HR Analytics Nov 2012.PDF

2009-2012, Smart Wave Education

No part of this presentation shall be copied, reproduced, shared or transmitted without the prior approval of the author

Learn Connect Collaborate Succeed

HR Analytics

Session 1: Introduction

Nurturing leaders of tomorrow

Why are we here?

Your expectations from the program


"Numbers, their use, which ones, and how to capture them have always fascinated me"

-John Buchanan, Coach of Australian Cricket Team

What does 100% strike rate mean?


What does 100% strike rate mean?

100% tells us very little. I wish to know out of those 100 balls, how many does he actually use?

Our intention is not to increase a players' run rate, but the number of balls they scored from

So, we discussed what kind of existing shots the players had that he was comfortable with?

Whether they could use those shots to different balls?We looked at periods of the game or bowling types when

there were more balls which he wasn't scoring?Whether he could develop new shots in their repertoire?


What does 100% strike rate mean?How do we take this example to HR issues?

What does 1% increase in EE mean? What is the duration of impact of an EE

initiative? What does 1st division with 5 year

experience mean? Does the incentive scheme affect sales?

How does EE affect Store Performance ? Factors affecting Store Performance ? Factors affecting EE? Is there a relationship? How many types of data required?


List 3 HR metrics used in your firm

For each functional area


List 3 HR metrics used in your firm

What are the challenges associated with these metrics / implementation?


The new Finance Director, is spending the morning with two of her most senior team.

She pulls down the current management accounts and her expert eye scans through the figures, taking in the pattern to review trends and spot any anomalies.

She knows that her first Board meeting is on Friday and it is important that she provides an authoritative grasp of the companys financial position.

So every now and again she asks for more information and one of her colleagues at the PC clicks a button to drill down into more detailed analysis by region, by product to interrogate the data and highlight specific issues.

After four hours she congratulates her colleagues on their good work, impressed with the speed and efficiency with which theyve worked.

The HR Director knows that on Friday there is a planned Management Review.

Although there is no great pressure he still wants to make a strong impression.

By Monday afternoon, however he is worried. The trouble is there is no data!

There is no financial equivalent of the succession plan. Instead his PA has pulled out a few charts dated from last year. Glancing at it he sees that the structure is hopelessly out of date and that many of the candidates named as potential successors have either left or been moved sideways.

Could we pull down the data for the top 100 execs? he asks. She disappears to return 30 minutes later, holding armfuls of folders.

This is mainly CV stuff, but I do know that the execs went through an individual assessment exercise. The trouble is I dont know where the reports ended up.

Monday - A tale of 2 Directors


Why are we here?

To get familiar with key developments in following areas of HR:

Evidence based HR

How HR can emerge as TALENT DECISION SCIENCE

Enable HR to emerge as a strategic partner


Program Outline Fundamental HR Metrics

Criteria for Metrics selection linking metrics to business goalsAssessment of Metrics based on business alignmentAssessing Metrics based on quality and feasibilityMetrics PrioritizationSelecting : Key Strategic Measures , Key Functional Measures and Impact metrics, HR DashboardsTypes of HR dashboards,Crafting the HR Dashboard for your business environmentImplementing new HR dashboards in your organizationCase studies HR Data Analysis in MS ExcelData Handling, Sampling, Collection and Presentation,Questionnaires, surveys and Interviewsusing graphical and numerical techniques like statistical and probability theory, hypothesis testing, variance analysis, Predictive ModelingUse of Analytical Tools for determination of cause-effect relationship, regression, multiple regressions, cluster analysis, factor analysis,Computer aided Research in HR


New Methods

Big Data

Predictive Analytics

Collaborative BI

Moores Law of Information?


MIT Study: The New Intelligent Enterprise, 2010

Users of Predictive Analytics

Top 5 Industries

Insurance

Healthcare

IT/ITES

Financial Services

Retail

Top 5 Functions

Finance

Strategy

Operations

Marketing

R&D


MIT Study: The New Intelligent Enterprise, 2010

Use of Predictive Analytics


Teams score on average once every 180 possession of the ball; this is near constant for any division of English football

and the World Cup

Two-thirds goal comes from possessions gained in the other half of the field

- Neil Lanham, Soccernomics

Predictive Analytics


HR CHALLENGES

Cost of absenteeism

Cost of employee separations (turnover)

Employee health, wellness, and welfare

Employee attitudes and engagement

Financial effects of work-life programs

Staffing utility

Economic value of job performance

Payoff of enhanced selection

Costs and benefits of HR development programs

Talent-investment analysis

Cascio & Boudreau have developed an excellent and comprehensive list of ways to use analytics to address and improve these talent-related issues

