Just the other day, I was deep into a workshop on agile practices. While tackling an exercise, I steered the discussion towards data and its role in driving business success. This discussion took me down memory lane to an event I had attended earlier this year. The event was put together by the passionate data enthusiast, Felipe Flores. The event, known as Data Futurology, left a deep imprint on my mind.

I can still clearly recall the feeling I had that day- like a country girl stepping into the bustling metropolis for the very first time. The atmosphere at Data Futurology was charged with inspiration, and I felt my mind expanding as each session unfolded.

Among the numerous intriguing ideas presented that day, one concept that left a lasting impact on me was viewing data as a product. It may sound straightforward, but its implications were deeply profound. It was one of those enlightening moments when a single idea significantly shifts your perspective. It may seem like I’m exaggerating, but that’s genuinely how I felt.

You might be wondering, “How can something intangible be likened to a physical product?”. It took me a moment to wrap my head around it. As the presenter proceeded, the light came on, and it dawned on me just how ingenious it is to perceive data that way.

So, in this blog, I’m going to share this innovative concept of looking at data as a product and how it contrasts with the conventional approach to managing data. This concept was popularized since the publication of the data mesh introductory article by Zhamak Dehghani in 2019. And to give this notion of “data as a product” a more tangible feel, I’ll draw parallels with a physical product.

Traditional vs Data as a Product Approach

Traditionally organizations see data as an incidental byproduct of operations. They collect, store, and sometimes use it, but rarely manage it actively or even see it as a valuable component of the organization’s success. Data typically exist in silos, hidden away in different departments, and the full potential of the data remains unrealized.

Contrast this with the 'data as a product' concept, where data is seen as a primary output. It's thoughtfully crafted, packaged, and presented, just like a physical product in a supermarket. Data is standardized, updated, and 'consumed' by the users. This proactive approach is geared towards understanding and catering to the needs of data consumers, and it fosters a data-centric culture within the organization.

The path from viewing data as a byproduct to a product is not a walk in the park. It calls for significant effort and time. But when we consider companies like Uber, Airbnb, Google, and in my industry, IQVIA, we can see how fruitful this investment can be.

But let’s keep it real, realizing this is no easy feat. It demands commitment, patience, and a shift in mindset from everyone involved. It necessitates open communication, transparency, and shared responsibility. But once these hurdles are crossed, the outcomes can be rewarding.

Organizations that treat their data as a product can reap numerous benefits. They're likely to have higher data quality, better data accessibility, and more effective decision-making processes.

Drawing Parallels with a Physical Product

To help understand this concept better, let’s draw parallels with a physical product: a carton of milk. It comes with clear labelling (the equivalent of metadata in the data world), which tells you what it is, where it comes from, and its nutritional information. It has an expiry date, signalling when it should be consumed for optimal freshness.

Now, imagine if our data were handled in a similar fashion. Metadata would provide context about the data: its origin, purpose, and other relevant details. We would ensure data freshness, getting rid of outdated or irrelevant data and regularly updating our datasets. We'd package the data properly, structuring and organizing it in a way that makes it easy for the data consumer to use. By doing so, we can dramatically increase the quality and utility of our data.

Yet, bringing this concept to life is not an overnight task. It involves rethinking how we collect, store, and use data. It requires establishing new procedures and practices, investing in the right tools, and training our team to adapt to the new approach. Most importantly, it demands a mindset shift. Everyone in the organization needs to see the value in treating data as a product, and this shift often starts from the top.

A key aspect of the 'data as a product' approach is having a dedicated product manager. In the same way, a physical product has a product manager who oversees its lifecycle and works to optimize its value, data too could benefit from such a role. A data product manager would be responsible for understanding the needs of the data consumers, ensuring data quality, and constantly seeking ways to improve the data product.

As a part of the implementation, a company also needs to develop a 'data catalogue.' This would be akin to a product catalogue, providing detailed information about the various data assets available within the organization. It will help users find the data they need, understand its context, and use it effectively.

This transformation, however, is not a walk in the park. It can be challenging, and it requires a considerable investment of time, effort, and resources. But as the saying goes, "the best things never come easy." The value that can be unlocked by viewing and treating data as a product can significantly outweigh the challenges.

There you have it—data as a product. Next time you stroll into the supermarket- I hope it’ll remind you how we should think about data. Notice everything around you. From the meticulously arranged juice and milk cartons to your much-loved chocolate bar, that's always in just the right spot. Everything is carefully managed and presented for your convenience. To realize data to its full potential, it should be thought of and managed like a product.