Maersk Drilling - The Use and Usefulness of Employee Engagement Surveys: Myths and Realities

The Use and Usefulness of Employee Engagement
Surveys – An HR Analytics Approach
This session will take the participants through how The Maersk Group is using HR Analytics to work with
Employee Engagement Survey data.
(c) Copyright 2013, A.P. Moller - Maersk. All rights reserved. Not to be distributed without written consent

Name
Peter V.W. Hartmann, M.Sc. & Ph.D.
Business Intelligence Expert
Responsibilities in Maersk
Currently:
Responsible for HR Analytics/HR Business
Intelligence in Maersk Drilling
Previously:
Responsible for Assessment Tool & HR
Analytics in Maersk Group HR
Background
Research in Personality & Cognitive Ability
Psychometrics & development of
Assessment Tools
Contact information
Peter.hartmann@maerskdrilling.com
Phone: +45 63 36 18 22
Personal data

The Maersk Group
What do we do?

The Maersk Group
The A.P. Moller - Maersk Group is a diversified conglomerate with
89,000 employees in over 130 countries
We serve customers worldwide mainly in the shipping & energy area-
operating in five business segments:
 Container shipping
 Terminal operation
 Tankers, supply, towage & logistics
 Oil and gas activities
 Offshore Drilling

Who is Maersk Drilling?
2.102.000.000
Financial turnover 2014 (USD)
903.000.000
EBITDA 2013 (USD)
478.000.000
Net Operating Profit After Tax 2014 (USD)
1972
Founded
3.899
FTE (2015)
Core business: Fleet in operation 2015
22 (+1 coming in 2016)
Marketing &
Distribution
RefiningStorageTransportation
Exploration &
Development
Production
Upstream Midstream Downstream
Maersk Drilling provides offshore drilling services to the upstream segment.
To large international customers – e.g. ExxonMobil, Statoil, Shell, BP, ConocoPhillips.

HR Analytics at Maersk
How do we see it?

The practical application of metrics, statistics and research
methodology to provide valid information for better decision making
Descriptive
To use simple
statistics to display
standardized key
metrics in a user-
friendly format for
the purpose of
tracking progress
Linkage
To use statistics and
research
methodology to
generate new
insights and
translate these into
recommendations
Predictive
To use statistics
and research
methodology to
predict future
events and
recommendations

General Framework & Working Model
Communicate results
Display information in intuitive
fashion
Construct easy understandable story
illustrated with own data
Provide recommendations on actions
Analyze internal data
What data do we have? What does our data say?
What does our ”existing knowledge”
say about this?
Consult existing knowledge to build model
What do the theories say? What does the research suggest? What Maersk experts do we have?
Identifying the key areas of interest
What is our strategy focusing on? What is critical to the business? Who internally can we partner with?

What have we worked on & how?

Descriptive HR Analytics
To use simple statistics to display standardized key metrics in a
user-friendly format for the purpose of tracking progress
Descriptive
To use simple
standardized key
metrics in a user-
friendly format for
the purpose of
tracking progress

Survey facts
Number of people invited
2014: Approx. 84.500 2015: Approx. 87.000
Survey Types
2014: Paper = 23,500 (27%)/
Web = 62,500 (73%) 2015: Transition to fully digital...
Languages 2015
22 Languages
5 point rating scale
strongly
disagree
disagree neutral agree
strongly
agree
N/A
Responses rates
2013: 87% 2014: 90%
Vendor & start date
Vendor = IBM Started 2006
2015 EES key dates
Survey opens
26 Aug
Survey closes
02 Oct.
Results available to managers
04 November

Employee Engagement Trend
Four core elements of engagement at The Maersk Group:
Satisfaction with the company
Pride in working for the company
Willingness to recommend company to others
Intention to stay at the company
64
66 66
67
69
75
76
72
73
58
60
62
64
66
68
70
72
74
76
78
2006 2007 2008 2009 2010 2011 2012 2013 2014
Engagement External Top 25% Benchmark
External Top 25% Benchmark is derived from 400+ organisations surveying in 200+ countries. On average, the distribution of
blue-collar and white-collar respondents is 60/40 compared to Maersk Group’s 41/59.

Manager and Business Unit reports
•Each manager receives a detailed report on his/her direct reports
and indirect reports (for leader of leaders) IF he/she has more than 5
respondents (to preserve anonymity)
•Reports provide information on item and scale level with comparison
to
a) External benchmark (if any);
b) Maersk Group results;
c) Last year´s result
•Group HR provides “tool boxes” for
a) HR partners, on how best to facilitate the managers´ use of survey data
b) Managers, how to discuss results with team and decide on actions
c) Managers and HR Partners on how to address low engagement scores
through focused follow up
•Engagement survey results can be on individual managers scorecard

Linkage HR Analytics
To use statistics and research methodology to generate new insights
and translate these into recommendations
Linkage
research
methodology to
generate new
insights and
recommendations