Cascio, Wayne & Boudreau, John, Investing in People: Financial Impact of Human Resource Initiatives




HR Analytics

Session 2: Evidence based HRM


How the role of HR is changing


Forces that are changing the workplace

FROMLocal markets, operations

Manufacturing, clerical work

Hierarchy

Intermediaries; face-to-face

Obedience to formal authority

Stability, efficiency, control

Full time job

Customer service

Work done by employees

Fixed work location

Management prerogative

Loyal service

White, male workforce

Financial performance

TOGlobal markets, operations

Service, knowledge work

Networks

Direct access/virtual relationship

Questioning of formal authority

Change, creativity, flexibility, order

Part-time and project work

Shareholder, stakeholder value

Work done by many contributors

Diverse work locations

Social license

Marketable knowledge, skills

Diverse workforce

Triple bottom line


Welfare Officers

Personnel Officers

HR OfficerBusiness Partner

???

Paternalistic

Caring for injured/sick

Protection of disadvantaged

Manage absenteeism

Training

Productivity

Trade Union

Line Manager

Efficiency & Effectiveness

Compliances

Expert

New Technology

Compliances

How the role of HR is changing

Today, there is a movement to find the critical human levers for improving business results


Some thoughts

To be a true strategic partner HR needs to evolve into a bimodal structure (similar to Finance and Accounting or Marketing and Sales)

- Boudreau and Ramstads in their book Beyond HR (2007)

Talentship:A term coined by Boudreau and Ramstad which envisionsHRM as Talent Decision Science

We do much order-taking and mistaking talk for action

- Jim Moore, Director HR, Sun Microsystems


How does HR become a strategic partner

Focus on Strategy

Strategy may be defined as how the company creates value for its stakeholders (Kaplan and Norton)

Part of any strategy are Goals. Measurement of these goals is through Key Performance Indicators

KPIs must beAligned with the strategyQuantifiable and measurableRecognized through the Firm


Sabermetric

Using Statistical Analysis to find hidden dimensions of success and acquire talent accordingly

Finding the right metric


Which major league baseball team holds the record for the longest consecutive winning streak?

Boston Red Sox

Los Angeles Dodgers

New York Yankees

Oakland Athletics

St. Louis Cardinals

Philadelphia Rangers


WINNING

So, how did the As manage such success with 1/3 the resources of their biggest competitor?

2nd lowest payroll in MLB


MONEYBALL PREMISE

Theory - One could construct a better team using better & different methods of selection

Opportunity - All baseball teams were using outdated methods of player selection

Approach - Rigorous statistical analysis identified undervalued players who manufactured cheaper runs, resulting in more, less expensive wins


If somebody is right 30% of the time using gut feel

and you can find a way to increase it to 35%,

you create 5% winning arbitrage"

- Billy Beane, Oakland A

Analytics


Key Questions for us .

source talent differently?

manage compensation/benefits differently?

prepare for the oncoming baby boomer worker gap?

schedule my workforce differently?

decide which countries are most profitable to operate in?

manage my high potentials differently?


Quality of data is the key to analysis

Causal relationship must fulfill three criteria:

1. Strong relationship exists between the 2 factors2. Cause factor occurs before the effect factor3. Analysis must account for other possible influences

Data must be collected in a timely manner

Evidence based management can be a powerful toolbut only if were clear about what constitutes

valid evidence. Unless we can distinguish hard facts from

questionable data, we may not get very far, no matter how good our intentions may be

- Prof. Rosenzweig


Identify 3 causal relationships

1. ..

2. .

3.


3 relationships which are not causal

1. ..

2. .

3.


How BMW used analytics


People analytics @ Google


Do Managers Matter?

X-Y Plot Performance Review Employee Satisfaction Score

Top v. Bottom Performance QuartileTeam productivityEmployee satisfactionEmployee retention

Great v. Struggling ManagersBehavioral indicators


In order of importance1. Is a great COACH

2. Empowers; does not micromanage

3. Is concerned about team members personal success and well-being

4. Is result oriented

5. Is a Good Communicator

6. Helps with Career Development

7. Has VISION for the Team

8. Is technically good

8 behaviors of Great Managers


In order of importance1. Undergoing a TRANSITION

2. Lack of consistent focus / vision

3. Not concerned with Career Development

4. Spends too little time in Communicating

5. Less focus on 1 to 1 meeting

Behaviours of Struggling Managers


Which staff category has maximum impact on customer experience at Disneyland?

