For much of the past decade, the technology industry has moved in one clear direction: toward the cloud, third-party platforms, and increasingly centralized digital infrastructure. Companies traded control for convenience, handing over more of their data in exchange for tools that were easier to deploy, scale, and manage.
But as artificial intelligence reshapes how businesses build products and interact with customers, that trade-off is coming under renewed scrutiny. AI models are becoming more accessible and increasingly commoditized, shifting the competitive advantage elsewhere. The question is no longer simply who has access to the most powerful model, but who owns the proprietary data needed to make that model genuinely useful.
For Onur Alp Soner, co-founder and CEO of Countly, this is not a new conversation.
A technologist and self-starter, Soner bootstrapped Countly from the ground up with a different philosophy: companies should be able to understand and interact with their users without giving up control of their data. Built as an open-source, self-hosted digital analytics and in-app engagement platform, Countly challenged the prevailing assumption that centralization was the inevitable price of scale and innovation.
Under Soner’s leadership, Countly has grown into a trusted platform for enterprises worldwide that want to innovate quickly while keeping user privacy at the centre of their growth strategies.
Today, the principles behind that early decision are becoming increasingly relevant. As organizations rush to become “AI-ready,” many are discovering that collecting vast amounts of data is not the same as being able to use it. Fragmented systems, third-party platforms, and a lack of genuine data ownership can all stand between companies and the AI ambitions they hope to achieve.
In this edition of TechNews180’s Founder Series, Soner discusses why proprietary first-party data could become one of the most important competitive advantages in the AI era, what executives misunderstand about AI readiness, and why the best technology may ultimately be the technology that quietly fades into the background.
1. For more than a decade, the tech industry moved toward centralization: put your data in the cloud, use third-party platforms, outsource complexity. Countly was built on almost the opposite philosophy. What did you see that everyone else missed?
I don't think we saw something everyone else missed. At the time, people were making a reasonable trade-off. Cloud services were easier to use, easier to scale, and often free or inexpensive because your data was part of the business model. However, we simply weren't comfortable with that trade-off. We felt organizations should be able to understand how their products were being used without giving that data to someone else.
That's why Countly was built as an open-source, self-hosted platform from the start. The technology has evolved a lot since then, but the motivation hasn't changed. We're still focused on building infrastructure that gives organizations real control over their data, because we believe that ownership shouldn't be something companies have to trade away.
2. You’ve argued that AI models are becoming commodities while proprietary first-party data is becoming the real differentiator. Do you think we're heading toward a future where companies are divided into data-rich and data-poor organizations?
To some extent, yes, although I think it's more nuanced than simply data-rich versus data-poor. Most organizations already have plenty of data. The bigger question is whether they understand it, control it, and can actually use it.
In the end, having large amounts of data doesn't automatically create an advantage. If it's fragmented across different systems, poorly structured, or locked inside third-party platforms, it's difficult to build reliable AI on top of it.
As AI models become increasingly commoditized, competitive advantage will come less from the model itself and more from the behavioral data that reflects how your own customers use your products. That's something no one else can replicate, provided you actually own the data and have the ability to put it to use.
3. A lot of companies claim they're AI-ready because they've collected massive amounts of data. From your perspective, what's the most common misconception executives have about what it actually takes to build effective AI systems?
A lot of the conversation starts with the model, when it should really start with the data underneath it because that's ultimately what determines whether AI creates value. In many organizations, that data is spread across different systems, sits inside third-party platforms, or has been collected in completely different ways over the years. Before AI can work reliably, companies first have to bring that together.
That's why I think companies need to spend as much time thinking about the data layer as they do about AI models. Whether AI delivers value ultimately depends less on the model itself and more on the quality and organization of the data behind it.
4. Founders often start companies to solve one problem and end up discovering a much bigger one. What problem do you think you're actually solving today, and how different is it from the one you started with?
The core problem hasn't really changed. We've always believed organizations should be able to understand and use their data without giving up control of it. What has changed is how that problem is perceived and how important it's become: when we started, analytics was mostly about understanding what had happened through page views, sessions, and user numbers, whereas today organizations expect data to help them improve products while people are actually using them.
5. For years, privacy was often framed as a compliance issue. Today, it's increasingly becoming a competitive advantage. Do you think the market finally caught up with Countly's original vision, or are we still in the early beginnings of that shift?
I think the market is moving in that direction, but I wouldn't say it's fully there yet. Privacy is no longer viewed only through the lens of compliance as more organizations are starting to see that ownership and control have a much broader impact on how they build products and use AI.
At the same time, many companies are still working with architectures that were designed years ago, when ownership wasn't really part of the conversation. Changing that isn't as simple as replacing one analytics platform with another.
6. Imagine it's 2038 and Countly has achieved everything you hoped it would. What do you hope the company's lasting impact is—not on analytics software, but on how organizations think about ownership, trust, and technology?
I hope we've shown that organizations don't have to give up control of their data to build great technology. That's the idea Countly was built on, and I think it's even more relevant today than it was twelve years ago.
I'd also like Countly itself to become less visible, not more. The best technology often fades into the background. If Countly quietly powers products that adapt in real time while allowing organizations to keep control of their own data, I think we'll have achieved what we set out to do.
