[PPT] - HR Analytics Workshop 6th June 2018 C R A F T E D B Y C O N C E N T PowerPoint Presentation

SLIDE 1

C R A F T E D B Y C O N C E N T R A

HR Analytics Workshop

6th June 2018

SLIDE 2

2

Giles Slinger, James Gardner, Min Bhogaita

HR CIPD Workshop

Welcome & Introductions

SLIDE 3

3

C R A F T E D B Y C O N C E N T R A

Giles Slinger

Director, OrgVue & Managing Director of Concentra Netherlands Giles.slinger@orgvue.com

Introductions

James Gardner

Engagement Lead, Analytics Solutions and Services, Concentra UK James.Gardner@concentra.co.uk

Min Bhogaita

Analytics Director, Concentra UK . Min.Bhogaita@concentra.co.uk

SLIDE 4

4

C R A F T E D B Y C O N C E N T R A

Agenda for the day

Section Content Timings

1. Welcome & Introductions

Purpose, background and capture additional items attendees want to cover on the day 9:30am-9:40am

2. Industry Insights

What we have seen across industry, emergent themes from HR functions, consumer led expectations and user experience 9:40am-10:00am

3. Tomorrow’s World

New applications of technology within the HR space; such as unstructured text analytics and game-based assessment 10:00am-10:45am COFFEE, CALLS & EMAILS BREAK

4. Analytics for Insight

How to get alignment, traction and a plan for Analytics in your Firm;

Running an analytics workshop
Introducing key concepts such as the data-driven organisation and competency

frameworks 11:00am-12:30pm LUNCH PROVIDED

4. Analytics for Insight (cont.)
The importance of demos, or Proof of Concepts
Practical exercises
Following up

1:15pm-2:15pm

5. Analytics for Change
Modelling systems in multiple-dimensions
Examples of impact
Practical exercise

2:15pm-4:00pm

6. The one thing…

What’s the one thing you’ll take from today? 4:00pm-4:30pm

SLIDE 5

5

C R A F T E D B Y C O N C E N T R A

Last time around, many participants had no analytics background and almost half came from small & medium sized companies

SLIDE 6

6

C R A F T E D B Y C O N C E N T R A

In 2016, building capability, getting meaningful insights and data quality were the biggest challenges

SLIDE 7

7

HR CIPD Workshop

Industry Insights

Change in industries Myths about HR analytics Opportunities for HR analytics

SLIDE 8

8

C R A F T E D B Y C O N C E N T R A

Businesses are being transformed by data analytics

Typical questions across the organisation

What recruitment sources are most likely to provide us with good employees?

Procurement Analytics Customer Management Employee Records Stock Control

What should we recommend to this customer? Which materials are in inventories across multiple sites? How much do I spend with each supplier?

SLIDE 9

9

C R A F T E D B Y C O N C E N T R A

Exciting new tools are coming to HR … are we ready?

SLIDE 10

10

C R A F T E D B Y C O N C E N T R A

“The skills of uncovering “insight” and being able to communicate this effectively as a “story” that correctly influences human capital decisions, is of increasingly critical importance in the global economy.” “Senior managers, line managers and employees all want HR to be proactive - independent minded, helping managers look ahead and tackling issues of strategic importance.” “Having achieved a valued role supporting strategy, we now need to be proactively visualising and preparing for the future.” “Work will become more insight-based, specialist and collaborative. In the war for talent our strategies must become smarter, and more focused on the performance of teams. HR will become the source of most competitive advantage.” If the CIPD were asking you about the future of HR, and the biggest challenges it faces today, what would your answer be?

Practitioners point to proactive data analytics as being the future of HR

SLIDE 11

11

C R A F T E D B Y C O N C E N T R A

Myths and answers

But a range of myths have been provided as excuses for not embracing a data-centric approach in HR

1. Our organisation doesn’t value analytics 2. Everything needs to be on one system before we can start 3. Our HR data isn’t good enough 4. It’s an 8 year journey to get results 5. Analytics is too hard Practise turning intuitions into business impact Map the owners of existing datasets and talk impact Focus on what’s important, create rules, and if need be, crowdsource more Not true – the sooner you start, the cleaner the data will be Not true – today’s tools reduce the cost of being curious

A systematic process can help you make the most of your data and overcome these traditional blockers.

SLIDE 12

12

C R A F T E D B Y C O N C E N T R A

The average large organisation has 8 different databases. The spread of

data across disparate systems generates multiple versions of truth.1)

>90% of companies have chronic problems with their data, e.g. missing,
utdated, inaccurate, typos…1)
Less that half of organisations have the tools for proactive HR; only 14%

are able to visualise data to plan for the future.

In many organisations, data analytics is still the preserve of data scientists

and statisticians.

HR too tied up in the ebb and flow of daily administration and not given

necessary tools to produce genuine insight.

HR Leaders’ responses to “We are getting significant returns on analytics investments.” Senior Business Leaders’ responses to “HR Data Analytics has led me to change a business decision in the past year.”