Beyond HR: the New Science of Human Capital John Boudreau & Peter Ramstad


Disneyland : Mickey or Sweepers?

Sweepers at Disney make a huge difference to the customer experience

While they keep its parks clean, they interact with customers frequently. helping them navigate through the park pointing them to catering and merchandisingcoping with the unexpected problems

Sweepers are front line customer representatives with brooms in their hands.

Investment in the recruitment, induction and training of its sweepers have more business impact than directing resource to its costumed characters

Beyond HR: the New Science of Human Capital John Boudreau & Peter Ramstad


AmEx

Measurement in central to HR Activities1. All measurements are metric focussed

2. PeopleSoft as the backbone for data quality and consistency

3. Rolled out across 83 countries


CapOne and Nextel


Cost of replacing an employee

1. ..

2. .

3.


Cost of replacing an employee




HR Analytics

Selecting HR Metrics


What gets measured, gets done

Attrition Analytics @ Convergys

We used consumer marketing technique, conjoint analysis. When used with employees, it helped figure out the mix of benefits most likely to encourage people to staywith the organization.

Using this tool, we learned, that employees weremore likely to stay with the company if they got half their annual pay raises every six months instead of the entireamount only once per year.


Key Questions from HR PerspectiveWHY measure?

WHAT to measure?

HOW to be measured?

How frequently to be measured ?

What gets measured, gets done


What gets measured, gets doneThe challenges lie in being first able to measure it effectively especially across the multiple service lines, for example, a high-end consultants productivity may need to be measured through a combination of billing rate and utilization, while a developer may need to be measured using the effort needed to code either a certain number of lines or a function point.

- Nandita Gujar, SVP, Infosys

If you put it all under one larger metric, it can camouflage some of the more critical issues

- Prabir Jha, SVP, Tata Motors

At Fortis, we track four essential areas: customer satisfactiondeveloping people

developing processes financial impact


Metrics to measure employee productivity

At Infosys, employee utilization is measured as person hours deployed on billable engagement

person hours available for billable engagement

Productivity is measured as- revenue per person - operating margin per person


Metrics that predict organizational performance

Some metrics consistently predict organizational performance

leadership

employee engagement

access to knowledge

workforce optimization

organizational learning capacity

and some metrics do not

turnover rate

time to fill

total hours training

Bassi and McMurrer (2007)


Efficiency Metric v. Effectiveness Metric

Efficiency =

Effectiveness =


Teams measure goalkeepers on what % of goals they stopped, This favored gatekeepers from big teams

because of tighter defence;

However, analysis of % of goals stopped from within the penalty area gives entirely different result

- Daniele Tognaccini, AC Milan

Analytics


Leading Metric v. Lagging Metric

Leading Metric =

Lagging Metric =


Leading Metric v. Lagging Metric

Turnover

Cost to hire / time to hire

Absenteeism

% of PRs completed

% of DPs in place

Disciplinary activity

% internal movement :

external hiring

Revenue/employee

% of new hire retention

Avg tenure : average PR measure

Avg L&D spend/employee : avg PR

Rating

% of succession plans in place

against critical roles

% of ready nows for critical roles

Engagement


Valero - Talent Supply Chain

Grew from 2,000 (Yr. 2000) to 22,000 (2007) employees and Revenue - $118 million (2000); $75 billion (2007)

The challenge recruiting operations were entirely paper-based start-up of a new refining complex in USA Outsourced engineers in Canada Programmers in India Manufacturing workers in China

Launches world-class staffing program in 2002

In 2002, it took us 41 pieces of paper to hire, more than 120 days to fill an open

position and cost us $12000 per hire


Valero Recruitment Metrics and Dashboard

Talent needs communicated to the best sources Started maintaining suppliers record

efficiency (cost, time) effectiveness (quality of hire, reliability)

Continuously monitors adverts placed on Internet Portals (one source) to assess the cost, speed, quality and dependability of the supplier

Captures data on the quality of talent to rate Suppliers job performance, Fitment with organization and its culture Retention

Dashboard shows Green = suppliers performance is good Yellow = Okay Red = Problems identified

In 2007, it is all automates, takes 41 days to fill a position and cost is $2300 per hire


Efficiency Metric v. Effectiveness Metric

Efficiency and Effectiveness Metric for

Recruitment

Training

Retention

.


