Gestion des produits de données - De la surcharge de données aux données activées en trois étapes

Version française prochainement disponible

Today's enterprises teeter on the edge of data overload.

Historically, we’ve had a one-track mind when it comes to data: gather as much of it as possible, from as many sources as possible. This relentless stockpiling, however, has turned data from a valuable resource into an embarrassment of riches.

Efforts to manage this unwieldy body of data have resulted in multiple sources of truth, eroding confidence in the data itself. As a result, data quality, governance, and trust have all suffered—affecting decision making at various levels. Recent Gartner® research, in fact, revealed that “44% of D&A functions are effective at providing value to the organization."¹ This data alone underscores a significant opportunity for improvement.

For many of our customers, building a new data strategy involves more than just finding new partners or solutions. Instead, it requires a reevaluation of how data is managed on an organizational level, by...

• Auditing your team’s data pain points and goals
• Developing a product management approach to data
• Entering the data marketplace to generate new revenue streams

Auditing your team’s data pain points and goals

To effectively audit your team’s data ecosystem, collaboration is key. Stakeholders from various lines of business must unite to clearly define their data pain points and needs. This collaboration sets the foundation for addressing your enterprise’s data challenges, as well as for setting short- and long-term targets for your data-driven outcomes and insights.

Take, for example, one of our customers in the financial services industry. Like many modern enterprises, this customer had, over the years, become the caretaker of a sprawling data repository, riddled with duplicates, redundancies, and access barriers.

They knew it was time for a new data strategy but hadn’t yet figured out what that looked like for them and their operation. One thing was clear though: their current data management methods were undermining their larger business objectives.

For instance, many of their teams felt there was a lack of a single source of truth for customer data—which was not only preventing them from optimizing customer experiences, but also dragging down revenue realization and productivity.

Historically, the company’s data had been stored across various legacy systems, including mainframes and multiple legacy data warehouses. This complexity made it increasingly difficult to catalog data sources efficiently. So, when it came to compiling a comprehensive view of the company’s customers, the teams were left with no choice but to mine multiple data sources, increasing the chance of errors and oversights.

Developing a product management approach to data

Addressing data management challenges such as this one might begin with assembling a cross-functional team of SMEs, data scientists, and data analysts, (in this instance, primarily from the marketing and sales departments). This team would then put their heads together to ideate and design data solutions tailored to their organization’s needs.

And this is where that product management approach comes in.

When teams start to treat data as a product, they naturally establish a product management workflow, which boosts transparency and maximizes value. The goal of a data product approach is to create data tools or solutions that can be used and reused across teams for a variety of use cases. In other words: a single source of truth.

For example, an organization—such as the financial service provider referenced above—that is struggling with customer data might start by leveraging existing data to create a unified, 360-degree view of customer data. In the past, this organization might have typically discarded data after a single use, limiting its utility and overall efficiency. In contrast, a data product like the 360-degree customer dashboard would provide a reliable, unified source of customer data to call on again and again.

However, building out data products such as this 360-degree customer dashboard is just one part of the journey. It is equally important to cultivate a product management mindset across the entire data portfolio.

Adopting a product management approach to company data, for instance, often requires assigning a dedicated product manager to each data product. It means integrating automated data product pipelines with agility and intelligence. And it calls for governance of each product through a federated system for observability.

This strategic approach ensures that data continues to be a valuable business asset, consistently driving value both in the present and the future.

Entering the data marketplace

Embracing a product management mindset for your data is a powerful tool for enhancing your data’s accessibility and usability. But the implications of a data-as-a-product strategy stretch still further, beyond your enterprise’s borders, paving the way for the distribution and monetization of that same data.

By packaging your data into products, you can deliver that data—clean, comprehensive, and governed—to data marketplaces. There, suppliers as well as other downstream B2B partners can source that data to, for instance, run AI models more comprehensively or create more personalized user experiences.

As sectors such as financial services begin to expand into new paradigms, such as open banking, adjusting your data strategy to seize these opportunities is not a matter of if, but when and what resources it will require.

Now is the time to start answering those questions.

Naveen Kamat is Vice President & CTO Data and AI Services at Kyndryl

¹ Gartner, CDAO Agenda 2023: Presence, Persistence and Performance, March 2023

GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.