Deetken Horizon is the machine intelligence
team at The Deetken Group.
Our approach is to combine leading edge
analytical techniques, including machine learning and economic
modelling, with deep experience in behavioural economics (i.e. the economics of
how firms and individuals make choices). This approach allows us to surface
keen insights to develop strategies and solutions that ultimately improve
services.
The problem
A healthcare client we were working with was experiencing challenges with the technical support they were receiving from a service provider. The support was being provided through a call centre.
One common issue was a “false close”.
This means that the service provider agent at the call centre would mark an issue - say slow login time or consistent freezing of an application - as resolved, even if the issue persisted for the user.
The impact
This was having an adverse impact on workflow and productivity. It was also causing a great deal of frustration on the part of users.
The service provider had a standard set of ticket data related to the calls they were receiving and how they were being addressed.
But these were not providing all the insight we needed to figure out why these shortcomings - these “service failures” - were persisting.
Our solution
We used a number of machine intelligence tools and techniques to try to uncover answers.
Topic modellingAn example of our work with an international client.
At the Intersection of Machine Intelligence and Behaviour Economics
There’s lots of great work going on in this space but Deetken Horizon is doing something different.
Many people work in the area of advanced analytics and a lot of firms have great skills in economic modelling and in the broader area of behavioural economics - the economics of how firms and individuals make decisions. But these worlds rarely collide.
We applied our advanced analytics skills and machine intelligence tools to get to the root of the service failures, which we identified as “false closes”.
Our work revealed that the service provider was incentivized to close a ticket as quickly as possible because their performance was being measured against how quickly they could address an issue. In other words, speed was one of their performance metrics.
The problem
The impact
Our solution
Topic modelling
At the Intersection of Machine Intelligence and Behaviour Economics
A Different Approach to Measuring Performance