*Source: CEB (2013), The Analytics Era: Transforming HR’s Impact on the Business

So it is still rare for organisations to access the full potential of People data

Data is incomplete and under-used “For 85% of business leaders, more data is not adding up to better business decisions”*

SLIDE 13

13

C R A F T E D B Y C O N C E N T R A

Corporate activity: mergers, divestments
Changing structures: new organisation configurations
New versions of analysis
Changing systems: new applications, new systems, new databases
Changing markets: new products & services, new activities, new

teams

Changing people: skills, names, responsibilities, grades, salary, job

titles (slow moving data)

Don’t panic! 1. Data will always be messy

Shire goes shopping after AbbVie deal falls through

Shire is reportedly back on the acquisition trail, after AbbVie's board turned its back on its £32bn takeover attempt last week. The FTSE 100 company is believed to have reopened discussions with various takeover targets, including NPS and Cubist, it was reported

ver the weekend…

Lauren Davidson, Daily Telegraph, 19 Oct 2014

“>40 different systems”

Reasons for messy data

SLIDE 14

14

C R A F T E D B Y C O N C E N T R A

Limitations to the software being used (e.g. excel)

gradually being overcome with visualisations

Data from multiple sources can be connected to

uncover connections across data types: e.g. information about same employees scattered across datasets

Faster, simpler tools can help overstretched HR

departments; ‘reducing the cost of being curious’

2. Tools are becoming more effective

Reasons HR can aspire to better analytics

SLIDE 15

15

C R A F T E D B Y C O N C E N T R A

The reactive mindset came from being trapped with the worst tools in the business
HR can turn the tables and make the business take ownership of poor data
HR can combine its datasets with Finance to align definitions and establish one source of the

truth

HR can combine datasets with business outcomes (sales, profitability, CSAT, mortality) to

connect people decisions to strategic results

Communication across departments can be developed: overcoming organisational ‘silos’
HR can move from ‘reactive’, day-to-day tasks to proactively defining and delivering strategy

and planning the long-term To-Be

3. HR can take a proactive and strategic role

Why HR can move from reactive to proactive

SLIDE 16

16

C R A F T E D B Y C O N C E N T R A

Data types
Analytical stages

HR data and HR analytics

SLIDE 17

17

C R A F T E D B Y C O N C E N T R A

What data is covered in HR analytics?

HR analytics join HR metrics to Organisational outcomes

SLIDE 18

18

C R A F T E D B Y C O N C E N T R A

HR analytics as a journey

HR Analytics methods are sometimes shown as a step-by-step journey

Added Value HR Analytical Capability

HR metrics Benchmarking Surveys Tracking/Trends Correlations Insights Predictive Self Serve Any others??? Experience Source: Bersin By Deloitte 2013 Source: Jeff Nelson & Nathan Adams, Aviva 2015

Must we progress step by step through the type of analytics we carry out?

SLIDE 19

19

C R A F T E D B Y C O N C E N T R A

…

Discuss: Where are you in the journey? What do you think about the idea of step by step progress?

SLIDE 20

20

HR CIPD Workshop

Tomorrow’s World

New types of data collection:

Free text
Game based
Physical: the Buzz Box

SLIDE 21

21

C R A F T E D B Y C O N C E N T R A

Employee Feedback using free text - Workometry

Traditional employee feedback surveys are confirmatory. Workometry is a feedback platform that uses machine learning and free text analytics to analyse responses to open questions spotting themes without preconceived bias

SLIDE 22

22

C R A F T E D B Y C O N C E N T R A

Free text to Themes..

Employee Feedback using free text - Workometry

How it works.. Benefits.. Further Information..

Email: Andrew.Marritt@organizationview.com www.organizationview.com

SLIDE 23

23

C R A F T E D B Y C O N C E N T R A

Game Based Aptitude & Personality Traits Assessment - Arctic Shores

Candidates are more successfully matched to roles through trait profiling. It aims to eliminate recruitment bias,

vercomes diversity-linked timed test bias, creates a positive candidate recruitment experience
Measures aptitude, cognitive and personality traits

SLIDE 24

24

C R A F T E D B Y C O N C E N T R A

Outputs and Feedback..

Game Based Aptitude & Personality Traits Assessment - Arctic Shores

How it works.. Benefits.. Further Information.. Email: Tomas.Kuzmickas@arcticshores.com www.arcticshores.com

SLIDE 25

25

C R A F T E D B Y C O N C E N T R A

SLIDE 26

Physical Data Collection – the Buzz Box from ‘We Love Surveys’

Buzz Box

Quick and simple surveys Ask up to 5 questions Tops are interchangeable Excellent response rates Less than 20 seconds to complete Anonymous Report experience by the minute

Prism Application

Real Time Feedback
Works Offline
Simple or Complex surveys
Completely bespoke design
Anonymous (unless
therwise requested)
Perfect for fixed and mobile

use

Multi-Lingual

Web Surveys

Real Time Feedback
Invite via email, social

media, website etc

Simple or Complex surveys
Responsive as standard
Bespoke design
Multi-Lingual
Anonymous (unless
therwise requested)

