Johannes Hovi: You’ve had a long and successful career in data management space. What have been the biggest changes or advancements in the industry recently? Big data and Hadoop was a huge thing some time ago and there is certainly a big buzz around AI right now, isn’t it, or how do you think about this?
Donna Burbank: Artificial Intelligence (AI) and Machine Learning (ML) are clearly part of the current “buzz” and value proposition for many organizations today. The ironic result of this focus on new technology, however, has been a return to fundamentals such as Data Governance and Metadata Management. The largest demand outside of Data Strategy in our organization has been Data Governance, as more and more organizations realize that in order to take advantage of new advances in AI, ML and predictive analytics, a strong, high-quality data foundation needs to be in place.
Although data governance has been around for many years, I am seeing a change in how this being rolled out, which is typically led by the business with a higher focus on collaboration and innovation in order to achieve business goals, rather than starting with rules and policies. While rules and policies are clearly still important, generally a focus on the “carrot” rather than the “stick” is more effective in the long-run.
JH: Your class is about Data Strategy in practice. What are the key elements of an efficient data strategy?
DB: At Global Data Strategy, Ltd., we use the following framework to guide the key components of a data strategy.
At Level 1, the focus is on aligning Business Strategy with Data Strategy. Does the organization want to be the next Netflix of the movie rental business, or are they looking for a more moderate approach, looking to improve their current business efficiency through better reporting and analytics? This is key to get right in order not to over or under-develop the associated data strategy and architecture.
Level 2 focuses on managing the people, process, policies, and culture around data, in other words: Data Governance. The data governance maturity of an organization at this level – or the lack of it – can determine the options you have for using your data strategically, as well as the timeline for putting it into practice.
The third level encompasses the various Data Management practices that help leverage data for strategic advantage, such as Data Quality, Master Data Management, Data Warehousing, and others.
The fourth level focuses on coordinating and integrating disparate data sources. For some organizations, the first step is to create an inventory of the data available, which is often unknown. Once these sources are known, it is critical to integrate them effectively with the right level of business and technical metadata for them to be successfully leveraged for business use.
Level 5 focuses on the various data sources and platforms available in order to align technology with the business use cases and drivers highlighted in Level 1. With the exponential growth in technological advancement, there are a number of options available from real-time streaming to big data storage, as well as the tried and true relationship database.
While all of these levels need to be addressed in some capacity, they do not need to be implemented in a particular order, or with the same level of focus. Generally the Business Strategy at Level 1 is the first place to start, which dictates the level of focus needed in the other areas.
JH: In digital companies such as Google or Amazon there is no data strategy, data and business are pretty much the same thing. Should companies have separate data strategy at all or is it sort of a subcategory of a digital strategy?
DB: Many organizations are pursuing a digital strategy and data is at the foundation of any digital strategy. Without core data management fundamentals, a digital strategy is unlikely to succeed. That said, many other organizations also focus on data strategy without a specific link to digital transformation. Many organizations are using data in a number of ways to improve efficiency, drive revenue, etc. through better analytics or operational data utilization. It depends on the industry and use case whether an organization sees themselves as a “data company” where the business and data are inextricably linked, or a company that uses data to support their core business model in a more effective way.
JH: The most downloaded mobile games come from Finland such as Angry Birds and Class of Clans. These companies are obviously very good at utilizing data in the business development. However, most large companies are not digital gaming companies. How would you recommend “traditional” industrial or retail company to get started with data strategy?
DB: We have some core tools in our toolkit that help outline core business drivers at a high-level and then map them to the data assets of the organization. These tools answer a series of questions that help pinpoint the areas most suited to expansion around data. For example:
- What are the core business drivers of the organization? (e.g. product expansion)
- What are the market forces outside of the organization that might affect our strategy? (e.g. online shopping, social communities)
- What are the core data assets of the organization and how can they be linked together (e.g. customer usage patterns with product development).
- What new business models or business enhancements can arise by combining market and organizational needs with current and future data assets?
Combining the answers to these questions into an actionable data strategy is a bit of art and a bit of science, as the saying goes. We’ll provide some of the tricks of the trade in the course to hopefully make the journey a bit easier.
JH: How data strategy should be implemented in practice? Some companies do have Chief Data Officers, but not all. Is it an IT department’s job to come up with strategy road map?
DB: Data strategy should have core support from the business stakeholders in order to be successful. One way we often summarize this is that data strategy should be “Business-driven and IT-supported.” New roles such as Chief Data Officer are typically a mix of business vision with data management technical skills, with an emphasis on the former. i.e. business vision is increasingly more important than core technical skills. That said, technical staff who are able to think strategically and explain complex technical data topics have a unique role to have a “seat at the table” in a company’s vision and roadmap. I work with many C-level executives who are eager to work with technical staff who can explain difficult concepts in an intuitive way and align technology roadmaps with business vision. It’s a great time to be in data management if you’re the type of person who is interested not only in technology but also in business innovation. That’s a key reason I’m still working in the industry myself.
JH: It is absolutely amazing to have you here to run a class. We have your books in the office about data modeling and your life work has inspired us a lot. Which would be the key takeaways of the training?
The goal of this training is to provide clear, practical guidance in building an actionable data strategy. The concept of a data strategy can seem overwhelming and this class will provide some clear tools that attendees can use when they get back in the office right away to begin to show value.
Donna Burbank will come to Helsinki Finland to run a class on Introduction to Data Strategy – A Practical Approach on 25.02.2020 – 26.02.2020.
More information and tickets here.