Human Capital Metric

HC ROI = Human capital return on investment measures the return on each dollar invested in human capital after adjusting for the cost of financial capital = (Profit Financial Capital Costs) Human Capital Costs

HC Productivity = Amount of revenue generated for each dollar invested in human capital, after adjusting for the costs of materials and financial capital = (Revenue Material Costs) (Human Capital Costs + Financial Capital Costs)


The status

Gates study of 104 HC Execs71% have some form of HC measures

Only 7% have metrics capturing information on efficiency, effectiveness, and impact (on business process and strategic outcomes)

Luketic survey of 307 HR Professionals79% using some form of HC metrics

Only 11% are strategically focused

Stephen Gates, Strategic Human Capital Measures, The Conference Board (2008)

Helen Luketic, Human Capital Metrics Trends 2008 Survey Report, ResearchVOICE (2008)


A US manufacturer of cars and trucks

400,000+ employees

Sales $200B +

Profit $ 5B+

Situation

To quantify HR support towards business objectives Identify metrics clearly aligned with business objectives Metrics to provide the highest quality information

Metrics should be feasible to collect

Metrics that have maximum impact on the business

Case of a global automobile maker

As the HR function transitioned from transactional to strategic, it was critical

to improve the process of identifying and prioritizing metrics.

VP of HR


Business Objectives of the firm

Focus on Brand

Grow Market Share

Increase Profits

Business Objectives HR Objectives


HR Objectives of the firm

Focus on Brand

Grow Market Share

Increase Profits

Top Company to Work Leverage global network to provide

unmatched career opportunity Hire Right Person at the first

instance

Appreciate Diversity and Culture Increase People Return on

Investment

Increase PROI Build firms capability for higher

targets, goals, competition To achieve targets:

Ensure people have right skills Increase people motivation

Business Objectives HR Objectives


What CRITERIA should be used

for selecting

the

METRICS


Screen the Metrics for

Alignment Quality Feasibility

Ch

arac

teri

stic

s


Screen the Metrics for

Alignment Quality Feasibility

Ch

arac

teri

stic

s Linked to achievement of Business Objectives

Measures how well is the firm using human capital

Depicts quality of HR performance in supporting business objectives

The Red Flags, to show decline in performance

Ease of data collection

Resources required in implementation


Step 2: Identify potential metrics to support business Objectives

Metrics like level of attrition, recruitment mix, and bench strength have a direct bearing on cost and organizational performance.

Other key focus areas include optimal resource utilization, and impact on top line and bottom line.

Finally, continuous competency enhancement including leadership pipeline affects delivery excellence and hence, has a positive effect on organizational performance.

These are all people metrics and we look very closely at all of them.

-Dr. Ritu Anand, VP, Global HR, TCS

Internal Sources Input from Line Managers Input from Line-Based HRPrevious HR Metrics

External Sources Best Practices Case studies


Step 3: Select Metrics based on Alignment

Direct Metrics Supporting Metrics


Step 3: Select Metrics based on Alignment

Direct Metrics Supporting Metrics

These indicate actual performance which impacts the achievement of business objective.

Example: Employee productivity (vehicles per employee) has a direct impact on the ability to achieve the objective Build firms capability for higher targets, goals, competition

Positive performance on these indicators contributes favorably to the achievement of business objective.

Example:Competitive compensation package contributes to, but does not directly impact the ability to achieve the objective Having the Right People at the Right Time.

Our approach is to categorize talent and focus on retaining the essential

- Intercontinental Hotels


Step 4: Select Metrics based on Quality

Better quality means greater business impactQualityCategory

Category Description Ratings

Subjective Metric cannot be communicated as a number. Subjective terms are used to indicate performance (e.g., employee morale).

0

Quantifiable Metric can be communicated as a number. A numeric goal can be established, or trend goal visually apparent (e.g., hours of training).

1

Results Positive performance will directly impact company financial performance (e.g., gap between current and required workforce skills).

1

Qualitative Metric reflects the quality of performance (e.g., utilization of training). 3

Predictive Metric will raise the red flag when performance begins to decline, allowing management an opportunity to put in place immediate countermeasures designed to reverse poor performance. Metric is conducive to problem solving and process improvement. Best-practice performance can be isolated, globalization of practice implemented, and financial impact forecasted e.g., strength of employment brand

6


Step 5: Ease of Implementation

QualityCategory

Category Description Ratings

Easy to Measure

Metric exists already or data to support metric iscurrently available (low cost and low resource utilization).

1

Moderately Difficultto Measure

Metric does not currently exist. Data may currently existbut is not readily available (moderate cost and resourceutilization).

-1

Difficult to Measure

Metric does not currently exist. Data may not currentlyexist (cost and resource intensive).