SLIDE 27

Physical Data Collection – for Everyday Engagement

In many organisations there is a reliance on annual employee surveys and customer mystery shopping Following these surveys reports can take a significant time to be received Mystery Shopping is a quick snapshot and can be inconsistent

Scores very often do not reflect the associated comments.
2 points in the year cannot give a reflective view of the customer

experience

Quite often the results of staff surveys are out of date before being received

Questions are often not about things that the local manager can

impact

We Love Surveys believe in everyday feedback

We give front line managers and organisations information about how their customer and colleagues feel every day, allowing them to react quickly to challenges and issues. We help you to increase customer satisfaction and turnover through better service delivered by happy people

SLIDE 28

Example: moving from annual surveys to monthly team-level pulse to address the local drivers of engagement

Image credit: 2012 Orange, Concentra Analytics Ltd

SLIDE 29

29

HR CIPD Workshop

Analytics for Insight

Running an Analytics Workshop
Introducing Key Concepts
Understanding Commercial Objectives
Art of the Possible
Understanding your Vision
Project Prioritisation

SLIDE 30

30

C R A F T E D B Y C O N C E N T R A

Running an Analytics Workshop

SLIDE 31

31

C R A F T E D B Y C O N C E N T R A

Overview of the facilitated workshop approach

Notes

Workshop should be as interactive as possible; our role is to facilitate not consult
All outputs should be captured and wrapped up into a playback deck of the day
Avoid discussions on technology selection; the purpose is to agree priorities and next steps

A facilitated workshop will create alignment on business priorities, lock in sponsorship for analytics and develop a 30-60-90 plan to accelerate your capability

Introductions & Scope setting

Outputs:

30-60-90 day roadmap
Initial Vision for Analytics

Agree short term roadmap Project Selection and Prioritisation Setting your Analytics Vision Demonstrating the Art of the possible Brainstorming tough questions Understanding the Business Strategy and Commercial objectives in scope Introduce key Analytics Concepts (DDO) Rules of the Workshop

1 2 3 4 5 6 7 8 9

SLIDE 32

32

C R A F T E D B Y C O N C E N T R A

Start with the end in mind..

Introductions & Scope setting

1

SLIDE 33

33

C R A F T E D B Y C O N C E N T R A

Get everyone in the room..

Rules of the Workshop

2

SLIDE 34

34

C R A F T E D B Y C O N C E N T R A

Introducing Key Concepts

3

SLIDE 35

35

C R A F T E D B Y C O N C E N T R A

What is a Data-Driven Organisation?

“An organisation that leverages insight in its decision making and achievement of its commercial

bjectives, continually developing the right mix of people, process, data, strategy and technology to

achieve this.” Benefits for being a Data-Driven Organisation

Faster

Make faster decisions as a Business when you reduce your ‘time to insight’

Cheaper

Improved use of technology, data, and process can reduce the production cost of current reporting and allow for further innovation

Smarter

Increased commercial effectiveness through the improved use of data in

perational and management

decisioning

Introducing Key Concepts

3

SLIDE 36

36

C R A F T E D B Y C O N C E N T R A

How does Data drive action?

Consider all elements of people, process, data, infrastructure and strategy you need throughout the process of insight generation to mature your internal capability

Data Insight Decision Analysis Action

Source : Gartner

Introducing Key Concepts

3

SLIDE 37

37

C R A F T E D B Y C O N C E N T R A

Analytics Maturity Model – where are we?

Consider all elements of people, process, data, infrastructure and strategy

Emerging drivers for

developing Analytics

No/Basic Analytics

Strategy

Low appreciation for

potential value/use cases

Low data quality and KPIs

fragmented, poorly aligned and no clear ownership

Low appreciation for data

governance, or how data can be managed as an Asset

Forming link between

business outcomes, drivers for change and Analytics Capability

Analytics Strategy

Forming across Functional Areas

Proof of Concepts/Value

used across the

rganisation as part of

Analytics methodology

Synergies emerging and

being acted upon

Progressing to single

source of truth and basic governance

Business Strategy and

Analytics form an active ecosystem

Insight delivery model and

Service catalogue in place across all functional areas

Innovation and disruption

baked into roles and toolsets

Analytics linked to business

impact and commercial

gains. Moving to self-

funding.

Clear data governance

practices, known supply chains, business owners

Initial Awareness Building Traction Integrating Capability Widespread Value Generation Fully Integrated 1 2 3 4 5

Introducing Key Concepts

3

SLIDE 38

38

C R A F T E D B Y C O N C E N T R A

…

Discuss: Where would you place yourself on the Analytics Maturity Model?

Emerging drivers for

developing Analytics

No/Basic Analytics

Strategy

Low appreciation for

potential value/use cases

Low data quality and KPIs

fragmented, poorly aligned and no clear ownership

Low appreciation for data

governance, or how data can be managed as an Asset

Forming link between

business outcomes, drivers for change and Analytics Capability

Analytics Strategy

Forming across Functional Areas

Proof of Concepts/Value

used across the

rganisation as part of

Analytics methodology

Synergies emerging and

being acted upon

Progressing to single

source of truth and basic governance

Business Strategy and

Analytics form an active ecosystem

Insight delivery model and

Service catalogue in place across all functional areas

Innovation and disruption

baked into roles and toolsets

Analytics linked to business

impact and commercial

gains. Moving to self-

funding.

