Diary of a CSP Retention Manager: How Predictive Analytics Fosters Long-Term Subscriber Loyalty

5 Mins read

August 2nd

Had a brutal meeting today. Our churn is up, and the team needs to totally rethink the way we target and retain broadband customers before they cut the cord altogether. The numbers are pretty grim at the moment. Our targets are very clear, we need to reduce churn by 15%, and boost CSAT scores by 20 points. And all by the end of Q2 next year! 


August 5th

I’m supposed to have a plan in place by the meeting tomorrow, but all I can find from my research is the bad news! The Telecom industry has been hit harder than almost any other since the COVID-19 pandemic, loyalty is down 22%. And then even if you break down different types of CSPs, internet service providers are known to score the worst CSATs!

I keep thinking that we need to get a better view of what’s actually happening in the connected home. I already know that a lot of our subscribers don’t want to pick up the phone, so we really need a solution that gives us that missing visibility into what’s happening. I think a lot of our current resources are being wasted, and with a bit of context – we could see what’s working and what isn’t. 


August 29th

A lot has happened since the beginning of the month! The execs loved the idea of using technology to see what’s happening inside the home, and our team found a software solution that offers what’s called Contextual Intelligence which does exactly that. The technology is called Quality of Experience scoring, and we’ve agreed to start working on a POC with Veego on their AI-based technology. 

What’s great is, it doesn’t replace our existing data and customer analytics tools, it’s an added layer on top to augment it. Through a lightweight agent, easily integrated onto the home router, the technology collects information on what devices and applications are being used, and uses this alongside the ideal performance requirements for each app. From analyzing each session, every home gets its own QoE score, which translates all of their data and internet metrics into a dynamic score that describes the subjective user experience. 

Because this data is being collected in context, we’re only alerted to a problem if it’s impacting the user. For example, latency issues are only flagged if they are having an impact on a gaming session. There were more than a few happy faces in the team meeting when they realized that meant we could target support and reduce operational costs, let me tell you!

October 31st 

Wow, I can’t believe it’s only been three months since that stressful meeting about reducing churn. I remember being told my targets for retention, and feeling like the directors were expecting me to have x-ray vision into how subscribers were feeling. It seemed like an impossible task – and yet I’m starting to see exactly how I can make that happen with technology. 

On one dashboard, our team can now see all the homes that experience a consistently poor QoE score, and this is an obvious segment who we can consider to be in the “red zone.” We’ve already been reaching out to support subscribers in this cohort, either with intervention behind the scenes such as implementing band steering to solve choking issues, or more overtly with targeted outreach. Yesterday one of the team shared a list of homes with coverage issues where the problem can be solved with WiFi extenders, and we’ve handed that over to Customer Care. Bonus – it’s an upsell opportunity for us, too! Next step – a pilot project!


December 19th

Had a great meeting today about our progress. Really happy we decided to expand from a POC to a pilot that covers 100,000 homes, it really gives me a lot more data to use to show the technology’s worth. The main point I wanted to get across is that before we implemented Veego, we only had insight into what I call “loud” churn. Calls, complaints, service degradation that impacted a whole area. Now, we classify households in a whole different way. Sure, repeated calls to the call center, or low satisfaction scores are a part of it. However, we now have insight into the silent majority, those who are unhappy but never tell us about it. Those who simply leave, and don’t ever feed back what we could do better. They’ve been the thorn in our side for so long! 

The kind of predictive analytics we have now covers all households – so the company has much larger ground over which to perform retention strategies.


March 1st

The numbers are starting to reflect the hard work we’ve been doing, and the pilot has been even more successful than we anticipated. 25% of homes have shown a significant improvement in Home QoE score over the past 3 months. Got an email from my manager today with a “well done” for all the hard work. Looks like we’re definitely on track to get that 15% churn reduction by the end of Q2 next year. 

Our primary goal was to improve loyalty – and so we’re working hard on making that happen at the next stage. Plans are in motion to create  personalized offers and offer remediation at scale. I can see the potential through Veego that hundreds of thousands of subscribers (and the right subscribers) will soon be provided with targeted recommendations to improve service degradation, and how we will then  see those results reflected on churn reduction and improved CSATs. Literally couldn’t have imagined doing this without Veego.

We’re also uncovering patterns in the root cause of service degradations, for example specific network conditions that impact certain applications, or device upgrades where they are all having the same negative effect on QoE. That’s helped us to make larger targeted changes that have a real impact, putting resources towards projects that will move the needle on customer experience. 

As we’re tracking consumption patterns, we’re starting to think about additional segmentation, too. At the moment we’ve been focusing on identifying suffering homes that can help us reduce churn and increase loyalty. But it’s so easy to see on the dashboard who are our streamers, our gamers, those who work from home, or those who rely on smart devices. We can use this data to offer greater personalization across the board, which is a really exciting value-add. 


June 22nd

Signed the paperwork today on 5 million additional homes to support with contextual intelligence. It was an easy sell, as we’ve hit and surpassed our original targets for CSAT scores and churn reduction, and added so much value to the business besides. 

What a difference 9 months makes! We’re seeing less direct complaints to the call center, stickier customer relationships, glowing NPS and CSAT scores, and a reduction in operational costs, too. Can’t complain about my surprise summer bonus package either! Super excited to see what we can do with Veego technology by the end of the year! 

Want the same diary entry by the time next summer rolls around? Schedule a demo of the Veego platform and see what Contextual Intelligence and Predictive Analytics can do for you.

You May Also Like:

Skip to content