Techopedia has a clean succinct definition[i]:
Data-driven is an adjective used to refer to a process or activity that is spurred on by data, as opposed to being driven by mere intuition or personal experience. In other words, the decision is made with hard empirical evidence and not speculation or gut feel.
Many if not most companies have said – We can do that – and started down the road towards the data-driven nirvana. The basic premise has long been deploying a data warehouse, populating it with your company data, hiring a cadre of business intelligence analysts and in no time, we’ll reach nirvana as it’s just over the horizon.
The journey is rarely as easy as originally planned and scoped. This is the first in a series of blog posts discussing common obstacles, and why they are often much tougher to overcome than the initial promise. We’ll start by focusing on culture. Changing an organization’s culture is never easy.
What happens in the real world?
First, hiring one or a cadre of analysts rarely delivers the capacity to inform every critical business decision. Often the result is that much of the organization’s staff continues to operate as they always have. Basing decisions on what worked last year, or last month, or just feels right. Many will do their own simple or very complex analysis using an Excel model populated with out-of-date and often dirty data which delivers invalid analytics. Even when the hard analytics are very good, the analysts often lack the soft skills to craft and tell a compelling story.
Second, executive teams, particularly when informed with a weak story regardless of the quality of the analysis, will often ignore the analysts’ recommendations and do what they’ve always done. Repeat past decisions or simply follow a hunch – make the decision that feels right. That approach built the company we are! Reality often comes faster than you can respond. Whether it’s lower prices from a competitor who has figured out how to undercut you, disruptive innovation from an upstart, or pressure from stockholders executive teams are under continual pressure to improve execution.
Third, the more complex an organization the greater the likelihood that data is siloed, and analysis is therefore siloed too. Correct answers often require sharing data across the organization and it must be current, correct, and easy to join across silos. A strong data-driven culture, therefore, requires strong data governance and metadata management skills embedded across the organization.
Fourth, emerging software-as-a-service (SaaS) analytics solutions make it easy for individuals and small teams to choose their own tools (shadow IT) which empowers them in their eyes to be data-driven. What often happens, however, is these individuals and teams lack the skills to ensure security and protect the privacy of data which creates reputational[ii] and potential financial risks[iii] for the entire organization.
Fifth, many organizations have successfully built classical business intelligence processes which look backward at performance often delayed by days and sometimes months. Competing in today’s markets requires near real-time views of performance and the ability to predict likely outcomes. Tomorrow’s forecast predicts rain. What actions will deliver superior outcomes for a coffee shop on a rainy Wednesday?
What does a data-driven culture look like?
It starts with your people. Data skills across the organization, with a strategic human resource investment that ensures you have a:
Data-literate team where everyone who works with data (which is often everyone in your company) understands the basics of data, analytics, and storytelling. This is your foundation.
Data-savvy executives who expect all critical business decisions to be informed with data; who strive daily to instill and foster a data-driven company culture.
Data-experts placed strategically across the organization. Whose mission is to manage data as a core business asset.
A data-driven organization requires a culture of data ethics. Simply put -- your Key Performance Indicators (KPIs) across your organization must respect your ethics. Just because the data says doing X will be more profitable than Y, if doing X could harm your reputation then it’s probably not the right course of action no matter how cool/innovative X may be. Further, a strong foundation of data ethics makes it easier to comply with a changing regulatory landscape such as emerging privacy regulations in Europe.
The data-driven culture is aware of the organization’s data-driven maturity level and striving to constantly improve it. It is aware that being data-driven is a journey, not an end-point because the goal posts are always moving. It is open to change but understands that change requires rigor. That chasing trendy shiny objects without process and requisite rigor could upend the organization.
In my next post, I’ll expand on the concept of data literacy for everyone.