Clear data governance

practices, known supply chains, business owners

Initial Awareness Building Traction Integrating Capability Widespread Value Generation Fully Integrated 1 2 3 4 5

Introducing Key Concepts

3

SLIDE 39

39

C R A F T E D B Y C O N C E N T R A

Analytics Capability framework

Developing Analytics as a competence has a number of elements under people, process, data and infrastructure governed by a coherent strategy all of which focus on insight generation

Data Infrastructure Strategy Process People

Data Footprint

(incl. measures and KPIs)

Data flows &

Sources

Data Governance
Security
Business

Ownership

Architecture
Vendors and

Partners

Innovation

Sandbox

Cloud vs On-

Premises

Insight delivery

model

Alignment to

commercial strategy

Analytics

Roadmap

Benefits tracking
Internal comms
Service

Catalogue

Requirements

management

Demand pipeline
Reporting

lifecycle Mgt

Business Process
Embed/Adopt
Sponsorship
Ownership
Talent

Management

Best Practice

Sharing

Learning

Introducing Key Concepts

3

SLIDE 40

40

C R A F T E D B Y C O N C E N T R A

Why do organisations struggle to become Data Driven?

Typical Blockers Identified

Lack of consistent KPIs and measures Incoherent Data model Culture change and ways of working Poor implementation experiences Lack of executive sponsorship Overwhelmed by internal data volumes and variety Challenging relationship between Business and IT Poor adoption or embedding

Introducing Key Concepts

3

SLIDE 41

41

C R A F T E D B Y C O N C E N T R A

What are the people and skillsets that DDOs develop?

Combining Technical and Commercial skillsets to solve business problems Technical Competences

Creates effective visualisations Structured analysis & Data Modelling Possesses software competences Ability to extract & manipulate data Performs technical and hypothesis testing

Business Competences

Understanding the commercial context Awareness of Business Process and Performance Measurement Ability to build compelling narratives; story telling Possesses soft skills for stakeholder engagement Applies technology to solve business issues

Introducing Key Concepts

3

SLIDE 42

42

C R A F T E D B Y C O N C E N T R A

Understanding commercial

bjectives and associated

people challenges

SLIDE 43

43

C R A F T E D B Y C O N C E N T R A

Link commercial objectives to people challenges..

Understanding Business Strategy & Objectives in Scope

4

SLIDE 44

44

C R A F T E D B Y C O N C E N T R A

Link people challenges to tough questions..

Brainstorming Tough Questions

5

SLIDE 45

45

C R A F T E D B Y C O N C E N T R A

Words, words, words – but what’s the underlying message you are

conveying in everything from job descriptions to annual reports?

The impact of chat-bots on the HR department, now and in the future
Industry leading tools for Analytics

Art of the possible

SLIDE 46

46

C R A F T E D B Y C O N C E N T R A

Art of the Possible with Min

Min Bhogaita

Analytics Director, Concentra UK

Words, words, words – but what’s the underlying message you are

conveying in everything from job descriptions to annual reports?

The impact of chat-bots on the HR department, now and in the future

Art of the Possible

6

SLIDE 47

47

C R A F T E D B Y C O N C E N T R A

Analytical tools vary in their features and strengths

Overview of common analytics tools

Basic data handling & models Suitable for beginners Flexible and well supported Manipulating / integrating data Suitable for capable Excel users Visually manage complex workflows Free trial link Workforce analytics and planning HR Analysts & senior managers Integrating multiple sources & visualising Request demo Statistics and graphics Analysts with stats degrees Multiple stats tools, allows coding Download link Visualising data / dashboarding Suitable for beginners Fast, intuitive drag & drop Free trial link Org Design, HR analytics, Transition Management Suitable for capable Excel users Messy people data Request demo Creating an analytics “portal” Suitable for capable Excel users Great UI - no coding required Personal version link Embedded OEM analytics All HR users No data integration required

? ? ? ? ? ? ? ?

Integrated OEM analytics

Art of the Possible

6

SLIDE 48

48

C R A F T E D B Y C O N C E N T R A

Combine different tools to find the solution that works best for you

Data storage Data manipulation and analysis Data visualisation

Develop a reliable single data set with latest information that is structured to provide fast analysis Build reporting and modelling tools to answer questions about the past, present and future Create an accessible environment to publish analysis and share insights OrgVue Manipulation of complex data for integration into user portal Automated processing of messy data for intuitive visualisation Standard HRIS integration for HR analytics/ modelling Examples:

Art of the Possible

6

SLIDE 49

49

C R A F T E D B Y C O N C E N T R A

Understanding your vision for Analytics

SLIDE 50

50

C R A F T E D B Y C O N C E N T R A

A simple exercise for building a Vision statement..

Setting your Analytics Vision

7

SLIDE 51

51

C R A F T E D B Y C O N C E N T R A

Sample Vision Statement

Setting your Analytics Vision

7

SLIDE 52

52

C R A F T E D B Y C O N C E N T R A

Project Identification and Prioritisation

SLIDE 53

53

C R A F T E D B Y C O N C E N T R A

Translate your tough questions into potential projects..

