Solving the customer experience disconnect
Old Mutual was the first bank to offer its customers a cup of coffee when they visited its offices. Some of its older customers would even pack sandwiches when they visited a branch so that they could have their coffee with the teller handling their financial queries.
Sharing this story during her presentation at CEM Africa in Cape Town this week, Sarina de Beer, client experience director at Ask Afrika, noted this is how close the business was to its client base.
The relationship was so personal that when the bank tried to move to a self-service model, it was these customers who expressed outrage at the idea because the tellers they befriended over the years would likely lose their jobs.
Yes, we need artificial intelligence (AI), robotics and chat bots to make us more agile, more responsive and more able to understand our clients and what they expect from our business, explained De Beer.
And data does offer a wealth of insights into clients and what they want. But we cannot disregard the human factor. These resources and innovations must be seen as an enabler; not the be all and end all. “Today, we are probably more connected than we have ever been, and yet, in the same breath, we are actually more disconnected than we have ever been.”
While using a popular app recently, De Beer came across an error and the only way she could address the issue was by communicating with an automated chat bot on the brand’s Web site. But, for whatever reason, the bot couldn’t provide the answers she needed.
“It’s all very well for me to be able to ask a question at any time of day thanks to the introduction of an AI bot, but the fact that this system couldn’t fix my problem left me feeling even more disconnected and made the brand feel even more inaccessible.”
According to De Beer, businesses want all of the answers from technology and data, which can make it very easy to get stuck in what she dubs “data fundamentalism”. This phenomenon sees businesses believing data can do anything and everything for them. It’s all very well to have large data sets but if it can’t be interpreted in a meaningful, the organisation could be making some real mistakes when attempting to satisfy customers’ needs.
Offering an example, De Beer explained that she recently helped a colleague find an adventure boot camp class in her area and later found herself inundated with adverts for the outdoor exercise programme. “Anyone who knows me, knows that I will absolutely die before I do boot camp.”
Nothing in the data that is associated with her online persona indicates she would have any interest in a boot camp, and yet, one search and she was drowning in promotional content.
“This is because a machine can’t apply logic; it lacks context and an algorithm can only do so much. The only way to prevent these interpretive errors is for businesses to be close enough to their consumers that you understand their behavioural patterns,” De Beer noted.
Were a person analysing this data, they may have been better equipped to flag this search as an “anomaly” and would likely test these new assumptions before they started sending advertisements her way, she added. Technology can cause inaccurate assumptions about customers when what we should be doing is taking time to speak to people and asking them what they actually want.
“Don’t get me wrong, I’m huge proponent of AI, machine learning and chat bots. I think these tools are essential in terms of how we engage with all of the consumers out there and handle their expectations. But I don’t believe that any of these things can replace humans,” she concluded.
“As crucial as AI is in making us faster and more able to deal with information in a slicker way, the real power lies in the interpretation.”