SHORT
CUT
ALIGN THE BUSINESS
How to easily master data analytics transformation
It’s common sense across industries:
Proficiency in data analytics (DA)
helps businesses make better and objective
decisions, cut costs, improve
quality and reduce risk.
With its potential, DA solutions using deep
learning or artificial intelligence techniques,
only two methods within the vast
spectrum of DA which ranges from diagnostic
and descriptive to prescriptive solutions,
empower enterprises to turn their
data into business value. However, over a
third of data analytics professionals say
their company never, or only sometimes,
puts their piloted DA solutions to use 1 .
That isn’t because these companies don’t
possess a small army of data scientists or
cutting-edge technology. Many companies
have built up dedicated teams and initiated
various DA initiatives. But it seems these
aren’t the only requirement to do DA well.
So, what’s going wrong? goetzpartners has
identified eight root causes that are often
responsible and has devised a clear strategy
and operating model for organizations
to overcome them and release the full
power of DA.
Common stumbling blocks
EIGHT ROOT CAUSES FOR
DATA ANALYTICS FAILURE
LACK OF
MANAGEMENT
SUPPORT
LOW
RELEVANCE OF
USE CASES
INSUFFICIENT
PLANNING
2
3
INSUFFICIENT
DATA SETS
1
4
DATA
ANALYTICS
FAILURE
To understand the root causes of DA failure,
it is first important to recognize the role
organizational structure plays. Typically,
all digital and innovation activities – such
as a DA initiative – begin in a separate area
of the company from the main operational
environment (e.g. data labs, innovation hubs
etc.). This innovation environment is where
DA pilot projects will be developed, before
being integrated into the operational environment.
It is at this point – the integration
phase – that the challenges arise and cause
many businesses to not advance their
DA initiatives past the pilot phase to
realize the benefits.
The first and always one of the most important
issues when it comes to significant
5
8
INSUFFICIENT
PRIORITIZATION
WITHIN IT
6
7
LACK OF
COMMITMENT
UNDERESTI-
MATED INTERNAL
RESISTANCE
LACK OF
RESOURCES
Data Analytics and
Artificial Intelligence
are not tools that just
need to be implemented
– they completely
change the way we
work.
changes in the way we work may be a lack
of management support. Despite higher
management stating their belief in DA,
they are often less committed to enforcing
implementation at the operational
level. Another root cause lies in the fact
that the DA use cases often have low relevance
to the business strategy and challenges
the business is facing. In fact, it’s
frequently the case that an application
will be developed and piloted because it is
possible – not because it is needed.
Insufficient planning may also be an issue
which can occur in many forms: There is
no realistic business case for the project;
the relevant people in the operational
environment or in the IT department are
not involved; there is no realistic timeline
in place or no legal sign-off. Another issue
which can founder a project is insufficient
data sets in terms of poor quality and/or
a lack of quantity. Challenges also arise
due to insufficient prioritization within IT.
The IT department is typically required to
implement the DA solutions into the operational
environment, but capacity and prioritization
issues can be a stumbling block.
Some businesses lack a strong-enough
commitment to DA, and thus business
units aren’t willing to integrate the new
1) Data Science Survey, Rexer Analytics 2017: http://www.rexeranalytics.com/data-science-survey.html