Project selection and Prioritisation

8

SLIDE 54

54

C R A F T E D B Y C O N C E N T R A

Prioritise your projects..

Projects or Proof of Concepts are plotted based on Business Value and Ease of Implementation

Notes :

Projects or PoCs should be achievable within 90 days
Spread out the ‘wins’ across the period
Not everything should be a technology project; think about stakeholder engagement

activities too!

Long List of Projects and Pilots (max. duration 90 days)

Ease of Implementation Business Value / Impact Best suited for the 30-60-90 day plan

A vs B

Create our Short List Plot them in a 2 x 2

Project selection and Prioritisation

8

SLIDE 55

55

C R A F T E D B Y C O N C E N T R A

Priority setting by crowd-sourcing: A vs B, Crowdoscope examples

Diverse methods for collecting priorities

Project selection and Prioritisation

8

For more info go to www.silvermanresearch.com For more info go to www.orgvue.com

SLIDE 56

56

C R A F T E D B Y C O N C E N T R A

Sample Prioritised List of Projects

Project selection and Prioritisation

8

SLIDE 57

57

C R A F T E D B Y C O N C E N T R A

30-60-90 day plan

No one left unassigned..

Action Item Business Owner Target Date

Agree short term roadmap

9

SLIDE 58

58

C R A F T E D B Y C O N C E N T R A

Follow-up Checklist

Every action assigned
Every person has at least 1 action
Get the agreed actions out quickly
Set up diary items at the 30, 60 and 90 day markers for the workshop attendees
Write up the content, including images of the day then distribute!
Maintain communication and ‘face time’ to keep momentum across the group
Showcase the workshop event and the outcomes as wide as possible (lunch n learns)

Agree short term roadmap

9

SLIDE 59

59

C R A F T E D B Y C O N C E N T R A

3 traps
Power of visualisation
Analytics and context

HR Analytics in practice

SLIDE 60

60

C R A F T E D B Y C O N C E N T R A

Some practical questions:

How complete and how ‘big’ do the data need to be?
Who should be the owner of People Analytics? HR? Finance? Someone else?
How to build a team? Internal vs. external capabilities
Who is the audience for HR Analytics?
What impact can HR Analytics have?
What tools should we use? Existing vs. new tools
What could we stop doing?

Discuss the question on your table with your group

…

SLIDE 61

There are many traps in HR analytics, that you need to be aware of. But there is too much value here to ignore

Common issues include:

Sample size
False correlation
Meaningfulness
Prediction
Treating people as mechanisms

Common opportunities include:

Predictable issues that the organisation prefers to

ignore – especially in workforce planning

The cost of attrition and absence
The link between manager action and staff

performance

Middle data
Differences in performance in role

SLIDE 62

62

C R A F T E D B Y C O N C E N T R A

Common statistical traps

When performing analysis, keep in mind that data can often be misrepresentative – think hard to avoid jumping to erroneous conclusions

The Ecological Fallacy An ecological fallacy is a logical fallacy in the interpretation of statistical data where inferences about the nature of individuals are deduced from inference for the group to which those individuals belong. Taking correlation for causation A correlation is just a number. It happens to calculate the strength of a linear relationship between variables, but it does not carry any information about causation. Ignoring statistical significance The greater the standard deviation, the less likely you can draw the conclusion. The smaller the sample size, the greater the danger. Be aware that the world is full of wrong and dangerous analysis and statistics Trap 1 Trap 2 Trap 3

SLIDE 63

63

C R A F T E D B Y C O N C E N T R A

In the 2004 US presidential election, George Bush won the 15 poorest states

Conclusion: Poor people vote Republican and the wealthy vote Democrat.

Bush vote by state in 2004

Trap 1

Gelman, A, (2008) Red State, Blue State, Rich State, Poor State, Princeton University Press

SLIDE 64

64

C R A F T E D B Y C O N C E N T R A

At the individual level, wealth is positively correlated to tendency to vote Republican

Bush vote by state in 2004

Trap 1

Bush vote in 2004 by individual income

The Ecological Fallacy

Gelman, A, (2008) Red State, Blue State, Rich State, Poor State, Princeton University Press

SLIDE 65

65

C R A F T E D B Y C O N C E N T R A

Are genders paid exactly the same in this organisation?

Trap 1

Source: The gender gap and statistical bear traps, Concentra blog http://concentra.co.uk/blog-gender-gap-and-statistical-bear-traps

SLIDE 66

66

C R A F T E D B Y C O N C E N T R A

Within each skill level group women are being paid less than their male equivalents

Trap 1

How is this possible?

Source: The gender gap and statistical bear traps, Concentra blog http://concentra.co.uk/blog-gender-gap-and-statistical-bear-traps

SLIDE 67

67

C R A F T E D B Y C O N C E N T R A

Aggregating data can hide distributions within the data: here, a high % females in the high skill group masks lower pay for females throughout

Simpson's paradox

Trap 1

Source: The gender gap and statistical bear traps, Concentra blog http://concentra.co.uk/blog-gender-gap-and-statistical-bear-traps

SLIDE 68

68

C R A F T E D B Y C O N C E N T R A

Do sharks like ice-cream eaters? It is more likely there is a common cause...

