A Consumer’s Guide to Understanding Data - Data Fluency: Empowering Your Organization with Effective Data Communication - page 68

over another. For example, an executive dashboard that highlights quarterly
sales and new customers is authored by someone who celebrates growth.
Implicitly, the same dashboard minimizes the importance of customer retention
and cost control. You might be growing revenue, but at what cost? Priorities
will shine through.
4. What are the relationships between the different elements of data
shown? Much of data analysis and visualization is aimed at revealing the
connections and relationships in data. A trend chart shows the relationship
between a metric and time. A scatterplot chart shows how two metrics relate
to each other. Some of the relationships may already be obvious to you from
your experience; others may surprise you with new insights into how the world
works. View the data product as a tool for uncovering these relationships.
5. At what level of granularity is the data shown? There are different types
of insights that can be gleaned from data depending on whether it is sum-
marized or granular. Summarized data can show the big picture but glosses
over important relationships and drivers of results. For example, the average
height of a basketball team may be 6 foot, 4 inches, but this hides the fact that
there is a 7-foot center and two guards under 6 feet. Granular data exposes
individual correlations, outliers, and exceptions at the risk of believing a small
pattern applies broadly.
6. What is the scope of the data? To draw accurate conclusions from a data
product, you need to understand how the data has been filtered or narrowed.
Time range is a good place to start because data for the last week may tell a
different story from a year of history. A good data product often tries to cut
out data that isn't relevant to the discussion—perhaps there are discontinued
products, non-standard transactions, or categories of customers that would
skew your results.
7. Do you have a shared understanding of the meaning of all the data
fields? As you examine the data product, are there terms, calculations, or labels
that require more explanation? Discussions about data are often derailed by
different understandings of how a metric is calculated or what a dimension
represents.
What Can I Learn from It?
To become a savvy data consumer, you need to transition from understand-
ing “what it is” to “what it means.” This section considers how to find insights