-2


Step 6: Prioritize MetricsMetric Hire Right

Person at

the first

instance

Ensure

people

have right

skills and

people

motivation

Build

firms capability

for higher

targets

Top

Company

to Work

Increase

PROI

Appreciate

Diversity

Quality

Score

Implement

ation

Score

Overall

Score

HR:FTE Ratio

Competitivenes

s of

Compensation

Package

New Hire

Turnover

Turnover by

Employee

Segment

Gap Between

Current and

Required

Workforce Skills

Employee

Productivity

(Vehicles/Empl

oyee)

Inter-region

executive

transfer


With the recruitment of Carlos Tavez, David Silva, Adam Johnson and Yaya Toure our pass rate in the final third has

gone up by 7.7%"

- Gavin Fleig, Head of Performance Analytics, Manchester City

Talent Metric


Step 7: Select high Impact Metrics

After individually evaluating potential metrics, the HR team meets as a group to select the metrics they believe will have the highest impact on HRs contribution to business objectives.

The team discusses the quality and implementation ratings assigned to the metrics by each team member and ultimately comes to consensus on the final rating for each metric.

Finally, the team selects the two or three metrics with the highest ratings for each objective


Step 8: Final HR Dashboard

Focus on Highest-Value Metrics: Dashboard includes only the highest-value metrics based on the HR planning teams assessment of alignment, quality, and ease of implementation

Concise Collection of Measures: Each objective is mapped to no more than two to three metrics

Organized by Business Objective: The dashboard is organized by business objective to clearly indicate HR performance in supporting these objectives


Step 9: Select high Impact Metrics

Parameter Measure

Focus on Quality Organizational Excellence

HR Accelerated Leadership Program

No of workshops

Employee Engagement Plan Conformance to Plan

Global functional alignment Linkage Matrix

Transform HR Service Performance to Plan

Grow Mkt. Share Critical Capabilities

PMS Implementation % Population

Increase Profit

HR Budget Performance $M

People Cost $M

Perf. To Cadence Hire Plan Performance to Plan


Metric Assignment: Strategy of firm




HR Analytics

Recap


I. Create your Firms Business Plan-Define your business objectives

II. Map your HR objectives which will support this business plan

III. Identify potential Metrics that can be used to support each HR objective

IV. Score each potential metric in terms of Quality and Feasibility

V. Hold a consensus meeting to agree on a final score for each metric

VI. select highest impact Metrics-One or two for each metric

VII. Ask for HR feedback

VIII. Freeze the Final HR Dashboard

THE PRIORITIZATION PROCESS


Analytics at Sysco thru' ESS Three gross measures

work climate and employee satisfaction

Productivity,

Retention

Across seven dimensions such as supervisors effectiveness, diversity, quality of life etc.

Results: Units with highly satisfied employees have higher revs, lower

costs, better retention and customer loyalty

Identified management actions with greatest impact

Tracks each units ESS

Increased retention rates of delivery agents from 65% to 85%

$50M savings in hiring and training costs

Davenport, Harris, Shapiro, Competing on Talent Analytics, Harvard Business Review, 2010


What area might offer the greatest financial benefit for your company if addressed?

Improving quality of hires

Reducing absenteeism or turnover

Improving employee engagement

Increasing effectiveness in job

Optimized labor allocation/scheduling


THE METHODOLOGY

Four steps to help you leverage analytics

1. Identify the right things to measure (i.e. KPIs)

2. Select the best analytic method to apply to your project

3. Effectively communicate the results

4. Build a follow-through strategy

Bassi, Laurie HR Analytics Handbook, 2010


Analytics Value Chain

Opinion Data Metrics Analysis Insights Actions

A Gut Feel

I think so

Based on experience

Basically Raw

Difficult to make sense

Could be structured

Ratios

Trends

Too many metrics become confusing

Identify relationships

trends

Influences decision

Leads to action


Background Fortune Best Company Strong culture, low turnover by industry comparison Experienced Senior Management Team

Challenge Large training investment 240 hours in year 1 alone Discovered that majority of WF was leaving within 1-2 years

Question Can we do a better job of identifying those who will stay

versus those who will leave?

