In brief

In brief

  • Today, most companies run separate projects for each use case they want to solve with data. This approach is slow, and leads to duplicated work.
  • Data products offer far greater ROI and lower cost-per-use than data projects, because they evolve to support multiple use cases over time.
  • Data products can help organizations create new revenue streams, improve decision-making and boost efficiency.
  • To create and manage data products effectively, companies need the right skills, the right technology, and the right strategy.


Data products productize data to accelerate value, improve decision-making, and even create entirely new revenue streams

Data is a powerful new form of capital. It’s essential for organizations to survive and thrive in today’s fast-paced business environment. And it’s generated everywhere from humans, machines and Internet of Things (IoT) devices to edge systems and beyond.

Now, there’s a new opportunity to turn all that data into a competitive differentiator. How? By creating data products.

Simply put, they’re products that facilitate end-goals through the use of data.1 They contain data packaged with everything someone needs to understand that data and use it to solve a new use case—even if that person works in a different team or outside the business altogether. Just like consumer products, data products are designed for specific purposes. Let’s take a closer look at what data products are, and how they differ from traditional approaches to using data.

Project vs product approach to data

Today, most companies approach enterprise data with a project mindset. Each time a business function has a problem that it wants to solve with data, the organization starts a project to acquire the data, cleanse and prepare it, then analyze it for that specific use case. And each time a new business problem or use case arises, it follows a similar process to acquire, prepare and analyze data for its specific need.

This project mindset may sound familiar. It’s how most data teams currently operate. But this approach has some significant drawbacks:

  • It may lead to duplicated work.
  • It’s relatively slow.
  • The outputs from each project typically can’t be repurposed to solve other use cases.

Data products give companies a better way to address their data needs. They’re designed with the entire organizations’ data needs in mind, and they can be reused to support numerous use cases across multiple functions. Data products can be:

  • Datasets—reusable datasets (eg for design, manufacturing, finance and operations), data streams, data feeds, or APIs that meet the needs of the whole enterprise, as well as each business function.
  • Code—feature code and transformation snippets (small blocks of reusable code) or data models (which show the logical structure of how different data elements fit together).
  • Analytics models—reusable machine learning models (eg for predictive maintenance).
  • Dashboard reports—reusable dashboards (eg risk dashboards) and other visualizations.
Data products typically offer far greater ROI and lower cost-per-use than data projects, because they evolve to support multiple outcomes over time.

Data products typically offer far greater ROI and lower cost-per-use than data projects. Why? Because although the upfront costs may be higher, they evolve to support multiple outcomes over time and accommodate new use cases that emerge. The product mindset keeps the focus on realizing the business use cases. What’s more, data products offer an innovative way to decouple data from specific applications and use cases to maximize its value. And they help break down data silos across the enterprise.

Data products also offer several key advantages for the people who use them. These include:

Speed

Time-to-insight is much quicker, because the data product is pre-built…People don’t have to start a new project each time they want some data.

Trust

Users know that data products have gone through rigorous quality control.

Real-time information

Unlike static datasets, data products provide real-time data for decision-making.

Accessibility

The relevant data is already available, so people don’t have to go and ask another team for it.

Usability

A well-defined data product with a well-defined interface is much easier to consume than a raw dataset.

View All

Data products in action

You may not realize it, but you’re probably already familiar with some data products. In fact, many digital natives—such as Google, Uber and Netflix—have built their entire businesses around them. They compete on their ability to drive actionable insights from their data.

But data products aren’t just for digital natives. They offer huge potential for established companies, and they can even open up entirely new revenue streams. For example, a medical device manufacturer can start supplying medical-grade data services to healthcare providers to help drive better patient care. An oil and gas producer can achieve greatly enhanced efficiency in its plants. Or a media and entertainment company can serve personalized content to its customers.



How to make data products a success

To successfully implement data products, organizations must support the entire lifecycle—from conceiving and designing the product through to building it, rolling it out, supporting it, then retiring it when it’s no longer needed. This responsibility should sit with a new organizational function devoted to data product management. It’ll need people with a broad range of skills around data, business analysis, DevOps and more. Plus, the teams creating data products need knowledge of the relevant industry and domain.

Organizations should develop data products in a data platform that’s built in the cloud. Why? Because cloud enables scale, agility, and the opportunity to drive reinvention. It allows for data to be connected as a part of a larger continuum. And by tapping into the Cloud Continuum, organizations can productize their data—wherever it resides.

Investing in data products can really pay off. After all, data products empower organizations of all kinds to leverage data to achieve critical business outcomes. It’s too big an opportunity to miss.

1 Definition from Data Jujitsu: The Art of Turning Data into Product, 2012, by Dhanurjay “DJ” Patil, former Chief Data Scientist at the United States Office of Science and Technology Policy

Teresa Tung

Cloud First Chief Technologist

MORE ON THIS TOPIC

Why data & analytics needs a multi-speed approach
AI for data: Data capital management @ scale with AI

Subscription Center
Stay in the know with our newsletters Stay in the know with our newsletters