� Page of 1 16
Ac#onable BI Analy#cs for Managing the Global Workforce
� Page of 2 16
As a result of technology innovation, the business value of data analytics and business intelligence is beginning to have a t a n g i b l e i m p a c t o n t o t a l t a l e n t management. The realized business value is only as good as how well an organization understands how to use data analytics—both tactically and strategically—for sourcing and managing talent. Those that do so put themselves in a position to win the war for talent.
Table of Contents
Sabermetrics …………………………………. 3 - 4
Talent Data Analytics from the Top ........ 5
"Data, Data Everywhere …” …………………6
Figuring Out Where to Start ………………..7
Challenges of Talent Management ………8
Decision-Making and BI Outcomes ….….9
BI Maturity Model ……………………….…….10
Actionable BI Frameworks ..………. 11 - 12
Requirements for Actionable BI….13 - 14
Five Key Takeaways.…………………………..15
Endnotes ………………………………………….. 16
� Page of 3 16
Sabermetrics
The potential impact of using data analytics to make
intelligent business decisions was brought to life in
Michael Lewis’ book Moneyball: The Art of Winning an Unfair Game that was turned into a film called Moneyball
in 2011 starring Brad Pitt and Jonah Hill. The central
premise of the book and movie is that data analytics on
player performance can predict how well they will
perform in given circumstances. Through rigorous
analysis, Oakland A’s General Manager Billy Beane and
his staff concluded that on-‐base percentage and
slugging percentage are be=er indicators of offensive
success than other outputs such as batting average, runs
batted in, and stolen bases.
A small-market Major League Baseball (MLB) team with
a payroll dramatically less than big-market teams such as
the New York Yankees, the Oakland A’s had to find other
ways to compete beyond identifying and keeping “best”
players, at least based on the standards typically used to
evaluate performance. Beane and his staff found the
answer in what is known as Sabermetrics, which enabled
them to rebuild their team with cheaper players based
on predictable outcomes using data analytics. To do so,
he hired a team of data scien#sts and business analysts
to help develop the algorithms needed to predict future
performance.
The premise of the book Moneyball: The Art of Winning an Unfair Game and the movie
Moneyball is that data analytics on player performance can predict how well they will
perform in given circumstances. This turns into strategic-prescriptive analytical model that
teams use to plan their rosters, draft prospects, and make trading decisions.
� Page of 4 16
With the success of the Oakland A’s in 2002 and 2003,
other MLB teams began employing Sabermetrics to
evaluate talent. Suddenly, every team hired data
scientists and business analysts to help them develop
the right algorithms and siA through mounds of data
on thousands of players in the major and minor leagues
to intelligently predict what their future performance.
The same tools that are being applied in MLB to win the war for talent are also being leveraged in the private
sector. Human resources (HR) and procurement leaders
employ ac#onable business intelligence to help their
organizations to recruit and source better candidates
(both full time and contingent), pay at competitive
market rates, gain strategic market intelligence, and
m a ke b u s i n e s s p r e d i c t i o n s r e g a r d i n g t a l e n t
requirements.
When Sabermetrics are applied to con#ngent talent management organiza#ons can…
• Recruit and source be=er candidates • Pay compe##ve market rates • Gain strategic market intelligence • Make data-‐driven business predic#ons
regarding talent requirements
� Page of 5 16
Talent Data Analytics from the Top
75% of business executives report talent data analytics are a
critical issue for their organizations.
However, only 8% believe their
organizations are doing a good job.
Data analytics can play an important operational and
strategic role if used correctly. Seventy-‐five percent of
business executives report that talent data analytics are
a critical issue for their organizations, enabling them to
achieve be=er results and gain a compe##ve advantage.
Yet only eight percent believe their organizations are
strong in this area.1 The outtake is that much work
remains to be done.
Executives are not wrong in their assessment.
Data-driven analytics such as talent data analytics
have substantial potential to impact bottom-line
results. For example, a study in Harvard Business Review from a few years ago found that
organizations using data-driven analytics see a five to six percent improvement in profitability.2 Of
course, to achieve these types of results, and
organization must possess a high degree of
readiness; saying versus doing is quite apropos in
this case.
� Page of 6 16
“Data, Data Everywhere …”
One of the most famous poems written in modern
western civilization is “The Rhime of the Ancient Mariner” by Samuel Taylor Coleridge. The poem is a
ballad that relies heavily on repetition and archaic
diction and is told by an old mariner to a man on his way
to a wedding feast.
The “ancient mariner” recounts how he shot an albatross
that was serving as an escort to the ship with his
crossbow. The previous luck he and his fellow sailors had
experience under the guidance of the albatross
evaporates and the ship is left to aimlessly drift when all
wind disappears. The en#re crew except for the ancient mariner dies, though they hang the albatross
around his neck before perishing to the depths of the
ocean. Eventually, through ghostly visions and
supernatural interventions, the ancient mariner is
spared. intervention
There is one stanza in the poem that is particularly
salient. Dying from thirst and without hope of any
intervention, the ancient mariner issues the line: “Water,
water everywhere, nor any drop to drink.” Data analysts
likely feel that way today.