Case Study


Collected 20 years of employment history data (26,000 employees & 50+ attributes per employee)

Employment duration made known for half of employees(training the model)

Model identified variables best at predicting outcome(5 attributes retained from 50+)

Outcome was hidden when model applied to other half (testing the model 85% accurate)

Model applied to current employees, giving them a score(applying the model)

STEPS INVOLVED




HR Analytics

Session 5 & 6 : Data Analysis


Truths, half truths and statistics

Raw data is:

Voluminous

Difficult to handle

Decision makers hate it, hence, often prefer to take decisions without data


First 100 weeks of

15 148 142 10 14359 6 96 0 1398 17 6 42 8

41 114 43 7 2335 5 11 18 1557 0 34 2 1480 11 85 61 13

24 111 179 36 488 1 54 15 1555 6 40 4 79

10 0 10 169 17727 73 4 9 3168 50 0 35 34

119 9 52 9 721 165 2 7 4711 78 24 15 11316 62 122 88 677 28 31 92 0

15 104 177 4 13640 71 74 83 6


How would SHE be doing in the next 10 weeks

15 148 142 10 14359 6 96 0 1398 17 6 42 8

41 114 43 7 2335 5 11 18 1557 0 34 2 1480 11 85 61 13

24 111 179 36 488 1 54 15 1555 6 40 4 79

10 0 10 169 17727 73 4 9 3168 50 0 35 34

119 9 52 9 721 165 2 7 4711 78 24 15 11316 62 122 88 677 28 31 92 0

15 104 177 4 13640 71 74 83 6


5 number summary

Min

Q1

Median

Q3

Max


Outliers : Unusual observations

It affects analysisNeeds to be IDENTIFIED

Analyzed separately

May not necessarily be BAD data

How to identify OUTLIERS?


How to identify OUTLIERS

IQR Test, IQR (inter-quartile range) = Q1 Q3

If data is Q3 +1.5 IQR

Q1 - 1.5 IQR

How to deal the OUTLIERS?


2 Critical Aspects

Central tendency

Spread

How to does OUTLIERS affect centrality and spread?


CTC of TLs

Given below is a sample of monthly CTC (Rs.K) for 70 Team Leaders in BPOs based in Gurgaon.

The data are presented in ascending order.

425 430 430 435 435 435 435 435 440 440

440 440 440 445 445 445 445 445 450 450

450 450 450 450 450 460 460 460 465 465

465 470 470 472 475 475 475 480 480 480

480 485 490 490 490 500 500 500 500 510

510 515 525 525 525 535 549 550 570 570

575 575 580 590 600 600 600 600 615 615


Mean

n

xx

i

N

xi

The mean of a data set is the average of all the data values.

If the data are from a sample, the mean is denoted by x-bar

.

If the data are from a population, the mean is denoted by mu


80.49070

356,34

n

xx

i

425 430 430 435 435 435 435 435 440 440

440 440 440 445 445 445 445 445 450 450

450 450 450 450 450 460 460 460 465 465

465 470 470 472 475 475 475 480 480 480

480 485 490 490 490 500 500 500 500 510

510 515 525 525 525 535 549 550 570 570

575 575 580 590 600 600 600 600 615 615

CTC of TLs


Median

The median is the measure of location most often reported for annual income and property value data

A few extremely large incomes or property values can inflate the mean


The median of a data set is the value in the middle when the data items are arranged in ascending order.

For an odd number of observations, the median is the middle value.

For an even number of observations, the median is the average of the two middle values.

Median


Median = 50th percentile

i = (p/100)n = (50/100)70 = 35.5

Averaging the 35th and 36th data values:

Median = (475 + 475)/2 = 475

CTC of TLs

425 430 430 435 435 435 435 435 440 440

440 440 440 445 445 445 445 445 450 450

450 450 450 450 450 460 460 460 465 465

465 470 470 472 475 475 475 480 480 480480 485 490 490 490 500 500 500 500 510

510 515 525 525 525 535 549 550 570 570

575 575 580 590 600 600 600 600 615 615


Mode is the most frequently occurring data

The greatest frequency can occur at two or more different values

If the data have exactly two modes, the data are bimodal

If the data have more than two modes, the data are multimodal

Mode


450 occurred most frequently (7 times)

Mode = 450

CTC of TLs

425 430 430 435 435 435 435 435 440 440

440 440 440 445 445 445 445 445 450 450450 450 450 450 450 460 460 460 465 465465 470 470 472 475 475 475 480 480 480

480 485 490 490 490 500 500 500 500 510

510 515 525 525 525 535 549 550 570 570

575 575 580 590 600 600 600 600 615 615


Percentiles

A percentile provides information about how the data are spread over the interval from the smallest value to the largest value.

Admission test scores for colleges and universities are frequently reported in terms of percentiles.


Percentiles

COMPA RATIO??