Correlation Causation

# of shark attack

See http://www.tylervigen.com/spurious-correlations for more ‘shark - Ice cream’ like examples

Trap 2

SLIDE 69

69

C R A F T E D B Y C O N C E N T R A

The risk of using bar charts and the power of the box plot

When there is large variation within a dataset, it’s harder to tell a story about what is happening to the whole group

Average performance score Box plots for performance score

Large standard deviation

Trap 3

SLIDE 70

70

C R A F T E D B Y C O N C E N T R A

Tips: Simple lines help to find a story within scatter plots

Source: "Correlation examples2" by DenisBoigelot, original uploader was Imagecreator - Licensed under CC0 via Commons

Performance Pay Performance Pay

SLIDE 71

71

C R A F T E D B Y C O N C E N T R A

Tips: Simple lines help to find a story within scatter plots

Source: "Correlation examples2" by DenisBoigelot, original uploader was Imagecreator - Licensed under CC0 via Commons

Performance Pay Performance Pay

Terminate
Training
Monitor
Promote
Pay rise

Overpaid / action needed Beginners Bargain / risk of loss Stars

SLIDE 72

72

C R A F T E D B Y C O N C E N T R A

What if the business is focused on 'Spans and Layers'?

Overlay business results: the aims everyone agrees on
Catch people doing things right
Explain value of converting managerial overhead to preferred ways to

invest time

Talk about the changes in the ways of working – what to stop – to allow a

reduced span of control and make the business more effective

What enablers are coming this way that will make the business cheaper to

run?

Tips: Analyse spans and layers in context

Mean: 6.3 Median: 5 Average Spans of Control Benchmark: Asia offices

9 8 7 6 5 4 3 2 1

7 7 5 6 10 15

SLIDE 73

73

C R A F T E D B Y C O N C E N T R A

Discuss: What tips do you have?

…

SLIDE 74

Visualisation Exercise

SLIDE 75

Find the best ways to look at the data based on your original hypotheses and the stories you are trying to tell

Visualisation options

Statistical representations Additional analytical elements Plotting a dimension over time Visualising the data early in the process will achieve buy-in and may also start uncovering some basic insights

6. Analyse & share data

SLIDE 76

To generate interesting insights, slice and dice the data and mash-up data

Chart showing heat map of org by years to retirement Scatter chart showing performance vs. engagement coloured by absence days Average current salary by department coloured by gender

How many employees in each function are likely to be here in 5 years? Are more engaged employees better in performance? Are male and female employees are paid equally?

6. Analyse & share data

SLIDE 77

Visualise to analyse

VS.

Sunburst coloured by engagement index Data table

6. Analyse & share data

SLIDE 78

Exercise: Each group is given a set of visuals. Look at the questions below and choose the visual(s) that can best answer each question

Questions: 1. Are there any opportunities to streamline span of control and layer mix? 2. How many high performers (scale 1-10 with 10 being the highest) in certain area (e.g. Birmingham or Head Office) have less than 2 years of tenure? 3. Are there any correlations between large spans of control and performance? And what about Absence days?

1 2 3

6. Analyse & share data

SLIDE 79

79

Modelling organisational systems in multiple-dimensions
Why does it matter?
Practical exercise

HR CIPD Workshop

Analytics for Change

SLIDE 80

…………..But organisational data is not big data

Image credit: Shutterstock

Big data methods are rarely required to answer organisational questions

SLIDE 81

81

C R A F T E D B Y C O N C E N T R A

HR analytics answers questions about the organisational system

Processes & Activities Employees Goals & Objectives Competencies Org structure & Positions Customers Strategy

Required competencies Responsibilities How are customers served? Employee competencies Gap analysis Employees fulfill roles Gap analysis What are my key goals? Headcount reports Is right sized?

?

Questions Links Reports

? ? ?

Percent time Responsibilities

?

Which competencies are core?

SLIDE 82

82

C R A F T E D B Y C O N C E N T R A

Organisations need to know if their people are spending their time effectively and on the right activities, in the right quantities

1. “How do people in the
rganisation spend their time?”
4. “How much time do we waste
n non-value adding activities?”
2. “What are the

most expensive activities?”

3. “How many people

are involved in each activity?

5. “Where does effort in one

business area influence a different business area?”

6. “Is the current activity

mix aligned to business effectiveness?”

