Behind seamless streaming: how data powers better experiences

Introduction Introduction

Viewers expect streaming to simply work. One moment of buffering or a single error can break that experience, and once frustration sets in, audiences disappear quickly. Video and app analytics give platforms a clear picture of what viewers are actually experiencing, and they make continuous improvement possible.

At Hypersolid, we partner with Conviva to help platforms understand and optimize their performance. We spoke with our Tech Lead, Jesse van Erkelens, about how data shapes better media platforms and why analytics have become essential for delivering high-quality experiences.

InterviewInterview

Q: How did your journey at Hypersolid begin?

I joined Hypersolid as an iOS developer and spent the first three years fully focused on mobile development. Over time, my interest shifted toward the broader process: how teams collaborate, how decisions are made, and what happens long before anything reaches development. I was already helping translate client and product owner wishes into clearly scoped tickets for the development team, and that’s when I realized I wanted to move into a more leading role.

After sharing that ambition internally, I experimented with a combination of project management and technical leadership. In the end, the technical path was the best fit as it allowed me to stay close to the craft. That’s what set me on the tech lead track, a role I’ve now been in for almost seven years.

Data gradually became a bigger part of my work as well. Through our collaboration with Conviva, I moved deeper into the data domain, creating measurement plans, shaping new ways of working, and supporting our data analyst with both technical expertise and client consultancy. When our analyst left, I even stepped into the role temporarily. That period really deepened my understanding of data, not just from a leadership perspective but through hands-on analytical work as well.

Q: You mentioned that your work with data really took off through Conviva. How did you first get involved with them?

My experience with Conviva goes back eight or nine years. At the time, NLZIET was already a Hypersolid client, and they wanted to start using Conviva for streaming analytics and involved us in the integration early on. As part of preparing the integration of Conviva into the NLZIET app, I joined a workshop from Conviva in Hilversum with another developer. I still remember that day clearly. It was storming so badly that public transport shut down , and we had to take a taxi home.

Back then, I was still working purely as an iOS developer, and Conviva was my first real introduction to video analytics. It immediately stood out from traditional analytics. Instead of tracking general behavior, you look at performance and usage: how the player behaves, what the viewer experiences, and where improvement opportunities lie. That focus on concrete improvements became something I applied continuously at NLZIET.

Q: What was the learning process like when you first started working with the tool?

That first workshop was a one-off, so everything after that I learned step by step by working with Conviva in practice. By applying it across different projects, I gradually got a real grip on the platform and grew into what Conviva itself would call a “power user.”

Over the years, I’ve worked on several major initiatives, including full platform rebuilds and first-time app launches. In the early years, I handled these integrations as a developer and later as a tech lead. This progression naturally deepened my understanding of the platform and strengthened my expertise.

Along the way, Conviva has continued to expand, and working with it regularly has helped me stay closely aligned with how the platform is structured and used in practice.

Q: How do you stay up to date with new platform features?

During integration projects, contact with Conviva is frequent, with regular bi-weekly calls to stay closely aligned. Once that phase is over, the cadence usually shifts to monthly sessions. In those meetings, Conviva walks through the tooling, explains new features, and often shares practical use cases. I always make time to explore updates myself, partly out of curiosity and partly to understand whether they add value for our clients. Sometimes a new feature immediately creates an opportunity to advise a client in a way that wasn’t possible before.

A fun part of this journey has been participating in Conviva’s alpha-release program. When Conviva was working on the next evolution of their platform, NLZIET was open to joining the program and exploring a transition from Google Analytics to Conviva Digital Product Insights (DPI). We guided that transition and evaluated what worked, what didn’t, and what we liked about the tooling along the way. Conviva applied several of our suggestions, and it’s great to see that I was able to influence how the tool is used today.

Because of that involvement, NLZIET was invited to a Customer Advisory Board meeting in the United States. Their CTO couldn’t make it, so he suggested I go. Suddenly I found myself in a room with CEOs and CTOs from major international broadcasters.

Moments like that show that being a Tech Lead isn’t just about integrations. It’s also about working with the tooling, interpreting the data it generates, and helping improve the platform itself. Today, that experience comes full circle at Hypersolid, where I run Conviva workshops for the media team so they can use the tooling effectively, advise clients, and spot early when something isn’t performing as it should.

Q: What makes Conviva such a great platform to work with?

What stands out is the level of accuracy and the ability to drill down into the data. Filtering and comparing the right data sets is essential when investigating issues or identifying improvements. With this level of detail, even small performance spikes can be identified down to the minute.

Conviva is also incredibly complete as a product. It can be configured for different teams and roles. Marketing wants to know what people are watching. Support teams need to diagnose a specific user’s issue and inspect their individual session. Developers look at overall performance metrics to determine what can be improved or whether something got worse after an update. Conviva supports all those perspectives within a single platform.

On top of that, the tooling is highly configurable. Dashboards, alerts, and metadata can be tailored to specific teams, platforms, and business needs. By working closely with Conviva, we’re able to fully leverage the platform’s capabilities, making it a key part of our end-to-end monitoring and analytics setup rather than just another reporting tool.

