11 October 2022
Data Driven Product Roadmaps: How customer experience can (and should) dictate technology choices
Successful products solve customer problems better than any competing solution. You’re taught this when you’re a product manager just starting your career. As an experienced product manager, and the VP of Product Development at Telenet, Stijn Eulaerts has developed numerous product roadmaps. The approach of building data-driven product roadmaps has allowed Stijn and the Telenet team to successfully solve customer solutions (yes, better than competing solutions!).
In a recent Beyond Digital session, Stijn shared the differences between technology-driven product roadmaps and data-driven ones, and how Telenet keeps customer experience and data at the core of their approach.
Letting data (and CX) drive the roadmap
“A roadmap for a product starts from a customer need. With a technology-driven roadmap, or a traditional roadmap, a lot of the roadmap planning comes from new technology that becomes available. This is typical in my business; talk to the Nokia’s, Alcatel’s and Erickson's in this world, and probably their product managers will come up with a great new feature to bring to market because it will change the world of their customer. How is this different my data-driven product roadmap? For me, we start from the same customer problem to solve. But the roadmap starts evolving, not based on what your technology partner is telling you, but based on what the data is telling you. And the data is, of course, based on what your customer is giving to you.”
(Want to join the next meeting of Beyond Digital to learn, share, and shape the future of digital in Belgium? Sign up for more information here!)
Escaping the product research trap
“So, if we want to remain relevant in the future, we better improve our ‘solution’ to the customer problem. The first rule of product management is always ask your customer. But when you do ask your customer, you need to realize there's such a thing as the product research or the product trap. Take the example of Henry Ford who said ‘if I had asked people what they wanted, they would have said faster horses.’ This is the typical product research trap. We see here that with product research, you have to ask the right question. The questions you need to ask are not about your solution; the question you need to ask your customers is ‘what is the problem that I can solve for you?’ In Henry Ford’s case, probably customers would have answered, I need to get faster from point A to point B.
“When you’re asking the right questions, you start discovering that the solution space is much bigger, and it goes well beyond just the use case itself. It's about how you discover the product. It's about how you buy it, how you get it delivered, how you install it. So product teams really start looking at products in an end to end journey. And once you realize the scope of your product is this broad, you realize you also need a whole different set of metrics to start measuring the success.”
Measuring the journey
In the case of Telenet’s latest challenge, Stijn knew where to start. “We went from one key metric that our engineers love to measure, looking at the modems and asking: do we deliver the fastest internet product? And we lay out the roadmap to improve that metric. Of course, there are a multitude of metrics measured across all the steps in the customer journey. Based on how these metrics are performing, we can decide where to improve the product – and suddenly, your roadmap can look completely different.
“Now, the good thing about metrics is that they give you insights into the past performance. But the bad thing is also that they don’t predict the future. And if you’re only collecting customer input, while a small set of customers will contact you with complaints, the majority won’t warn you when they find a better solution. Our response has been to build predictive data models.
“The good thing about metrics is that they give you insights into the past performance. But the bad thing is also that they don’t predict the future... Our response has been to build predictive data models."
“We have built over 10 different predictive data models that tell us how the customer is using our product and what kind of experience or potential issues they might have in the future.
"These are very technical models that look at several elements in the network that help us deliver your internet service. We track them at an individual customer level because well, the type of usage that I have is very different from that of my 77 year old mother, yet she is essentially buying the same internet solution.”
The culture behind the data
Stijn is adamant that data should be neutral. "It’s important to educate the entire organization; data can be abused and it’s important to have everyone speaking the same language. We want to preserve that the data is the customer’s voice. This is a culture shift we make across the entire organization. For example, we can get a tremendous amount of insight from our technicians in the field when we’re working the right way. Taking all of this intro account, these approaches have allowed us to make bold choices, of which we could predict that these investments would make the biggest difference for our customer, and of course, solving their problems better than anyone else."
Want to join in on the discussion? Ariad’s digital professionals knowledge sharing club Beyond Digital brings together future-forward individuals working across industries in the largest brands in Belgium. No member fees, just a community of ambitious minds! Apply here for more information about this monthly event.