Danish operator TDC has decided to implement a state-of-the–art analytical factory as 34% rate personalisation as the number one priority for customer centricity.
“Our analytical platform has enabled us to crunch and explore big OTT data,” Jonas Munk, head of predictive analytics, TDC, said during the Nordic Media Summit. “We use machine learning to detect consumer preference patterns beyond human capacity. We developed three recommender system prototypes, in order to enhance the customer experience.”
Sophisticated machine learning spot patterns hidden for the human eye, e.g. clusters of movies that are watched together. “Where associations are strongest cluster appear, the underlying drivers of recommendation. So for instance, people who watch new movies, continue to watch new movies. I like the machine to tell me how these combinations should be.”
Hidden insights into consumption behaviour show the type of movies watched and the relation to the time of day they are watched. “We start to see patterns of behaviour and gain insights into ‘time of day’ viewing patterns.”
TDC can now manage big piles of data, and is now ready to go into the next level of recommendations using more data sources, including type of device, time of day, day of week, location of viewing and more. Also, TDC is looking to embed recommendations in more touch points to activate the user, using Facebook, emails and the website.