a menu-based chatbot that will interact with users over WhatsApp via the Twilio API. I’ve been working on a very simple message flow to start, but we are hoping that we can use n8n to quickly build multiple more complex message flows. For example, we want to collect user information over several messages and save this to our system. In terms of deployment, the plan is to run n8n in Kubernetes attached to a DB.

  1. From the context provided, does it seem that we would be abusing this tool for this use case?
  2. Are there any considerations we should be taking regarding scalability? Our intention would be to use this tool for an early platform proof of concept and some fairly limited initial scaling.

Any and all feedback is much appreciated!

Welcome to the community @ddotzert!

To your questions:

  1. Honestly, n8n was not built for this use case. But did hear of more and more people creating chat chat-bots with i. So if it works fine for you why not. Anyway, because there is still some functionality missing, like being for example able to wait for a response within a workflow, is the bot-development-experience for sure not perfect right now.
  2. As long as all the workflows get started via the Webhook-Node, (which I guess is the case here) should it not be a problem. You can simply scale it horizontally. So you start as many n8n instances you want and then put a proxy in front of them which evenly distributes the traffic among them.

I hope that helps!

2 Likes

Thanks for the response, Jan! This has validated our initial approach and I’m looking forward to using this tool more in the future!

That is really great to hear. Have fun!

1 Like