Part of the “Enterprise Data Strategy” series, which explores the importance of leadership and accountability in directing an overall data strategy tied to business outcomes.
Years ago, when I went on my first date, I wanted my car to be clean and sharp-looking. I took care to clean it, inside and out. It looked nice and shiny for the date. The date turned out great, except for one small detail: The car ran out of gas at the end of the evening, about 10 minutes short of reaching my date’s home destination. I was totally embarrassed, and although my date was cool about it, her parents doubted my sincerity when I said it was an accidental oversight. In the end, the problem was resolved, and it all ended fine. I dated that girl many times afterward.
The foundation for mission success
That small story illustrates to me the difference between successful tactics and successful strategy, especially on the battlefield. Strategy is directed at winning the war, completing the mission, and achieving ultimate success. Tactics are focused on individual steps that move us toward completing the full mission. Tactics sometimes require you to lose the hill in order to win the battle or lose the battle in order to win the war.
For my first date, my tactics were mostly aligned (but not 100%) with the ultimate strategy – to complete the mission of having a successful date. I should have focused on the full strategy and not so much on the individual shiny components. Perhaps then I would have paid closer attention to my mission-critical data (the sensor reading on my car’s fuel gauge).
Enterprise data strategy can be like that. Teams of digital experts, business analysts, analytics practitioners, and data scientists can experiment with and build fancy algorithms, models, and tools that work well with the clean data sets that are specially prepared and curated for their experiments. The experimental models, validation tests, and demos can all look nice and shiny. But the real completion of the organization’s digital mission requires more than those tactical successes.
The reality that organizations face is that the fuel (i.e., the data) for those experiments may be sufficient for those limited tactical goals, but the data collections are insufficient to fuel sustained, successful, and strategic enterprise-level deployments. The truth is this: Most big data are collected with no strategy in mind, and the data sets that you critically need are often imperfect, dirty, mislabeled, siloed, or just plain missing!
A comprehensive, domain-specific road map
An enterprise data strategy (EDS) can provide the foundation for strategic mission success and market advantage. The EDS has been defined as the comprehensive vision, actionable foundation, and domain-specific roadmap for an organization to harness the full potential of its data-dependent and data-related capabilities.
There are several things that an EDS is not. It is not a wish list of things hoped for. It is not a laundry list of technology trends to try out. It is not a generic set of principles and platitudes (such as “Data is an Enterprise Asset”). And it is not a fuzzy, aloof vision.
Any strategy, including an EDS, must be specific, relevant, actionable, evolutionary, and mission-goal directed. If your EDS does not align with your organization’s North Star (your mission), then one or both of those need a mid-course correction to get back into alignment. Your mission should already be aligned with your market strategy: knowing your current place in the market, current market trends, your differentiator in the market, how you will achieve market advantage in that environment, and what components you must assemble to get there. Consequently, your EDS must be connected with and integrated into that strategy discussion.
Data strategy includes the data management principles of governance, security, privacy, access, architectures, and related design considerations. Those are important, but I want to focus now on the analytics and data science considerations.
Data science and analytics
In the era of massive data collections, your data can be your most critical and valuable strategic asset. Data science and analytics drive insights discovery, innovation, and value creation. Data science and analytics also induce strong employee retention, by empowering your people to explore, discover, and create value. They also generate analytics products that can be operationalized and monetized. These include data products, customer-facing applications, APIs, models, recommender engines, specialized data sets, curated collections for specific lines of business, open source tools, shareable data science notebooks, cloud services, data portals, reusable analytics workflows, and more.
Tactical components: the power steps
In order to operationalize, monetize, and commoditize those data analytics products, the EDS needs to focus on clean data; quality data; labeled data; curated data (data catalogs); shareable data; machine-accessible data; well-documented data (data dictionaries, employing metadata standards, including taxonomies and perhaps also ontologies); and data inventories (to determine what’s missing among the lists of essential data sets). These tactical components are power steps toward mission fulfillment, strategic success, and market advantage.
Ultimately, none of these components live in isolation. It is the integration and combination of data, analytics, business logic, and mission that explicitly induces the organization’s message, culture, and strategy to align with specific, actionable, and relevant business outcomes. Data are simply the input, albeit a significantly large amount of input. Similarly, data are not necessarily a key differentiator, since all organizations now hold (or have access to) massive repositories of data.
Trusted data for trusted outcomes
On the other hand, your mission-focused analytics outputs are your uniquely valuable output, your differentiator in an increasingly crowded digital marketplace. Your analytics must be fueled by clean, accurate, accessible, curated, labeled data – prepared for the complete mission and not falling “10 minutes short” of reaching the end goal of a fully successful digital transformation. Trusted information and data enable trusted innovation and outcomes.
I was fortunate to attend the SAPPHIRE NOW conference in Orlando this year, where I learned about the many ways that SAP enterprise software and services can support an end-to-end data strategy. What I most loved about the event was the focus on the “Experience Economy.” For me, the Experience Economy is the ultimate application and value proposition of our collected big data assets: the creation of better experiences for our customers, employees, and other stakeholders, through discovered insights and learned actions that are encoded in our data collections. A Customer Experience Management (CXM) platform that focuses on customer data can achieve its full potential when it is designed, developed, and deployed in alignment with an organization’s greater enterprise data strategy.
Success in the Experience Economy
In the Experience Economy, the CXM platform and the EDS work together as the sharp-looking vehicle that will drive your organization to the home destination of digital transformation, mission success, and market advantage. That’s not just a first date with your data; it’s a long-term strategic commitment to harnessing the full potential of your organization’s data-dependent and data-related capabilities.
For more information
- Read all five of the blog posts in the Enterprise Data Strategy series
- Check out the Data Strategy Guidebook: What Every Executive Needs to Know for more details
- Gain a better understanding of your current organizational maturity with SAP’s executive-focused, next-generation Database and Data Management Assessment, which is based on top KPIs and best practices
- Learn how the office of IT and data leadership can leverage information as a strategic advantage with Best Practices for Aligning Data Capabilities and Business Needs
- Find out more about the importance of data management in powering AI in the post Artificial Intelligence Without Data Intelligence is Artificial