Example: Looking at external research for Engagement to guide
analytics
Employee
Engagement
Safety
(-0.22)
Performance
(+0.14/+0.30)
Turnover
(-0.23)
Absenteeism
(-0.26)
Customer
Satisfaction
(+0.30)
Dispositional drivers
(e.g. Personality)
+0.24/+0.46
Situational & organisational
drivers
+0.3/+0.5

Example: Linking Engagement and Performance
Performance ->
Engagement
Engagement –>
Performance
Performance yr1 –>
Performance yr2
Engagement yr1 –> Engagement yr2
There is a strong link between
a team’s Engagement from one
year to the next
There is a moderate link
between Performance from one
year to the next
There is a small but relevant
link between Engagement and
Performance
There is a small but relevant
link between Performance and
Engagement.

Example: Linking Engagement and Business Unit PerformanceLongTerm
Performance
Long Term Engagement
BU´s Trend line
External research (e.g.
Judge et al, 2001;
Harter, Schmidt &
Killham, 2003) suggests
a moderate relationship
between Engagement
and performance.
Internal Maersk data
shows the same

Example: Addressing the issue of cause and effect in engagement
Willingness to
put in a little
extra effort
Individual and
team
performance
Observing that
discretional
effort matters
Feelings of
success and
accomplishment
Increased
optimism, joy &
team-spirit
Jury is still out on the true causality.
Review of available External research as well as internal Maersk data suggests
that it is likely to be subtle long term virtuous or vicious cycle where
engagement feeds into performance and vice versa.

“The Morale of the troops is vital for
winning the battle...”
Engagement is simply a practical way to
utilize that knowledge in a commercial
setting
The Engagement link to Customer
Satisfaction depends on how close the
business and customers are:
BU Distal BU Average BU Close
Engagement-
Customer
Satisfaction
Link
Example: Linking engagement and customer satisfaction

Example: Linking Engagement and Absenteeism

Example: Looking at the “spill-over effect” of the ability to engage
employees
Ability to engage
across levels
Team
Leader of
Others
Leader of
Leaders
r=xy
Ability to engage
across teams
Therefore, the
manager plays a crucial role
in engaging his / her team, over and above
the teams individual
perception/circumstances
r=xy
Therefore, the
Manager´s manager plays a role
in engaging his / her organization; but
Good managers engage their
employees irrespective of their
own conditions
Team A Team B

Drives
Engagement
Drives
Performance
Energetic
Sense of urgency; Flexible;
Results driven
Sociable
Self confident; Persuasive;
Team worker
Optimistic
Takes calculated risks; See
challenges instead of problems
Respect
Fair; Respectful; Inclusive
Conscientious
Predictable; Clarifies roles &
responsibilities
Trust
Honesty; Trustworthy; Truthful
We find high
performing leaders
with
ALL kind of
personality profiles,
as well as leaders that
can drive engagement.
However, some
trends
emerged when
analysing leaders
personality profiles
Example: Linking personality profiles to ability to drive
engagement and performance

Example: Leadership training impact through the Kirkpatrick model
Reaction
•What is the participant´s initial reaction to the training course?
•Can be divided into affective (”I like this”) and utility (”I can use this”)
•Important for “face value” and to avoid scaring people off. The felt utility seems to a better prediction for later effects
Learning
•What knowledge & skills have been acquired?
•Can be divided into what has been learned, what is initially retained and what can the be reproduced – all in the training
situation (can be further separated into content e.g. cognitive, interpersonal, psychomotor, etc.)
•Important for determining the “input” the participants are leaving with and potentially bringing back to the office
Behavior
•What “on-the-job” effect is there on behavior, performance, etc.?
•What “learnings” have been transferred e.g. actually brought back to the office and given the opportunity to use/display
(e.g. manager support, opportunity to change and implement, etc.)
•Important for the actual impact on how the job is done and what can be directly observed
Results
•What impact does the training have on the overall “hard-core business metrics”?
•Ultimately, overall profitability, but can be broken into customer satisfaction, overall productivity, etc.
•Important for the final evaluation of the return on investment. Training cost vs. increased profitability
External research suggest a moderate effect size of d = 0,6 (but varies) for training for each of the
four levels. However, the relationships between levels are weak (small correlation) and the spill-
over is diminishing the more distal they become; e.g. Learning (0,66d)=>Behavior (0,44d). So
one is not a ”guarantee” for the other, due to several factor influencing the ”flow between levels”
???
!!!
$

Example: Testing Leadership training impact through
longitudinal approach with control for sampling
2011
Q3
Collection of
EES2011
2012
Q1
Appraisal
scores for 2011
consolidated
2012
Q1-Q2
Possible
attendance
2012
Q3
Collection of
EES2012
2013
Q1
Appraisal
scores for 2012
consolidated
2013
Q2
Data
consolidated
and analysed
Gain for not-trained
Gain for trained
Training effect
Potential intervention

Predictive HR Analytics
To use statistics and research methodology to predict future events
and translate these into recommendations
Predictive
To use statistics
and research
methodology to
predict future
events and
recommendations

Unknown Never been in the
bottom in the
past 2 years
Been in the
bottom last year
Been in the
bottom the last 2
years
1 in 4 1 in 8 2 in 5 1 in 2
21-23% 12% 39% 50%
Linkage & Predictive HR Analytics
Example: Assessing risk of managers ending in the external bottom
30%
We need to intervene, otherwise the risk is very high for managers to end up being
in the bottom 30% again...
As a manager, what are your chances of falling in the bottom 30% on
Engagement next time?