Q: You mentioned anomalies. What happens internally when something like that occurs?

When an anomaly appears, we automatically receive an alert. There are two types. AI alerts are triggered when Conviva detects unusual behavior, such as sudden buffering on live content or a complete service outage. I set these alerts to post directly in our team channel so we can respond quickly.

Manual alerts are more specific. Instead of looking at averages, you define exact thresholds, for example when a certain number of errors occurs within a short period of time on a particular streaming provider. This approach is especially useful for applications where we deliver streams partially or entirely. It also allows us to set tighter thresholds, ensuring we don’t miss even small drops in performance. We monitor our own streams 24/7, and combining these Quality of Experience metrics with our Quality of Service monitoring gives us a more complete and reliable picture of overall performance.

Once an alert comes in, we look at the impact and which team member can take the lead. Sometimes I notice a repeat pattern. When that happens, I dig deeper. I look at when the issue began and how it behaves. Sometimes I immediately recognize the cause. Other times I reach out to a developer and say, “I am seeing increased rebuffering since the last update, let us investigate.” Then we look into it together.

Q: What are some of the latest developments happening behind the scenes?

Conviva is no longer just about video analytics. The company now offers app, website, and AI agent analytics through its Digital Product Insights offering, providing deeper insight into consumer experience, behavior, and outcomes across the digital landscape. Conviva overcomes the limitations of traditional product analytics in hyper-personalized web, app, and agentic experiences by automatically discovering stateful patterns that drive outcomes, identifying the strongest drivers, and prioritizing them based on business impact for planning and decision-making—rather than relying on manual tagging and pre-built funnels.

Conviva understands that simply collecting data is no longer unique because everyone collects data. The real value is in activation and in turning the data into actionable insights about performance, behavior, and engagement. That is where their current focus lies.

On top of that, Conviva introduced Nexa an AI agent interface into the platform. You can prompt it to analyze data and produce quick reports or tables. You can ask something like, “How did performance yesterday compare on iOS versus Android, and where do you see opportunities for improvement?” Nexa generates the table and highlights the most important factors that influenced performance.

Before Nexa, I had to apply filters multiple times, adjust date ranges, and manually dig through the data. Nexa can save a lot of time, and I hope clients and teams will begin using it more actively. With broader adoption, it could reduce the number of questions I receive, especially since most people are already familiar with writing simple prompts.

One of the more recent additions is support for connecting to Conviva through its MCP. This allows us to query Conviva data directly from LLMs, integrate insights into dashboards and reports for diagnostic purposes, and automate internal workflows, such as enriching alerts or triggering Jira tickets, using AI-driven processes. Given its potential, this integration is on our media team’s roadmap for 2026.

Q: Which Hypersolid clients are currently running on Conviva, and how quickly can we implement the tool?

Right now, the list includes NLZIET, ESPN (GOAL! app), Ziggo Sport Totaal, and Talpa. Talpa is the most recent one, and I led the integration for them. I support the developers, answer questions, and guide the process. We also passed the qualification phase with a major European broadcaster, which means we can join their tender together with Conviva as their integration partner.

The integration for video analytics usually takes about one month in total lead time, not one month of development. Typically, the initial integration takes a week, followed by delivery to Conviva. Then we spend a week validating everything, another week processing feedback, one more round of validation, and then we are done.

That is the normal scenario. Some platforms have limitations that make the process more complex. Smart TVs are a good example. Third-party players can also take longer. The advantage of those players is that they often include a pre integration of Conviva. The disadvantage is that if something does not work, we need them to release an updated SDK. The positive part is that we know these players well, and communication lines are short, which speeds things up.

The Conviva DPI integration, which focuses on precision analytics, requires relatively little effort from developers. Rather than just collecting data, DPI helps translate performance, behavior, and engagement metrics into actionable insights that can be tied back to business goals. When auto-tracking is used, the integration can be completed within a day, with only a few additional days required for customization. After configuring events, flows, dimensions, and other settings, meaningful insights can be available within a matter of weeks.

Q: What do you think is Conviva’s added value in working with Hypersolid?

Hypersolid is one of Conviva’s most trusted partners and possesses comprehensive, hands-on knowledge of Conviva Video Streaming Insights as well as Digital Product Insights. We have mastered the platform from both an implementation and a data consumption perspective, and have extensive experience with the wide variety of custom platforms that Conviva integrates with. That creates a natural exchange. We have implemented the tool many times and have close contact with end users, so we see firsthand the value it brings. That makes us fast, and it helps us understand the impact it creates for clients.

We are currently running demos for one of our clients, and I see how enthusiastic people become. Tasks that used to take hours now take minutes. During the demo, I can assemble charts with just a few clicks. It is simple and very fast.

Conviva is a powerful solution for video, app, website, and AI agent analytics, yet it is one of many tools our clients rely on. At Hypersolid, we follow a best-of-breed approach, which means we work with whichever platform our clients choose. What truly matters is understanding what your users are experiencing and turning those insights into better performance and better results.