Business questions

SLIDE 83

83

C R A F T E D B Y C O N C E N T R A

Understand As-Is Work Analyze As-Is Work Design To-Be Work Link To-Be Work to To-Be Org Structure

Process design involves building and analysing As-Is process map, then designing to-be process and structure

Fixed Process design steps

Build as-is process maps Perform Individual Activity Analysis Perform dimensional analysis Identify the areas of focus Define to-be processes Optimize work and decision making processes Link roles to activities (iterations) Detail to-be positions and

rg chart

1-1 1-2 2-1 2-2 3-1 3-2 4-1 4-2

1 2 3 4

SLIDE 84

84

C R A F T E D B Y C O N C E N T R A

The most important thing when building these taxonomies is to get the level of detail right – it is easy to define too much detail

MANAGE BUY PREPARE SERVE CLEAN MEAT VEG SAUCES RADISH POTATO CARROT WASH SLICE PEEL FRIES FEED PEOPLE Single person canteen Small 10 person restaurant McDonald’s

Example taxonomy tree

Depth 1 Depth 2 Depth 3 Depth 4 Depth 5 Depth 6

SLIDE 85

85

C R A F T E D B Y C O N C E N T R A

IAA, Activity mapping & analytics

Survey time spent, calculate costs and likely impact

Action

Stop the work Outsource Improve productivity Nothing

702K

Marketing

Value Stream 66.4K

Define Marketing Objectives Process

232K

Develop Marketing Strategy

Process

220K

Develop Marketing Plan

Process

183K

Review & Report

Process

15K

Sales targets Activity

3K

Competitive analysis Activity

9.2K

Verify ‘compellingness’ to content CC & CONS Activity

9.2K

Documented

bjectives

Deliverable

30K

Approve

bjectives

Decision

38K

Target audiences – espciallly CC + CONS Activity

19.8K

Prioritize markets regions and segments Activity

11.8K

Develop end user personas Activity

38K

Messaging Activity

30.6K

Competitive positioning Activity

23.4K

Use case development Activity

24.6K

Consumer insight development Activity

15.9K

Customer benefits Deliverable

30.2K

Approve strategy Decision

27.7K

Schedule Activity

20K

Media mix Activity

40.5K

Metrics Deliverable

19.5K

Budget asset list Activity

14K

Partner plan Activity

11K

Digital development Activity

15.5K

PR Activity

14K

Media Communications Plan Deliverable

33.5K

Cost breakdown Deliverable

24.2K

Approve plan Decision

30.2K

Gather Activity

14.3K

Analyze Activity

44.3K

Report Activity

14K

Post mortem Deliverable

15K

Metrics Deliverable Assets Deliverable

36.3K

Change product? Decision

29.3K

Change marketing plan

r process?

Decision

SLIDE 86

86

C R A F T E D B Y C O N C E N T R A

Marketing Define Marketing Objectives

Sales targets Competitive analysis Verify ‘compellingness’ to content CC & CONS Documented objectives Approve objectives

Develop Marketing Strategy

Target audiences – espciallly CC + CONS Prioritize markets regions and segments Develop end user personas Messaging Competitive positioning Use case development Consumer insight development Customer benefits Approve strategy

Develop Marketing Plan

Schedule Media mix Metrics Budget asset list Partner plan Digital development PR Media Communications Plan Cost breakdown Approve plan

Review & Report

Gather Analyze Report Post mortem Metrics Assets Change product? Change marketing plan or process?

Activity mapping & action planning

Map activity changes and their effect on the business

Action

(2) Stop the work (4) Outsource (7) Improve productivity (24) Nothing Develop end user personas Partner plan Digital development PR Metrics Metrics Target audiences – espciallly CC + CONS Prioritize markets regions and segments Report Competitive positioning Use case development Consumer insight development Cost breakdown Schedule Gather Sales targets Media mix Competitive analysis Analyze Verify ‘compellingness’ to content CC & CONS Budget asset list Messaging Approve

bjectives

Change product? Change marketing plan or process? Approve strategy Approve plan Post mortem Documented

bjectives

Assets Customer benefits Media Communications Plan Define Marketing Objectives Develop Marketing Strategy Develop Marketing Plan Review & Report Marketing

SLIDE 87

87

C R A F T E D B Y C O N C E N T R A

The RAD framework

Do away with the complexity from other systems such as RACI, RASIC, RAPID R

Responsible

You are ultimately responsible for ensuring the end result or output of the work is achieved. You understand and manage the delivery and approvals required to be able to deliver the end result.

A

Approve

Veto power on a decision and sign off actions

D

Deliver

Deliver the work, e.g. provide information, analysis and other support to person responsible

Keep it simple Throw out complexity

A* In the RACI definition, A stands for Accountable. I = Informs. Needs to be informed of the outcome of the decision or process C = Consulted. Needs to be consulted prior to an outcome/decision being reached

A* I C

Note: In the Data-Driven Design Book, it was the RAS rather than RAD model. S stood for Supporting the person Responsible in actually doing and delivering the work. This has now been made more explicit with D for Deliver.

SLIDE 88

88

C R A F T E D B Y C O N C E N T R A

Experiment and iterate the design by letting people interact with the data

Create gamified process cards

Creating process cards helps to physically

assign activities to different roles in the

rganisation
Specific activities can be identified for

improvement, outsourcing or can be handed

ver to different areas of the organisation
Map activities to roles and iterate an
rganisation seeing how the organisation

physically takes shape

SLIDE 89

89

C R A F T E D B Y C O N C E N T R A

Now start to design the roles – assign activities to roles based

n who is or will be responsible for doing the relevant work

Roles

SLIDE 90

90

C R A F T E D B Y C O N C E N T R A

Sample questions to test whether each role makes sense

Is it doable, is the role over-loaded?
Would it be motivating?
Could you find someone to do it?
Is it one FTE or multiple?
If multiple, then how would you determine the number of FTEs required?
Is the level of the work consistent?
Would the role have sufficient purpose?
What decision-rights and responsibilities does the role entail?
What are the key process outcomes?
What sort of competencies would be required?