The p-th percentile of a data set is a value such that at least ppercent of the items take on this value or less and at least (100 - p) percent of the items take on this value or more. Arrange the data in ascending order.

Compute index i, the position of the pth percentile.

i = (p/100)n

If i is not an integer, round up. The pth percentile is the value in the ith position.

If i is an integer, the pth percentile is the average of the values in positions i and i+1.

Percentiles


90th Percentile

i = (p/100)n = (90/100)70 = 63

Averaging the 63rd and 64th data values:90th Percentile = (580 + 590)/2 = 585

425 430 430 435 435 435 435 435 440 440

440 440 440 445 445 445 445 445 450 450

450 450 450 450 450 460 460 460 465 465

465 470 470 472 475 475 475 480 480 480

480 485 490 490 490 500 500 500 500 510

510 515 525 525 525 535 549 550 570 570

575 575 580 590 600 600 600 600 615 615

CTC of TLs


Compa Ratio

= actual pay / midpoint of the grade ( or mkt)

Range penetration

= (Salary Range Min)/ (Range Max-Range Min)

Group Compa Ratio??

Compa Ratio & Range Penetration


Quartiles

Quartiles are specific percentilesFirst Quartile = 25th Percentile

Second Quartile = 50th Percentile = Median

Third Quartile = 75th Percentile


Third Quartile

Third quartile = 75th percentile

i = (p/100)n = (75/100)70 = 52.5 = 53

Third quartile = 525

425 430 430 435 435 435 435 435 440 440

440 440 440 445 445 445 445 445 450 450

450 450 450 450 450 460 460 460 465 465

465 470 470 472 475 475 475 480 480 480

480 485 490 490 490 500 500 500 500 510

510 515 525 525 525 535 549 550 570 570

575 575 580 590 600 600 600 600 615 615

CTC of TLs


Frequency Distribution

Used for large data set

Divide large data into smaller groups / classes with number of occurrence for each class

Class Frequency Percentage

425-449 18 25.7%

450-474 16 22.9%

475-499 11 15.7%

500-524 7 10.0%

525-549 5 7.1%

550-574 3 4.3%

575-599 4 5.7%

600-625 6 8.6%

70


Frequency Distribution

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

400 450 500 550 600 650

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

400 450 500 550 600 650

STURGEs RULE for Number of Classes: 2k-3


Measures of Variability

It is often desirable to consider measures of variability (dispersion), as well as measures of location.

For example, in choosing Recruitment Consultant A or B we might consider not only the average delivery time for each, but also the variability in delivery time for each.


Range

Interquartile Range

Variance

Standard Deviation

Coefficient of Variation

Measures of Variability

Range

The range of a data set is the difference between the largest and smallest data values.

It is the simplest measure of variability.

It is very sensitive to the smallest and largest data values.


The range of a data set is the difference between the largest and smallest data values

It is the simplest measure of variability

It is very sensitive to the smallest and largest data values

Range


Range = largest value - smallest value

Range = 615 - 425 = 190

425 430 430 435 435 435 435 435 440 440440 440 440 445 445 445 445 445 450 450

450 450 450 450 450 460 460 460 465 465

465 470 470 472 475 475 475 480 480 480

480 485 490 490 490 500 500 500 500 510

510 515 525 525 525 535 549 550 570 570

575 575 580 590 600 600 600 600 615 615

CTC of TLs


The inter-quartile range of a data set is the difference between the third quartile and the first quartile

It is the range for the middle 50% of the data

It overcomes the sensitivity to extreme data values

IQR = Inter Quartile Range


Inter-quartile Range3rd Quartile (Q3) = 525

1st Quartile (Q1) = 445

Inter-quartile Range = Q3 - Q1 = 525 - 445 = 80

425 430 430 435 435 435 435 435 440 440440 440 440 445 445 445 445 445 450 450450 450 450 450 450 460 460 460 465 465

465 470 470 472 475 475 475 480 480 480

480 485 490 490 490 500 500 500 500 510

510 515 525 525 525 535 549 550 570 570575 575 580 590 600 600 600 600 615 615

CTC of TLs


Variance

The variance is a measure of variability that utilizes all the data points

It is based on the difference between the value of each observation (xi) and the mean (x for a sample, for a population).


Variance

The variance is the average of the squared differencesbetween each data value and the mean.

If the data set is a sample, the variance is denoted by s2.

If the data set is a population, the variance is denoted by 2.

sxi x

n

22

1

( )

22( )x

Ni


Standard Deviation

The standard deviation of a data set is the positive square root of the variance.