The parallel stanza, “data, data everywhere, and not a drop to use,” is apropos. Indeed, surrounded by almost
unfathomable amounts of data, data analysts are oAen overwhelmed by its sheer volume and uncertain where to start. As a result, they are unable to leverage
the data or, at the most, merely scratch the surface of
possibilities.
“Water, water everywhere but not a drop to drink…”
� Page of 7 16
Consider the challenges facing data analysts and
scientists today. The amount of data is exploding: 90 percent of data is new every two years.3
Analyzing and making this immense sea of data
understandable and actionable is immensely challenging.
And HR is not unaffected. A report by IBM found that 60 percent of companies admit having disorganized HR
systems and moreover admit no way to make meaningful
data-driven decisions.4
Achieving actionable business intelligence from HR-
related data is most certainly a problem. Only 17 percent of HR organizations claim they actually use data
analytics, despite 73 percent conceding that data
analytics are important. 5
Figuring Out Where to Start
90% of data is new every two years
60% of companies admit having disorganized
HR systems and no way to make meaningful
data-driven decisions
Only 17% of HR organizations use data
analytics, though 73% concede analytics are
important
� Page of 8 16
Challenges of Talent Management
1. Controlling Costs 2. Talent Quality 3. Risk Management 4. Market Dynamics
Talent management is most certainly not without its
challenges. Controlling costs is a big issue. Multiple
talent-sourcing channels make it complex to manage.
When it comes to contingent workers, lack of
competitive bidding models drives up costs. Indeed,
organiza#ons may pay anywhere between 50 and 100 percent markup rates without competitive bidding and a
comprehensive understanding of market rates.
Talent quality is a second issue. Organiza#ons report they hire the wrong worker 27 percent of the #me.6
And when they do so, the cost is substantial. A single bad
hire costs more than $50,000. Differences in market
rates from one location to another and one skill set to
another create an additional set of challenges;
overpaying for workers drives up your costs, while
underpaying means you likely are losing high-value talent
to competitors. Further, the time it takes to source often
is the difference between securing the best talent and
losing them to your competitors.
When it comes to contingent talent, there are some very real tax and compliancy risks. There are a spate of
new tax and benefit laws on the books such as the
Patient Protection and Affordable Care Act and paid sick
leave laws. Organizations must understand the risks
associated with each of these and how to manage their
contingent workforce to mitigate them. In addition,
contingent labor often resides in silos, and thus
organizations lack enterprise visibility.
Finally, this on-‐demand workforce is highly dynamic;
market rates often fluctuate week to week, coupled with
multiple sourcing segments.
.
� Page of 9 16
Decision-Making and BI Outcomes
When it comes to talent analytics, there are different
types of decision-making. To begin, the fundamentals of
decision-making are either strategic or tac#cal. Strategic
decision-making examines issues across functions and
departments, often at a macro-level. The analysis looks at
emerging trends—both opportunities and threats—and
helps organizations prioritize them in terms of
importance. Tactical decision-making focuses on real-
time issues, enabling organizations and individuals to
make the best decision based on available data.
Decision-making also involves different business
intelligence outcomes. The base level is descriptive.
Descrip#ve analy#cs are retrospective in nature, using
data to explain what happened. Predic#ve analy#cs are
prospective in nature, using data to forecast future
outcomes. Prescrip#ve analy#cs—the most advanced
level—use artificial intelligence to show potential
outcomes and prescribe optimal recommendations.
Decision-Making
- Tactical
- Strategic
BI Outcomes
- Descriptive
- Predictive
- Prescriptive
� Page of 10 16
The intersection of strategic and tactical business approaches and the different business outcomes are overlaid on top of
each results in a BI Talent Maturity Model. For the purposes of this investigation, the focus is on contingent workforce
management. Contingent workers are defined as all non-employee talent (e.g., temps, independent contractors,
consultants, freelancers, et al.) sourced through staffing suppliers, independent contractors and consultants, freelancers
routed via freelance marketplaces, or supplied s part of a statement-of-work (SOW) project.
Descriptive: Data used to explain
what happened
Predictive: Data used to forecast
future outcomes
Prescriptive: Show potential
outcomes and prescribe optimal
recommendations
Tactical Strategic
Utilize individual behaviors and
hiring patterns in addition to profiles
and preferences. Graph algorithms
to map “people like you.”
Analysis of key insights on company-
wide hiring, interview trends, and
behaviors. Did you pass on top
candidates for the wrong reasons?
