We do receive around 1000 chats per day. I would like to gauge the quality of chats, is there a way to automate it somehow to know the trend and common questions asked over chats?
If I am assuming that you have all these chats per day stored(If not storing these chats on a data warehouse for reporting purposes is an option). There is software such as Nvivo which is great for qualitative analysis which you can use for these tasks. There are also packages for R, Python and even Excel for analyzing data.
Answered 5 years ago
It all starts with a proper Quality Assurance proccess in place. Benefits of having QA in place are twofold: it can act as a tripwire to capture knowledge gaps or discrepancies with a product itself. So not only reducibg support debt, but closing support to product feedback loop.
While automation is the best option it usually starts with agents tagging conversations manually and doing monthly or quarterly deep-dives to identify top call drivers (areaa of the product that creates friction and stops customers from achieving their goals).
While you can set up conversation-topic tagging by keywords automatically - it’s far from accurate and data on trends become too clunky.
Answered 5 years ago
Yes, definitely. I've seen companies before use something called 'keyword search', so for example if a lot of people are searching for the word 'account' then you'll be able to start to gauge if there's a common problem with their accounts.
Other companies I've worked alongside have tasked someone to weekly/fortnightly/monthly take a look at some transcripts and try to understand the trends and common problems that are going on. With 1,000 chats a day, you'll have plenty of information to look through and even 100 conversations will give you an idea if there's a common problem as it's unlikely that there would be 100 separate problems.
I'd love to help further, but I'd need to know a little more about your business to help advise on a more direct basis. Feel free to get in touch if you'd like to know more :-)
Answered 4 years ago