Your brand needs a virtual assistant, and you have the opportunity to build it — that’s great. But one of the burning questions on your mind right now might be — where do you even start?
The first step that you absolutely need to get right is use-case selection. Picking the right use-cases before you start investing in building a chatbot is critical to the success of your bot. Building a bot is a time-consuming and extensive exercise and spending all that effort building the wrong use-cases could be detrimental to the success of your Conversational AI Journey.
“If you don’t have the time to do it right, when will you have time to do it over?” — John Wooden
But figuring this out isn’t as easy as it sounds. It usually takes weeks and months of intensive planning, research, and ideation. You would have to talk to multiple teams, manually go through heaps of past conversations, tickets, calls, etc. to arrive at your key use-cases & customer queries.
Get it right the first time with Intent Discovery
You already know what your customers want and ask for. Your brand has been interacting with customers through tickets, existing chat data since the start. So why not use existing sources of information to build your bot and get it right the first time itself?
Intent Discovery uses your existing information sources like your previous chat transcripts, support tickets, and knowledge base to train your bot. Your chat transcripts and support tickets act as a corpus of information that will act as a guiding rod to help us select top use-cases. It also can be used as training data so that you don’t have to individually add intents to the bot. Not only will it eliminate your manual guesswork, but it also gives you a faster time to value by 3–4 weeks.
Don’t take our word for it — try it yourself.
- Go to your ticketing or helpdesk portal.
- Export a CSV file of support tickets or past chatbot conversations from the last 2 months.
- Upload it in the form below
Within 24 hours, we will send you an email report from our AI-driven clustering analysis containing all the top potential use-cases that your customers might ask your bot. Not just that, to simplify your life further, we also provide the sample queries which act as top-class training data for the chatbot.
Let’s understand how Intent Discovery works
Past Support tickets — Your support tickets are the biggest pool of all your customer queries asked in the past. What could be a better way to feed your bot with all the top use-cases basis exactly what your customers want. All it takes is to upload your tickets in a CSV file and then we tell you exactly which use-cases or intents your bot should be trained on.
Past Chat Records — If you already have a current chatbot but want to build a new one — we’ve got you covered. Instead of starting from scratch, you can use all the chat records from your current chatbot to train your new bot. Even better? You can also use your live agent chat records to provide additional training.
To Sum Up
We understand that building a great Intelligent Virtual Assistant can be time-consuming and tedious for you and one of the biggest challenges might be not knowing where to start.
Well, not anymore. With our Intent Discovery feature, you can leverage AI to arrive at the right use-cases by simply uploading existing customer tickets & conversations. To learn more schedule a demo with our product experts.
Originally published at https://www.haptik.ai on February 24, 2022.
Start building your bots instantly with Intent Discovery was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.