Example: Looking a stability of managers ability to drive
engagement
2011
2012
2011
2012
2011
2012
2011
2012
Engagement
Manager
Effectiveness
Values Index
Performance
Enablement
• Engagement, MEI, VI & Performance
Enablement remains fairly stable from
one year to the next.
• This year’s EES team results are
strong predictors for next year’s EES.
• That means that a team will remain
as Engaged or Unengaged as they
have always been,
unless something is done
about it...
What is the Engagement like for teams where
action has been taken since last year
EI MEI VI
2012
2011
We are unlikely to see a change in EES results, unless something is done about it...
r=xy r=xy r=xy r=xy

Example: Looking at drivers of engagement
= EI
+1
Focus
area A
+2
Focus
area B
+1,5
Focus
area C
+ 2,5 Focus
area D
+1,75
Focus
area E
+ 3
How much do we need to improve on the 5 focus areas to increase
Engagement by one point?

Incidents
With injury:
LTI & fatalities
Without injury:
Observations,
”near-misses”,
etc.
Predictors
Training
Engagement
Example: Linking Engagement and Safety in a BU
=> Lead to recommendation on how many LTI could be prevented
by raising lower engaged entities to the BU average
Injuries:
LTI & fatalities
Safety Climate:
Organizational
commitment to
safety
Personal Safety
Attributes:
Knowledge &
Motivation
Safety
Performance:
Compliance &
Participation
Safety violations:
“Near-misses” &
Observations
The scientific model The actual model

Drivers
HSE
Performance Customer
Satisfaction
HSE
– Material &
Personnel
Training
Safety
culture
Maintenance
of
equipment
Engagement
& Managerial
effectiveness
Operational
performance
Retention
Example: Looking at drivers of operational performance and
customer satisfaction
Safety is
*one of the causes of operation performance and customer satisfaction
*a consequence of performance excellence drivers

Lessons learned

HR Analytics
Data quality
Data knowledge
Data availability
HR Governance
HR IT Systems
HR Processes
First lesson learned:
Know your data and where it comes from
Knowing your
data will
increase the
validity of your
analytics
All data comes
from
somewhere

Second lesson learned:
Analytics requires several iterations
Descriptive
To use simple
standardized key
metrics in a user-
friendly format for
the purpose of
tracking progress
Linkage
research
methodology to
generate new
insights and
recommendations
Predictive
To use statistics
and research
methodology to
predict future
events and
recommendations
An iterative process
*Metrics & models need
to be constantly refined
*Insights need to fuel
adjustments or new
metrics
A long term investment
* A marathon, not a sprint
* Managing expectations
and communication
* No quick fixes
Knowing the data
*Systems, processes
& governance
* Availability & quality

Third lesson learned:
Success through relevancy, mind set & collaboration
The
right
question
The
right
stake-
holders
The
right
SMEs
Success
It is easier when HR Analytics is truly an
organizational mind set characterized by
open to discussion and willingness to
investigate and change rather than just an
analytics function
All are these factors are
necessary but not sufficient
to ensure success

Managing communication
Realistic expectations
Probabilities
A part of the answer
Takes time
Fixed expectations
Quick fix
One clear answer
Fuzzy expectations
Some insights
Cannot say
Managing expectations
Fourth lesson learned:
Managing expectations & communication is essential
Effective communication
Likelihood scenarios
e-(β*X + α) ≈ 42%
Recommendations
Technical communication
p = e-(β*X + α)
Complex communication
All details, assumptions
& disclaimers

Fifth lesson learned:
Analytics needs to be actionable
Actionable
• Ensuring that our insight are easy to
understand and communicate as well as
being actionable in order to support the
business
Accuracy
• With inspiration from external knowledge,
we strive to apply research methodology to
generate actual insights and to reduce the
margin of error of the recommendations

Thank you!
Questions?

Maersk Drilling - The Use and Usefulness of Employee Engagement Surveys: Myths and Realities

More Related Content

Viewers also liked (14)

Similar to Maersk Drilling - The Use and Usefulness of Employee Engagement Surveys: Myths and Realities (20)

More from CBS Competitiveness Platform (20)

Recently uploaded (20)

Maersk Drilling - The Use and Usefulness of Employee Engagement Surveys: Myths and Realities