Home Analysis Pay TV can boost profits by 30% with best-in-class data analytics

Pay TV can boost profits by 30% with best-in-class data analytics

Photo: iStock/PeopleImages
Share on

It is possible for Pay TV operators to increase profitability by as much as 30% if they use data analytics (with AI and machine learning) across all parts of the organisation, execute well against new-found business insights and then test, iterate and optimise based on the outcomes. That stunning figure is given by Jacques-Edouard Guillemot, SVP Executive Affairs at Kudelski Group, whose company has a comprehensive analytics solution called NAGRA Insight. He has been outlining some of the areas where improved analytics can drive business performance.

Not surprisingly, customer retention is a priority application in Pay TV and that can be noticeably improved by getting people to use a service more. The role of analytics in improving the UI/UX, including with recommendations, is well understood and, at its most impressive, analytics can be used to personalise an entire content offering before your eyes, so the service is presented differently to each individual user, let alone household.

One NAGRA customer, Canal+ International, has made data analytics work hard for the business, including when it comes to customer retention. Matthieu Petit, Chief Analytics Officer at the Pay TV provider, used Connected TV World Summit last month to reveal how the company has simplified the data links between consumption, subscriber value and churn, and proved to executives internally that a focus on driving usage is the most efficient way to reduce customer losses.

“We have AI and machine learning algorithms dedicated to scoring the fragility of each subscriber, based on their consumption and interaction with the platform, whether across the set-top box or on [the Canal+ streaming service] myCANAL. There is a daily and monthly scorecard for each customer, and this is what we use at every single customer touchpoint.

“So, if a ‘fragile’ customer rings our call centre, we can propose the most appropriate downgrade or promotion for them. We know how to adapt their package. In France we also re-engaged people to make sure they are taking the right offer from us. Sometimes that means taking less ARPU from a customer, but keeping them.”

Canal+ has reduced its churn rate in France by nearly a quarter (from 17% to 13%) since late 2016 – and Petit gives the analytics advances much of the credit. Part of this integrated strategy has been to drive customer usage, and a machine learning (ML) algorithm has been used to improve content discovery.

Jacques-Edouard Guillemot (at NAGRA) lists upsell as another area where a Pay TV operator can improve the bottom line thanks to analytics. Again, this is linked to a knowledge of consumption. “When someone calls, the call centre operative knows what you like to watch in your household and which packages to push as well as downgrade,” he explains. “Maybe it has become clear that you love squash, and the operator has squash in their sports package.”

This understanding also helps the call centre provide the best ‘compensation’ if something goes wrong, like a problem with a set-top box. “You can be proactive. The operator knows someone has been without TV for three days. You can propose a package to them [that is relevant to their tastes], free for a month. These kinds of things are really helpful.”

Analytics can also be used to improve your return on content investments, Guillemot points out. One NAGRA customer made a 20% saving on its content costs after realising which assets were hardly used in its catalogue.

Guillemot stresses that the industry must not get bogged down in the technology of analytics, and should remember that AI and ML are just a means to an end. ‘Business performance outcomes’ is the language that CEOs and CFOs talk, not algorithms. He says you need to show them the money (revenue increases, cost reductions, for example). “You need to see the face of a CEO when you tell them that 15% of churn is because of black-screen on a specific type of device.”

Guillemot says the power of AI is that it “scales talent”. Thus, a marketing specialist who could only cope with one major advertising campaign a year can now run lots, and they can be more targeted, focused on granular customer segments. “With AI you can assess every campaign and recommend what is the best approach for a given segment, and optimise it for next time.

“In [Pay TV] marketing, AI allows you to measure and iterate your approach and your spending for each territory, each customer segment and every type of content. You can act differently. It gives you elasticity in how you run the business.”

Guillemot notes that you need buy-in at the top of the organisation to drive the use of data analytics, partly because applications often require inter-departmental collaboration. NAGRA is so confident that its solution delivers business outcomes that its pricing reflects this – remuneration is linked to results.

The most difficult part of a new analytics project is getting access to the data, he reveals, and this is the first step. The usual implementation model uses a limited data set before scaling later. From the moment the data is available, it takes 5-8 weeks to start measuring business impact.

There is a definite need for vendors like NAGRA to help the Pay TV industry exploit analytics, Guillemot reckons. Recruiting data scientists is hard and you need more than someone who can craft algorithms – they have to understand the TV business context and be able to interact with engineers. “Recruitment will be the biggest limiting factor on data agility in this industry,” he suggests.

Share on