Pick one role and answer whether or not the role makes sense

SLIDE 91

91

C R A F T E D B Y C O N C E N T R A

The Future of HR Analytics

HR CIPD Workshop

Five views
Fundamental principles

SLIDE 92

92

C R A F T E D B Y C O N C E N T R A

View #1. HR Analytics journey will be like playing the piano

Description Correlation Causation Prediction Prescription

Very difficult to establish Causation in organizational data, because of the number

f uncontrolled variables

Prediction may be possible, but

ften you would want intervene –

so you will jump to Prescription, and by intervening, change the behaviour of the system

Hindsight Insight Foresight For our team to be able to do HR Analytics at each end

f the keyboard, from descriptive to predictive, not

seeing it as a series of necessarily sequential steps.

Placid Jover, Unilever CIPD HR Analytics conference

People often speak about gradually building HR analytical capability... In practice, the HR Analytics more resembles playing the piano

“ ”

SLIDE 93

93

C R A F T E D B Y C O N C E N T R A

View #2. The future will be Predictive Analytics

IBM Watson attrition analysis / Visier attrition prediction

Sources: Watson Analytics for HR: Retain your team, IBM Watson Analytics, https://www.youtube.com/watch?v=MUbmmuve1h8 Visier Blog, http://www.visier.com/workforce-intelligence-101/why-hr-needs-data-driven-workforce-planning-to-avoid-talent-shortfalls/

SLIDE 94

94

C R A F T E D B Y C O N C E N T R A

Examples of Predictive Analytics

Recruitment: which profiles of employees will be most effective in their jobs?
Demographics
Channels
Education (UtilityCo)
Psychological profiles (Elkjop)
Retention
% probability of departure in next year, forecast date of departure (Watson, Visier)
Attrition as a function of interventions e.g. New Starter Day (Maersk), training & coaching (Starbucks – after 2 years)
Performance
Engagement scores as fn (last year’s engagement)
Performance scores as fn (engagement, last year’s performance, training completion)
Accident rate as fn (engagement)
Engagement scores as fn (action taken)

SLIDE 95

95

C R A F T E D B Y C O N C E N T R A

View #3. The future of HR analytics will be big / external data

SLIDE 96

96

C R A F T E D B Y C O N C E N T R A

View #4. The future of HR analytics will be Real-time analytics

50-100 questions (wide range of topics)
1 year cycle

4 – 6 weeks designing survey, 2 – 4 weeks getting responses, 3 – 4 weeks analysing results by HR, line managers, execs, Endless action ‘planning’ workshops

Old engagement surveys New engagement tools

Source: The Dinosaurs of Employee Engagement Surveys, Jane Piper, https://www.linkedin.com/pulse/dinosaurs-employee-engagement-surveys-jane-piper?trk=hp-feed-article-title-comment

Fast and agile
Focused and targeted
Responsive

Example: eBay Europe Fast Feedback

SLIDE 97

97

C R A F T E D B Y C O N C E N T R A

View #5. The future will be driving organisational change in complex systems

The customer’s demand pulls through all other elements of the system

Customer Organisation Products & Services Teams Processes & Activities Roles Required Competencies Employees Actual Competencies Delivery Strategy Employees Competencies

SLIDE 98

98

C R A F T E D B Y C O N C E N T R A

Customer Organisation Products & Services Teams Processes & Activities Roles Required Competencies Employees Actual Competencies Delivery Strategy Objectives & Goals Workforce planning Transition Management

#5. HR Analytics needed at each stage so all elements work well together

Bench- marking System Analytics Gap Analysis Selection Right-sizing Activity- Based Cost RACI

SLIDE 99

99

C R A F T E D B Y C O N C E N T R A

1 Start with the business outcomes Align on pain points Build consensus on the priority questions to address Think about ‘credibility and capability’ in your roadmap Showcase learnings and benefits to build momentum and buzz

2 3 4 5

If you remember only one slide from today…

SLIDE 100

100

C R A F T E D B Y C O N C E N T R A

What one thing?

SLIDE 101

101

C R A F T E D B Y C O N C E N T R A

Our commitment to you

We will send you the slides from today’s session
We will send links to all tools and providers shown
We will publish the audience survey results in a blog
We will send you your postcard in 1 month from now
We will invite you to an Open Analytics follow-up in early September

Facilitated opportunity to compare notes with your peers on your analytics journey since this event

SLIDE 102

LONDON

PHILADELPHIA
TORONTO
THE HAGUE

Toronto Philadelphia London The Hague

GET IN TOUCH • GET THE EDGE

+44 (0)20 7099 6910 info@concentra.co.uk concentra.co.uk