Data is one of those rare birds that crosses the business/IT divide that exists in many organizations: while business managers depend on the insight data can provide to refine product and market tactics, IT can optimize operations based on data generated through monitoring of IT infrastructure and services. For both groups, however, data also represents a management challenge: if timely access to the right information must be secured to align information with business goals, the IT manager must support this access through the creation of systems that also deliver effective security, privacy, governance and management of the data tsunami that is part of our increasingly digitized world. But despite this cross over – or perhaps because each groups assigns responsibility to the other – the development of best practice in data management has been elusive in many organizations. Companies do not operate without a business plan, but many do try to function without a well formed data strategy, a juggling act that is becoming increasingly difficult as data grows in volume, velocity and variety – and value.
Lack of formalized process around data is having the expected consequence. Back in 2012, a Canadian market study found that the vast majority of small to mid-sized businesses had very little understanding and no plans for development of Big Data programs. Since then, awareness of the strategic importance of data has grown considerably. Interestingly, though, the share of organizations that are taking advantage of data to build value has not kept pace: according to 2015 research from Techaisle, only 26 percent of US SMBs (1-1,000 employees) report they will be engaged in Big Data initiatives in the current year, and Big Data ranks sixth in a list of the top ten priorities for this group.[1]
What accounts for this adoption lag, and what is to be done? One answer is education, a mandate that New York-based analytics firm Information Builders (IB) has taken on as an effective way to help customers derive value from their data, and ultimately drive adoption of advanced data solutions. For example, a recently released IB white paper, Driving Better Business Performance With a Practical Data Strategy, tackles this confusion around concrete data planning with a practical guide that can help businesses build data strategy that is unique to their individual organizational circumstances.
The white paper begins with the premise that data strategy is much like business strategy: it involves people, processes and technology, but is unique in that it aligns these with the organization, governance and sharing of data. This “three-by-three” matrix forms the structural core of the white paper, and by implication a good way of thinking about data strategy, defined as “a set of policies designed to help achieve a vision” within the organization.
Invoking the example of Napoleon Bonaparte, report authors note the importance of flexibility (reacting opportunistically to new circumstances that arise) and also single minded focus on a set of policies to the achievement of effective strategy. They also point to Bonaparte’s superior use of people (sufficient staff to communicate with foot soldiers), processes (execution of rapid troop movement and disruption of enemy communications) and technology (horse drawn artillery and less costly small arms). While Napoleon’s conquest of Europe is different in scope from the data challenges in many organizations, the authors argue that simplicity and ease of communications are two qualities developed by the little corporal that can be applied to many data environments.
Turing to data strategy, the paper outlines three key questions that may be used to align general and data specific imperatives in the matrix: How do we structure data to ensure that it meets company needs, how do we manage data to ensure its suitability for its purpose, and how do we provide data in ways that systems or people can apply it to business problems?
But the white paper does not rest on conceptual models, rather it discusses concrete steps that organizations can take to ensure successful strategy. In the ‘people’ realm, for example, IB recommends data strategy that brings together people at higher and lower levels of the company and across departments – IT and business in particular. People policy should have inherent balance, avoid favouring one group over another, and effectively communicate to stakeholders their and the organization’s interest in following the defined strategy. And “as with business strategy,” organizations should strive to “keep your communication simple and direct – more like marketing than technical documentation.”
On the ‘process’ front, the paper illustrates the perils of “redundancy of data-focused modules across different business processes,” where, for example, collecting, validating or analyzing customer information might be done on the company’s website and at the contact center. While the same function is being performed in both processes, the website and contact centre are likely to do so according to different rules, with conflation of information the outcome. A second issue arises when the same data is collected, validated or analysed in different ways through different processes – for example, all at once by an order entry system or over time by a marketing system. In both cases, good alignment across processes requires identification and rationalization of data collection, validation and analysis activities. The goal of “refactoring” is creation of a repository that allows changes made to the data in one place to affect use of the data across all processes. Better standardization in the organization’s approach to data governance and usage is the result, which engenders greater trust in the data, leading to more sharing, and ultimately to increased data value.
According to IB, ‘technology’ is in some ways “the least important part of creating a data strategy; it tends to fall out of the people and process sections of the strategy.” Happily, it is also an area that features relatively well-developed best-practice knowledge, such as how to gather technical capability and skill requirements to evaluate gaps and strengths in an organization’s hardware, software and talent resources. To effect this in a strategic way, IB recommends gap analysis across the organization – as opposed to analysis on a project-by-project basis – through outbound communication, but more importantly, through broad collaboration between all groups aimed at shaping future technology usage and buying patterns in order to encourage leverage of data management technologies across all business units.
And as final guiding principles for creating effective data strategy, IB asks readers to consider the following:
- Look at many successful projects to see what’s consistent across them; likewise with unsuccessful projects.
- When considering elements of your data strategy, evaluate them based on your particular culture, not in the abstract.
- Focus on improving things you can control, and containing those you can’t.
- Continually revisit your strategic principles in the light of changing circumstances.
Readers interested in understanding more about specific considerations around the organization, governance and sharing of data as they relate to people, processes and technology – and more about how the Ford Motor Company, RainTree Oncology Services and the Journal of Commerce broke new ground with effective data strategy can access the white paper here.
[1] Techaisle Research. The 360 on SMB Big Data: SMB & Midmarket Big Data Adoption Trends. 2015.
