Many companies aim to be data-driven. But what does that mean when context is missing? To explore this, we sat down with Deniz Minican, Director of Data Analytics at H&M and speaker at the Data Innovation Summit 2025!
With a career spanning multiple markets and expertise in building analytics capabilities, Deniz brings a people-first perspective to data. In this interview, he shares how teams work to bridge the gap between business and tech, foster a culture of curiosity, and create lasting impact through data storytelling.
Deniz highlights the challenges of aligning data with real-world decision-making, the importance of speaking a shared language across the organization, and why understanding the “why” behind your data is essential to building trust and driving value.
Read on for Deniz’s insights on why context is king, how to foster critical thinking with data, and what it takes to become a data-centric organization!
Hyperight: Deniz, you currently work as a Data Analytics Director at H&M. Can you share more about your role and the main focus areas in your work?
Deniz Minican: I belong to the central “AI, Advanced Analytics & Data” Tech Center. In this role, I lead several analytics teams who are based mainly in Sweden, India and China among others. We have two main focus areas:
First, we support our business through the AI & data product teams by setting up & following KPIs, being the translator between the technical colleagues and business stakeholders, supporting the Data Scientists with feature exploration & model evaluation, building dashboards and more.
Second (and imho more importantly), as part of the Analytics Center of Excellence we have in our area, we create ways of working and best practices, and follow these up by upskilling efforts through various analytics community efforts.
Hyperight: You’ve worked across various markets such as Spain, Russia, and Mexico at Inditex. How have those experiences influenced your approach to data analytics?
Deniz Minican: Like you say, I’ve been lucky enough to work in one of our largest competitors, learning from different parts of the business in their key markets at the time.
First of all, living in different countries and experiencing different cultures, I’ve always learned a lot about different ways of living and learned that there is always more than one way to tackle a problem.
Another thing I learned is, wherever you work in the value chain of the fashion business (or any business), you need data. It was the same when I worked in Expansion, analysing feasibility reports, or working with stores, taking decisions on operations, assortment allocation and so on.
Finally, being able to see different parts of the business and also working both with the head office and the other markets, I’ve built a more complete view of how the business works and how data can be structured & utilized accordingly.
Hyperight: What are some significant challenges you’ve faced in driving H&M’s transformation into a data-centric organization?
Deniz Minican: The main one is that we do not always speak the same language between the “business people” and the “data people.” I always say the same but our data, analyses, insights, and AI models are completely useless if the decision makers do not use them in driving the business. So, a continuous effort on data literacy for our business colleagues and vice versa is crucial for a successful data-centric H&M.
Another one is being able to meet business needs while we continue working on our data debt & future-readiness. It’s hard to find the balance for any business but we are always trying to find the most optimized balance between creating business value short-term and building & improving our solutions for a better future.
Hyperight: In your talk at the Data Innovation Summit 2025, the importance of understanding the “story” behind the data is one key point. Can you explain what this means in practice for businesses today?
Deniz Minican: I’ll try to explain this with a childhood memory I have. When I was in middle & high school, I’ve always been good at maths and science, but that wasn’t always the case with history. I’ve struggled memorizing facts, names, dates… Basically any information that I needed for my exams. And I thought that I was a hopeless case.
Only when I took a history class as an elective in my bachelors, I understood that that’s not the case. And that’s because we had this professor who turned the classes into almost like a Netflix series where you would wait impatiently for the next class to hear what’s gonna happen. By understanding the story behind the names, events, etc, I could understand the “why” of the pieces of information, which naturally connected the dots in my head and turned it all into almost like a knowledge graph.
For me, when you don’t fully understand the story, you don’t understand the “why” and how things are connected (even those you’d never think that they are), which can make you misunderstand the facts, miss the big picture, or focus on the wrong details of the story (i.e. your business KPIs).
Hyperight: From your perspective, what are some of the biggest misconceptions data professionals have when it comes to trusting and analyzing data?
Deniz Minican: One is how shortsighted we can get as data experts sometimes. Working with a focus on one business/data domain for a while, you start forgetting that all domains are somehow interconnected and you start thinking your data/numbers live in isolation.
The other is that we generally tend to think better upstream but not so much downstream. You may look at your data’s lineage and say “everything looks good” after seeing all is “green” from your source systems. But if you know your business domain well, you’d always have a look at the output and can identify if a number looks “off,” which would make you go back to exploring & fixing the issue.
Hyperight: Deniz, in your experience, what are some key strategies for fostering a culture of curiosity and critical thinking when working with data?
Deniz Minican: For me, the biggest issue generally is not speaking the same language between business and data, as I mentioned above. I spend probably the majority of my time talking to my data colleagues, explaining the business and talking to our business stakeholders across the company, explaining data and I encourage everyone I talk to to do the same.
Hyperight: What role do you see AI and machine learning playing in the future of data analytics? How does H&M plan to integrate these technologies into its strategy?
Deniz Minican: AI and machine learning will definitely change how we work with analytics today. We’re already seeing some of these changes today with coding assistants, chatbots, unstructured data analytics tools such as NotebookLM, and so on.
One part of it is how little we’ll rely on writing the actual codes & queries and building the dashboards manually as data analysts. Everything is moving fastly towards “self-service.”
In H&M, as we know that the demand is already there and the technology is getting more and more mature, we’re maximizing our efforts in two key areas to enable the future:
1. Having a robust & reliable data foundation and architecture with top of its class data management & governance capabilities. We already know that your AI solutions are as good as the quality of your data, so this is of utmost importance.2. Starting from our analysts, upskilling all our data colleagues so that they understand their business and data domain. And I put “business” in that sentence before “data” intentionally, because of the story/context drum that I’ve been banging on 🙂
If you’re curious about how to build a truly data-centric organization—one where business and data speak the same language—don’t miss Deniz’ session at the Data Innovation Summit 2025! He’ll explore why understanding the context behind the numbers is critical.
Deniz will also dive into how H&M is preparing for the future of analytics—from strengthening its data foundation to enabling self-service through AI and upskilling. Whether you’re a data leader, analyst, or business stakeholder, his session will offer guidance on making your data more trustworthy, actionable, and aligned with business needs.
Add comment