Page 81 - The Real Work Of Data Science Turning Data Into Information, Better Decisions, And Stronger Organizations by Ron S. Kenett, Thomas C. Redman (z-lib.org)_Neat

Educating Senior Leaders 71

Organizations Are “Unfit for Data”

Today’s organizations are “unfit for data” (Redman 2013d). As used here, “organization” consists
of four components, along with an example or two of companies’ lack of fitness for data:

• People. Companies lack enough people with the needed skill and expertise up and down
the organization chart.
• Structure. Silos are the enemy of data sharing, impeding the cross‐functional coordination
and work.
• Policy and control. Roles and responsibilities for data are misaligned.
• Culture. Even though they say they do, people and organizations do not value data and data
science.

There is a great deal here, so we will only pursue two points. First, we find that many people
confuse data and technology. In the past, this led companies to assign responsibility for data
to their information technology departments, to the detriment of both. But data and technology
are very different kinds of assets and require different styles of management.
Importantly, technology is increasingly a commodity. As Nicholas Carr (2003) pointed out
more than 15 years ago, basic storage, processing, and communications technologies are
readily available to all, at a fraction of the cost of just a few years ago. If anything, the trends
Carr called out are accelerating – witness the stunning progress of cloud computing, the
penetration of mobile, and easy access to advanced analytic and AI techniques.
These points lead us to conclude that the first step is to separate management responsibilities
for data and technology.
The second point is finding the right spots for data scientists, and too many companies,
perhaps unwittingly, set their data scientists up to fail (Redman 2018a). We take this topic up
in the next chapter.

Get Started with Data Quality
We took up data quality in the context of data science in Chapter 6. Across a company, most
data is in poor shape (Nagle et al. 2017), and the associated costs are enormous (think 20%
of revenue; Redman 2017c). Worse, people rightly do not trust the data (Harvard Business
Review 2013), and you certainly cannot expect them to make a data‐driven decision if
they do not trust the data. We find that most data issues have rather simple roots and can
be eliminated with relative ease. Thus, data quality is a great place to start a data program.
And the savings will provide the funds for everything contemplated here.

Implications

Educating senior management and helping guide overall data strategy is a tall order indeed.
The space is a confused mess and the topic is very charged and political. There are always
good reasons to delay, or simply avoid, the tough issues. But if a company or agency wishes
to enjoy more of the benefits data and data science offer, CAOs have no choice. Thus, the real
work of CAOs is building trust and gravitas so they will be listened to, sorting through the
many perspectives on data, leading the discussions necessary to help senior managers under-
stand the real issues, and helping chart a course.