Leverage multiple data sources:
market rates, unemployment
rates, who is hiring = predict what
you will pay.
Model KPI results and make decisions
for locations, divisions, and skill sets
based on empirical data to predict cost
savings based on initiative.
Status and real-time metrics:
SOW funds depleted, time to fill,
market rates, ratings and reviews,
feedback, etc.
Metrics, KPIs, performance across
locations and divisions to feed
strategic planning. Analysis of self-
sourced vs agency-sourced talent.
BI Talent Maturity Model
� Page of 11 16
BI Maturity Frameworks
1. Tactical-Descriptive. The focus for tactical-descriptive
analytics is on real-‐#me metrics and status. An example
here is a manager seeking to know the current status of
funds for a SOW project bucket. The managers can
analyze funds depleted and determine if the project is
running under, on, or over budget. They can also sift
through quotes and identify those that are above, below,
or within market rates per skill set and location. Yelp-like
ratings, reviews, and feedback help guide them in
identifying the right SOW vendor.
2. Strategic-Descriptive. Strategic-descriptive analytics
look at metrics, performance, and KPIs across loca#ons and divisions. These are used to feed organizational-
wide strategic planning. An example of strategic-
descriptive analytics would be a comparison of self-
sourced talent versus talent from staffing suppliers. The
comparison might look at data such as cost, performance,
and temp-to-conversion ratios.
3. Tactical-Predictive. For tactical-predictive data
analytics, an organization can leverage mul#ple data sources such as market rates, unemployment rates and
trends, and those who are hiring (including competitors)
to predict pay rates for both talent and SOW projects.
These types of analytics enable organizations to source
top-talent talent and project suppliers at competitive
market rates. The ability to make these types of decisions
quickly based on predictive modeling also allows
organizations to avoid scenarios where they lose out on
sourcing hard-to-find talent.
Watch This Webinar with PRO
Unlimited’s SVP of Product
Development Ted Sergott.
“Actionable BI
Analytics for
Managing the Global
Workforce”
� Page of 12 16
4. Strategic-Predictive. Using strategic-predictive
analytics, organizations can model KPI results and make staffing and SOW project decisions for loca#ons, divisions, and skill sets based on empirical data. These
KPI models predict market rate changes for locations and
skill sets based on external factors. For example, a
location for a new data center may initially appear to be
an excellent decision based on salary analysis for full-
time workers and bill rate analysis for contingent
workers. However, decisions by other companies to
relocate data center operations to the area, organic
growth and expansion of existing data center footprints,
and/or changes in the local job market skew the initial
findings. A strategic-predictive analytics model accounts
for those, enabling organizations to make decisions that span a #me con#nuum and consider broader market forces and changes.
5. Tactical-Prescriptive. The prescriptive layer gets even
more strategic than the predictive layer. At the tactical
BI Maturity Frameworks (cont.)
level, analytics examine individual behaviors and hiring pa=erns and map candidate profiles and preferences to those. Then, leveraging graph algorithms, analytics
become much more actionable by making candidate
recommendations to hiring managers based on the
success of current and former workers. For example,
workers with certain degrees, work preferences, alumni
backgrounds may have a proven track record of better
success with a hiring manager than other workers. As a
result, talent management systems will recommend those candidates over others, with the recognition they
are more likely to be successful in the role being filled.
6. Strategic-Prescriptive. For strategic-prescriptive
analytics, the intelligent recommendations extend beyond the manager level to analysis of company-‐wide hiring, interview trends, and behaviors .
Organizations are able to avoid the scenario where they
passed on top candidates for the wrong reasons (e.g., are
there social, educational, professional biases that are
impacting the sourcing and selection process).
� Page of 13 16
1. Internal / External Data 2. Technology 3. Visualizations 4. Humans
When it comes to actionable BI analytics, there are four
requirements that organizations need to have in place:
1. Internal and External Data. The first requirement is
that the data behind the business intelligence must come
from internal and external sources. These two data sets
must overlap on top of each other, providing deeper
insights and complimenting each other.
2. Technology. A technology platform is the next piece
that you need. There are two things organizations need
here, and both imply integration. On the one hand, a
talent analytics platform must include a strategic data
warehouse that is integrated into your ERP systems. On
the other hand, every organization requires tactical
integration for their sourcing platform that shows
market bill rates, staffing and project supplier
performance scorecards, and other analytics.
3. Visualizations. Visualizations are the third thing
organizations need to get to actionable talent BI. These
need to be customized and show various views of KPIs
and business measurements. Examples here might
include data visualizations on headcount, spend, and
engagements across departments, skill sets, and location.
