McKinsey Research Calls out the Business Value of Data Management
I recently read an interesting McKinsey piece, a gold mine of information about how companies are approaching data and analytics. The article, “The Need to Lead in Data and Analytics,” unveils the results of a new executive survey and covers much more than I could discuss here. Two bits of research really jumped out at me:
- 86% of those surveyed said they’re “at best only somewhat effective at meeting the primary objective of their data and analytics program.”
- The #2 success factor in effective data and analytics, ranked only behind senior-level exec support, was “Designing effective data architecture and technology infrastructure to support analytics.”
This tells me that, despite the proliferation of tools, platforms, and smart thinking on analytics, a lot of companies are still struggling to succeed with their data. And McKinsey’s research suggests that the biggest technical success factor is how you design and manage your data infrastructure.
Leading orgs master data management
McKinsey’s article includes a graphic (Exhibit 5) that breaks down the self-reported capabilities of advanced organizations, and the top three factors that separate leaders from laggards involve data management:
- Data is accessible across the organization, said 64% of high performers (vs. 33% of low-performing organizations).
- Tools and expertise enable the org to work with unstructured and real time data: 59% of leaders (vs. 12% of laggards).
- The org enables self-serve analytics capability for business users: 52% (vs. 23%).
McKinsey notes that exec sponsorship and talent development are also vital. Executive sponsorship is essential to drive a transformative business initiative like this. The talent development and career paths are essential to hiring and retaining top analytics and data talent – a major challenge in many organizations.
How data management drives results
Adding some of my own recommendations to McKinsey’s, here are five important steps around data management for organizations looking to drive business results through analytics:
- Secure the C-suite. The highest performing data and analytics organizations always have C-level executive sponsors. If you are lucky, you have that support. If not, you have to make the case to win them over. Win that approval with ROI analyses, a well-researched business case, and through pilot projects that start small, but show real value, on key initiatives, from the start.
- Know where you’re going. Yes, short-term needs or goals can dictate efforts around data and analytics, but make them serve a complete vision for where you’re going. That way, each short-term project is another step closer to your desired state. And a good data platform (Informatica, for one) will let you interoperate with existing analytics and applications for as long as you need them, rather than forcing a costly “rip and replace” approach.
- Have standards. You should standardize on your data management platform, and the challenge to your vendors should be two-fold. First, a platform should make you more productive by standardizing practices across all your data and use cases, letting you reuse skills and assets. Secondly, you need a platform that’s adaptive. The types and sources of data are always in flux, to say nothing of the tools and technologies you’ll apply to analytics.
- Make data management a career track. Data management, as a separate discipline from data science, is a crucial career track. True data scientists are in high demand, cost a lot, and are frankly too rare and costly to use for data prep work. Develop a data management track separate from analytics—one that develops specific competence in data management.
- Design for business-IT collaboration. While IT owns the data infrastructure, only the business knows the full context of the data, and which data is important to actively manage and which is not for a given analytics initiative. Collaborative environments will be the most productive in delivering business value.
Success with a holistic approach
Businesses have spent decades focused on process reengineering. We worked to improve marketing functions, and sales functions, and data was considered only in how it served whatever function was under scrutiny. Now we need to take a step back to set up and manage a comprehensive data infrastructure that manages data across the enterprise.
I think the five considerations I list above can help your organization achieve that holistic vision. For more, check out “Laying the Foundations for Next-Generation Analytics”, a detailed workbook we’ve created that walks you through the process of establishing your overall data management vision, and applying it as you assess each project in terms of stakeholders, goals, data sets, and tools.