I have a survey bot on my site on each post, and recently, I got this question:
Can I use Dialogflow without writing code?
And the answer, as in many things, is “it depends”.
The FAQ bot
One of the first bots I created for this site was an FAQ bot which simply gives links to popular articles on some famous blogs. As it turns out, creating this FAQ bot needs no coding whatsoever.
And unlike the typical chatbot you would create, it needs no entities or contexts either.
In other words, it is just a whole bunch of intents. So much so, you can use Notepad or your favorite text editor, list some answer links, and a bunch of question links below them, and copy the links and questions over into Dialogflow to create a chatbot. And then you would choose the ‘Web Demo integration’ inside the Dialogflow console to get some HTML you can use and embed the chatbot into your website similar to how you might embed a YouTube video. That’s pretty much all there is to it.
There are some limitations with the existing web demo integration. For example, you cannot have clickable hyperlinks in them. But if you don’t want your hyperlinks to be clickable, you can simply use this approach and you will have a working FAQ chatbot which you can embed into your site, and you don’t need to write a single line of code.
Rich content in your website chatbot
As it turns out, if you need clickable hyperlinks, the existing web demo integration is not sufficient. And also, you suddenly go from zero code to having to learn how the Dialogflow API works to be able to display rich content on your website.
And say you are building slightly more sophisticated chatbots, say a chatbot which surveys your users and stores the data somewhere. As I mentioned in a previous article, you cannot perform any business logic inside your chatbot without using webhooks.
Using webhooks means you need to write code.
Reasoning about the chatbot
Suppose you managed to get some help and created a somewhat sophisticated chatbot without having to write any code. I wouldn’t call them sophisticated, but none of my demo chatbots from my step by step guide need code.
In that case, you will need someone on your team with some basic knowledge of Natural Language Processing and perhaps some Machine Learning to make sense of the chatbot’s behavior. In addition, if they are also programmers (which is quite often the case), they can develop tools which help you reason about your chatbot to a much larger degree.
And lastly, if you are having a chatbot out in the wild, some people are interacting with it and you would like to know more about these conversations. This is known as conversational analytics, and we don’t have a Google Analytics like tool to do this (yet). So once again, you are looking at someone who can write some code to help you out with analytics.
It is possible you are building a chatbot which doesn’t need webhooks, or need to use the Dialogflow API, or need to get analytics from your bot. That leaves you with very little you can do, but as I have shown in this article, it is possible.
Suppose you are building a somewhat sophisticated chatbot. If you are getting into Dialogflow thinking it has such a cool visual interface and you don’t need to hire those pesky programmers, you are probably in for a surprise. 🙂
It is possible Google, which now owns Dialogflow, has a lot of these tools being built out as we speak. At the moment, though, I wouldn’t recommend bot builders choose Dialogflow unless they are willing to bring a programmer on board.
About this website BotFlo1 was created by Aravind Mohanoor as a website which provided training and tools for non-programmers who were2 building Dialogflow chatbots. This website has now expanded into other topics in Natural Language Processing, including the recent Large Language Models (GPT etc.) with a special focus on helping non-programmers identify and use the right tool for their specific NLP task. For example, when not to use GPT 1 BotFlo was previously called MiningBusinessData. That is why you see that name in many videos 2 And still are building Dialogflow chatbots. Dialogflow ES first evolved into Dialogflow CX, and Dialogflow CX itself evolved to add Generative AI features in mid-2023