RETHINK DATA: PUT MORE OF YOUR BUSINESS DATA TO WORK—FROM EDGE TO CLOUD | 35 |
DataOps: Getting to Customer Satisfaction and Prot Through Data
This survey established that a
smooth-functioning DataOps is key
to data management that enables
organizations to get more value
out of data and boosts business
outcomes such as prot and
customer satisfaction.
How can businesses get there?
As noted in the preceding section,
the human element of the equation
cannot be overstated. It’s the
people who keep data in silos.
Consequently, the way to institute
effective DataOps is not just about
having the right tools. To be sure,
the right tools are key. Virtualization
tools are immensely useful, if not
necessary, in that they allow for the
retrieval and manipulation of data.
Whether it’s a software virtualization
plane (such as Kubernetes, “a
portable, extensible, open-source
platform for managing containerized
workloads and services, that
facilitates both declarative
conguration and automation”)
or virtual machines that are an
abstraction mechanism for app
deployment, virtualization layers can
streamline data management.
But before DataOps avails itself
of virtualization, it has to start with
decisions about data. Putting tools to
work is the easy part. It’s trickier for
enterprise leaders to make decisions
about data via data governance and
processes around it.
Business owners need to start
with their goals—which commonly
include boosting customer
satisfaction and prot. To get
there, they need to interrogate
the data at their disposal.
They need to coordinate data
from various sources and solve
governance issues such as:
• Who has access to what data?
• How to classify data?
• Which data to keep and where?
• What to do with data after it’s
analyzed?
• How to make data available?
• How to interconnect data?
In order to make these
determinations, business owners,
data administrators, or CIOs
need to consult subject matter
experts (SMEs). Along with
SMEs, they need to identify,
evaluate, cleanse (detect and
correct inaccurate records),
and validate data. SMEs need
to be at the table because only
they possess deep knowledge of
certain types of data.
It’s with the involvement of the
SMEs, under the direction of
business owners, and facilitated
by data administrators that
determinations need to be made
regarding a key question: What
intelligence do you want out of
what data? What do you want it to
tell you? How do you want to use it?
There can be about 10,000 entry
point parameters collected for a
single product that a company
manufactures. If the company is to
retain all these points of information
for just one product, without a
clear data architecture dening
where it will all be stored and how
it will ow between environments,
the data risks drowning in a data
swamp. The decision-makers
need to confer with product design
engineers and quality engineers,
and ask: Which of the 10,000
parameters are the most critical?
Then, through interrogating and
tracking the curated data, they are
on a more efcient path to building
components and solutions.
The upfront work of data
governance—involving
coordination, discussion, analysis,
agreements around language,
and data classication (more on
this in the following chapter)—can
give way to the mostly-automated
downstream processes facilitated
by virtualization tools.
This twofold process of DataOps
can then reap the results of
customer satisfaction because
optimizing the governance and
ow of data leads to better quality
of offerings, which directly affects
how customers feel about their
purchases. As IDC analysts pointed
out in the previous chapter, speed
of delivery matters too. Getting
to results faster means greater
customer satisfaction—because
getting to data faster means
customers and business owners
can make decisions faster.
This is how a process that starts
with business owners setting out
to increase customer satisfaction
and prot attains these very goals
through optimizing data via DataOps.
It’s about being intentional on
making the most of data.
SEAGATE POV