AI Case Study
Bank of America improved call centre completion time by 23% by tracking and analysing behavioural data with machine learning that identified employee performance increases when they network and share ideas
Bank of America used smart badges to track call centre employee activity with smart badges. They recorded details such as employee office movement, who they interacted with, their tone of voice and whether they leaned into listen or leaned back. The data was analysed with machine learning and they identified that "network cohesiveness" could drive performance as employees were more likely to share ideas and help each other. This resulted in improved call centre completion time by 23%.
" A smart employee badge with a microphone, accelerometer, bluetooth connection and other tools typically found in a smart phone. (In fact, it looks a bit like a cell phone.) It measures things like how people moved through the day, who they interacted with, what their tone of voice was like, if they leaned into listen and other types of interactions that happen at every company every day. Each badge generates about 4GB of data per day, which gets uploaded to the cloud where Humanyze breaks down the 40 types of information it collects and pares it down to the 6 most important pieces for the client. They then link that to a key business metric dashboard, so the customer can see how behaviors are affecting the bottom line performance of the company."
The results reported were as follows:
* Network cohesiveness, a measure of how well a team communicates went up 18%
* Reduced stress (as measured by tone of voice) by 19%
* Happier employees and turnover rates went down by 28%
* Call completion time improved by 23%
However the results do not report on the impact of employees being observed constantly on job performance.
Humanyze records behavioral aspects of employees using a smart badge. This data is used to map correlations between communication patterns and thereby deriving insights about employee and team productivity.
Each badge generates about 4GB of data per day, which gets uploaded to the cloud where Humanyze breaks down the 40 types of information it collects and pares it down to the 6 most important pieces for the client.