Alex Dobromir on Product Thinking, AI, and the Future of Continuous Insights

Alex Dobromir, Product Manager at Product Hub by MMR Research and a 2026 Future List Honoree, is helping reshape how insights technology is built and used. Combining behavioral science, product management, and research innovation, he focuses on turning fragmented research into continuous learning systems that evolve over time. Passionate about agile development, data infrastructure, and human-centered technology, Dobromir advocates for faster, smarter, and more scalable insight generation across the industry.

Outside of insights, what are your passions and interests?

I’m driven by a mix of endurance, curiosity, and creative work. I’m an avid cyclist and every year I take part in week-long trips of around 300–500 km across the Netherlands, Belgium, and Germany. My personal one-day record is just under 100km, and this year I’m aiming for 300 km in four days. I also enjoy running and have structured training goals. So far I completed a quarter marathon in under one hour, and my target for the year is a sub-2-hour half marathon.

Beyond sport, I’m passionate about coaching and helping people grow. I coach colleagues in product management as part of my current role at MMR, and I also support students from Romania with their international university applications. I’ve spoken at alumni events for my former universities about careers in marketing and research. When I can, I like to give back to people who are starting out. Last but not least, I’m into photography (analog and digital) and am an avid film enjoyer, as they help me capture and imagine how people think, behave, and experience the world. I have a collection of personal photographs from last year that I am working on curating at the moment. 

Since starting your career in MRX, what would you consider to be your greatest accomplishment? 

My biggest accomplishment in MRX has been applying product management principles to shift how research technology is built, moving from a waterfall, delivery-first approach to agile, user-led development that reduces wasted effort and compounds learning over time. By introducing structured discovery interviews, opportunity solution trees and rapid prototyping, we were able to co-create with users, test assumptions quickly, and make data-driven decisions on what to pursue. These changes mean less rework downstream and a stronger focus on outcomes for our users, rather than outputs.

The results are strong: Product Hub has an NPS of 7+ and 0% churn in enterprise agreements which gave us a momentum that motivates us to keep building. When I started my current role as product manager for Product Hub, I focused on a problem I kept seeing across the industry: valuable product research data gets generated, but ends up buried in slide decks and forgotten, resulting in slow learning loops and repeated work.

For me, the real win is helping build a platform that turns insights into a system - something that gets smarter and more useful the more it’s used, and doing it in the most efficient way possible, especially for industries that have been left behind the digitalization curve such as physical product testing.

When did you know you wanted to enter a career in insights, and what inspired you?

I think knew I wanted to work in insights from when I was a student, when behavioral science stopped feeling like theory and started feeling like a practical way to improve real decisions. Reading Thinking, Fast and Slow by Daniel Kahneman and How Brands Grow by Byron Sharp was a big turning point as it made the impact of behavioral economics and marketing science feel obvious in everyday human behavior. In the past, I also worked on academic research projects across climate change and cognitive science, and I loved the rigor of research and the discipline of asking better questions. However, it sometimes lacked real life applications.

What really pulled me into research technology was seeing how much impact technology can have when it’s built around real user needs, which I learned by working in a startup studio. This felt like the right intersection for me: science, human behavior, and product thinking coming together to build tools that make insight generation faster, smarter, and more scalable.

How has market research changed since you first started your career?

I’d tell my younger self to stop rushing and overbuilding and committing to solutions too early. When we are young, we all want to make impact quickly.The biggest shift is learning to fall in love with the problem, not the end goal. In product management, especially research technology - you don’t win by shipping the most, you win by learning fast and building something that actually works for users, delivers value, and is viable for the business.

At Product Hub, we don’t start with “what we want to build,” we start with goals to achieve, like enabling better product testing insights, and we keep iterating until the definition of done is genuinely met (which can and sometimes does mean testing and rebuilding multiple times). The practical approach is simple: prototype early (low fidelity power users first, then high fidelity occasional users), run user acceptability sessions, and track whether the system is improving when launched - things like effort spent to achieve insight on a project, not feature delivery. Over time I’ve learned that learning is more important than shipping, because the best technology platforms evolve through many versions. Strong upfront design and repeated testing isn’t slow in the long-run, it’s how you build something that scales and stays useful.

Tell us about any advocacy/volunteer/association work you're doing within the industry. What issues are you trying to solve? Why is this work critical for the industry?

Advocacy is a significant part of how I approach my work. Through the Product Hub blog, conference speaking opportunities and client strategy discussions, I keep coming back to one core issue: wasted data and slow learning loops. The insights industry (especially product research) produces incredible information, but too often insights are treated as one-off deliverables - useful in the moment, then lost, and the next project starts almost from scratch. I’m focused on pushing the industry toward automated tech-enabled insight systems instead: databasing and tracking research outputs, benchmarks, action standards, and cross-study patterns, so that learning compounds over time.

With technology supporting better data architecture and AI improving retrieval and synthesis, it becomes more realistic to reuse knowledge at both product and portfolio level, rather than constantly rebuilding context. This matters because markets are changing faster than traditional research cycles can keep up with. Anti-fragility should be the goal: when market shifts happen, portfolios shouldn’t break, they should improve, because continuous testing and aggregated insights help teams spot trends early and respond with confidence for the next iterations or launches. 

Where do you see the future of insights heading in the next 10 years? 

In the next decade, I think insights will become more tech-first, more continuous, and more integrated into how organizations run, rather than something that happens on isolated projects. I see two major directions. One is digital twins: creating either audience segments or sensory digital twins of physical products and using machine learning models to predict consumer responses to changes, which can massively accelerate iteration and reduce unnecessary cycles by creating informed hypotheses and narrowing down the ideas/concepts/prototypes to be tested with users. This, in turn, will create more space (both in budgets and timelines) to understand the users more in-depth and to dig deeper into insights with expert AI-powered tools.

But the real shift won’t just be “more AI.” It will be moving from one-off studies to reusable insight systems where knowledge compounds through integrations, continuous benchmarking, and structured data foundations. The biggest risk is jumping into AI too early without getting the basics right: correct methodologies, the right data, good quality, proper databasing, and clear learning and business goals. Many of the best solutions won’t even be LLMs, but specialized solutions that are significantly more custom to their industry and more accurate. The winners will be the teams who build platforms (like Product Hub) to reduce wasted data, compress learning loops, and make evidence-based decision-making faster at scale.