4. Humans. The final area is that humans are required.
The power of talent-related data analytics is that they
have the ability to challenge established perceptions,
influence new behaviors, and enable business leaders to
make more intelligent business decisions that impact
business outcomes. The crux of the problem is that most
organizations that do employ data analytics for talent
management do so in a rudimentary manner. Further, for
those that are able to conduct some level of data
analytics, less than half admit that they can utilize data
from outside of their HR systems.7
Requirements for Actionable BI
� Page of 14 16
Requirements for Actionable BI (cont.)
Organizations are inhibited in embracing talent analytics
because many simply do not have the skills to interpret
and apply the analytics at their disposal. A study by
McKinsey found that less than 18 percent of businesses
have the in-house skill sets needed for actionable BI.8
One of the core recommendations the study makes is
that technology is not sufficient on its own.
Organizations must supplement data analy#cs with actual humans—data scientists, business analysts, and
other subject-matter experts—who understand how to
interpret and moreover apply the actual data findings.
Data analytics augment, rather than eliminate, the
knowledge and experience of domain experts. Indeed,
anyone who has listened to Billy Beane speak about how
the Oakland A’s employed Sabermetrics knows that the data scien#sts and analysts behind the scenes were requisite fundamentals to getting business insights out
of the data.
.
Fewer than 18% of businesses
have in-house skills sets to
produce actionable BI.
Technology plus humans
required to interpret and
apply the data findings.
Less than half of HR
organizations that do use data
analytics for HR decision-
making admit they utilize
external data.
� Page of 15 16
The Sabermetrics of Talent Management hold great
potential for an industry segment that is an outlier when it
comes to the use of data analytics and BI. The following
are some key takeaways:
First, when data analytics are used for talent recruitment
and management, business performance improves.
Companies that use data analytics and specifically talent
analytics for strategic and tactical decision-making see
higher rates of return.
Second, a talent analytics maturity model consists of six
different areas—strategic and tactical that are overlaid on
top of descriptive, predictive, and prescriptive business
outcomes. Organizations that want to realize the full value
of talent analytics and business intelligence will ensure
their talent management solutions use all six.
Third, organizations that view data analytics as a
replacement to human judgment and experience will fail.
Certain business insights are simply not possible without
the involvement of humans.
Fourth, business platforms utilize many data sources and
must be integrated at the product level. For contingent
talent management, this includes integration into
procurement and HR workflows.
Fifth, there is a war for talent, and organizations that are
able to source hard-to-find talent at competitive market
rates have a strategic advantage. Actionable BI analytics
are one of the essential building blocks that
organizations need to institute in order to stay a step
ahead of their competition.10
A joint survey by MIT and IBM discovered that organiza#ons with advanced HR analy#cs see:9
• 8% higher sales growth • 24% higher net opera#ng income • 58% higher sales per employee
Five Takeaways
� Page of 16 16
1“Global Human Capital Trends 2015: Leading in the New World
of Work,” Deloitte University Press, 2015. 2 Andrew McAfee and Erik Brynjolfsson, “Big Data: The
Management Revolution,” Harvard Business Review, October 2012. 3 Paul Zikopoulos, Dirk deRoos, et al., “Big Data Beyond the Hype:
A Guide to Conversations for Today’s Data Center,” IBM, 2013. 4 “Advanced Analytics Report 2015,” Advanced Business
Solutions, September 2015. 5 Ibid. 6 “More Than Half of Companies in the Top Ten World Economies
Have Been Affected by a Bad Hire,” CareerBuilder Survey, May 8,
2013. 7 “Unlock the People Equation: Using Workforce Analytics to
Drive Business Results,” IBM Institute for Business Value,
Executive Report, December 2014. 8 Matt Ariker, Peter Breuer, and Tim McGuire, “How to Get the
Most from Big Data,” McKinsey, December 2014. 9 Andrea Capodicasa, “Why Are Big Data and Analytics Such a
Game Changer for HR?” Capgemini Blog, September 29, 2015. 10 Richard Dobbs, Tim Koller, and Sree Ramaswamy, “The Future
and How to Survive It,” Harvard Business Review, October 2015.
Endnotes
PRO Unlimited possesses 25 years of con#ngent workforce management and holds a number of industry firsts. It has been working with global enterprises to leverage data analy#cs for ac#onable BI for 15-‐plus years. To find out how PRO Unlimited can help you to implement an ac#onable BI approach, contact us today:
Phone: 1-‐800-‐291-‐1099 Email: informa#[email protected] Website: www.prounlimited.com
PRO Unlimited, through its purely vendor-neutral and integrated managed service provider (MSP) and vendor management system (VMS) solutions, helps organizations address the costs, risks, and quality issues associated with managing a contingent workforce. A
pioneer and innovator in the VMS and MSP space, PRO Unlimited offers solutions for e-procurement and management of contingent labor, 1099/co-employment risk management, and third-party payroll for client-sourced contract talent.
©2015 PRO Unlimited, All Rights Reserved | 1.800.291.1099 | [email protected]
Top Related