Chatbot IntentsChatbot Training

Why do your chat and voice bots need to understand intent?

By ian tomlin | Jun 09, 2021

Why do your chat and voice bots need to understand intent? Cover Image

A new frontier in customer service

Customer Service Experience is the most frequently targeted area for digital transformations in the corporate world today. This is understandable. In a world that has come closer thanks to mobile telecommunications and the always-on web, buyers want to serve themselves with products, services and knowledge at any time of day. To achieve that level of service delivery capacity, firms are turning to bots.

To maximise the effectiveness of bots, they need to be trained not simply to react to known passages of conversation, they need to appreciate what the enquirer really wants.

What is intent all about anyway?

The biggest challenge facing the bot training industry is not the by now infamous ‘false positives’ — i.e., those moments when a bot thinks it knows the answer but posts the wrong one so far as the enquirer is concerned — it happens when bots have no clue what enquirers want.


A ‘missed intent’ happens when a bot doesn’t comprehend the intentions of the enquirer. This happens a lot, but thankfully, less and less the more your bot is trained.

Educating bots to understand intents

Sometimes, brazen headlines about the power and potential of artificial intelligence can leave you feeling that AI can solve anything. And who knows, in the future that may prove to be true. But, when it comes to bot training, much of the know-how to train bots has more to do with linguistics and the understanding of Natural Language Processing, than it has to do with AI. To interpret enquirer intents demands that bots know both WHAT is said, and WHY it is said. The reasons behind a request can massively influence the kind of response expected by the enquirer.

For example, consider these two requests:

‘I need a new credit card with a card strip that reads reliably cos mine doesn’t’, and:

‘I need a new credit card with a card strip’

It’s not uncommon for a bot to pick up the majority of words and phrases from this kind of enquiry to be picked up by an AI learning bot and assume the requester wants a new credit card. These slight nuances of intent–of emotion, of linguistics–can make all the difference between good and bad customer experience.

Maintenance service concept. Funny electrician robot handyman ready for repair housework. Robotic toy character with hand wrench, blue gray background. empty space for design

Interested to learn more? Ask us for an online walkthrough demo or request a download of our demo video.

How INTNT.AI Works

INTNT.ENGINE by INTNT.AI puts intelligence into your chatbot to know what site visitors want.  It employs a proprietary (patent-pending) model to predict user intentions by analyzing conversations to develop and evolve chatbots faster.  Using the INTNT.ENGINE takes 20x less human intervention. Consequently, chat and voice bots are smarter, and evolve 20x faster than if you relied on manual training. To resolve the issue of missed intents, INTNT.ENGINE can:

1. Train the chatbot with new training phrases.

2. Equip the chatbot with knowledge of intents.