Chatbots, schmatbots. When you’re talking contact centres, there are far more powerful applications for AI coming our way, predicts Shad Mortazavi, Managing Director, OpenIQ
Today’s contact centre bears little resemblance to its predecessor, the traditional call centre of yore. It’s likely to be hosted in the cloud, rather than on-premises, for starts. And, odds are, it will boast multiple channels – phone, email, webchat, SMS and even video – for businesses to connect with their customers, and vice versa.
Chatbots – computer programs which simulate human interactions and conversations – have become unremarkable and, in 2021 Australia, they can frequently be found embedded in contact centre messaging apps and telephone systems. They’re powered by Artificial Intelligence (AI) and can be incredibly useful when it comes to handling straightforward enquiries. And incredibly frustrating for customers when even a modicum of real intelligence or reasoning is required.
To date, this has been the primary use case for AI in the contact centre – its ability to power a form of low level human replacement, one that can engage in basic interactions in a ‘kinda human’ way.
Building on the basics
It’s a solid first step but I believe there’s a lot more value this technology will be able to add in the upcoming years.
One of the ways it’s likely to do so will be by helping companies use the wealth of data collected by their contact centres to do two things: measure customer satisfaction continually; and predict customer acquisition and attrition patterns.
Historically, getting a handle on customer sentiment has necessitated the gathering of feedback directly from the source. Companies typically email or text customers shortly after an interaction with the organisation has concluded to ask them for a 1-5 rating.
These mini report cards can provide some insight into how customers are thinking and feeling about an organisation and the quality of service it delivers – always providing people are prepared to spend the 30 seconds or so that’s required to complete them.
Many people are not. In 2021, ‘survey fatigue’ has well and truly set in and response rates as low as 10 per cent are now common. Customers may be willing to give you a score every six months or so, but they certainly don’t want to do so each and every time you, or they, initiate contact. Meanwhile, plenty can go wrong in the interim. A customer who gave your organisation the big thumbs up at first contact may be well and truly cheesed off by the time they’re next asked for an opinion, perhaps 10 or 20 interactions down the track.
Finger on the pulse
AI could help companies manage this challenge more effectively by enabling them to monitor their contact centre interactions at a far more granular level; comparing transcripts and agent scores with customer outcomes.
Modelling these results over time would make it possible to detect trends and predict customer feedback and satisfaction levels, without having to poll customers more frequently.
Armed with this intelligence, companies could tailor their interactions more effectively and intervene early if customer sentiment appears to be heading in the wrong direction.
Yes, it’s horizon technology right now but it won’t always be. In the high tech world, today’s blue sky thinking is tomorrow’s reality. I believe that by 2023 we’ll see companies exploring this technology to the full, to deliver better service and more satisfying customer experiences. Banks and other financial services providers are likely to be in the vanguard but forward thinking organisations across a range of sectors are likely to be hot on their heels. Meanwhile, those that don’t elect to make full use of the plethora of data the modern contact centre places at their fingertips may find it harder to hold their own.