It is measured in the same units as the data, making it more easily comparable, than the variance, to the mean.

If the data set is a sample, the standard deviation is denoted s.

If the data set is a population, the standard deviation is denoted (sigma).

s s2

2



The coefficient of variation indicates how large the standard deviation is in relation to the mean.

If the data set is a sample, the coefficient of variation is computed as follows:

If the data set is a population, the coefficient of variation is computed as follows:

s

x( )100

( )100


Variance

Standard Deviation


sxi x

n

22

12 996 16

( ), .

s s2 2996 47 54 74. .

s

x100

54 74

490 80100 1115

.

..

CTC of TL


Mean, Median and Mode

ApplicationCompensation Benchmarking

Time to fill

Training days

Attrition

..

.

119


2-Sigma

120


4-Sigma

121


6 Sigma

122


Impact of varying mean and standard deviation

123


124

Market CTC Data

Level Median 75P

7 8441 9921

6 4811 5367

5 2699 3079

4 1817 2032

3 1249 1445

2 841 885

1 479 523


125

Interpret this for Your Company

Level Median 75P YourCo

7 8441 9921 11000

6 4811 5367 3800

5 2699 3079 3576

4 1817 2032 1900

3 1249 1445 1440

2 841 885 980

1 479 523 471


126

Interpret this for YourCo

0

2000

4000

6000

8000

10000

12000

1 2 3 4 5 6 7

Median 75P YourCo


127

Right visual matters for right decision

0

2000

4000

6000

8000

10000

12000

1 2 3 4 5 6 7

Median 75P YourCo


Employee Engagement Surveys

Metrics Unit Top Box Score Ratio of Engaged to Actively Disengaged


Employee Engagement Surveys


Plotting the Bell Curve

130

Ratings% of Rated Employees

1 2 3 4 5

Your Co 0% 3% 50% 42% 5%

Best Practice 10% 20% 40% 20% 10%

Well Performing Co 5% 15% 40% 25% 15%


Plotting the Bell Curve

131

-10%

0%

10%

20%

30%

40%

50%

60%

0 1 2 3 4 5 6

Series1 Series2 Series3


Time to Fill

132

Position Time to fill (in days)1 342 453 924 555 126 247 358 479 52

10 2111 7812 4513 6714 11015 2816 3117 6218 4019 2620 8521 1522 5323 4424 3525 3926 4927 5828 3129 3730 4031 7132 1933 6234 2235 5636 2637 3338 3839 4140 80


Regression Analysis

1. Statically identify historical predictor of workforce size

Example: FTEs = a + b1 sales + b2 new customers

2. Only use equations with predictors found to be

statistically significant

3. Predict future HR requirements, using equation

Example: (a) FTEs = 7 + .0004 sales + .02 new

customers

(b) Projected sales = 1,000,000

Projected new customers = 300

(c) HR requirements = 7 + 400 + 6 = 413


Measuring relationships

Does sales increase with number of sales force?

How much waiters are required if seats increase?


Predictive Analysis

135


Hotel rooms and attendantsRooms Attendants

200 240150 210150 150120 140125 150220 225180 200240 240250 255180 230280 260300 260305 275325 275350 420400 470425 450440 500450 475500 450500 520520 490550 525600 620620 600660 640700 660800 900


1st Analysis

Rooms Attendants R/A

Total Population 10540 10830 1.03

Average 376 386 1.05

St Dev 191.4 191.8 0.13

0

100

200

300

400

500

600

700

800

900

1000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

Number of Attendants


Another Graphical Representation

0

100

200

300

400

500

600

700

800

900

1000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28


Hotel Staffing

Dependent and Independent variables ???


Size of Hotel and Attendants - Relationship

0

100

200

300

400

500

600

700

800

900

1000

0 200 400 600 800 1000



A = 0.979*R + 17.91R = 0.956

0

100

200

300

400

500

600

700

800

900

1000

0 200 400 600 800 1000



0

100

200

300

400

500

600

700

800

900

1000

0 100 200 300 400 500 600 700 800 900


Analyze relevant cluster

300

350

400

450

500

550

600

650

700

300 350 400 450 500 550 600 650 700 750


Analyze relevant cluster

A = 0.704*R + 160.4R = 0.876

300

350

400

450

500

550

600

650

700

300 350 400 450 500 550 600 650 700 750


t-statistic

146

Likelihood of = 0


Regression Standard Error

147

How much are the points scattered?


Manpower Planning using Monte Carlo Simulation

Case Study of a Loan Processing Unit of a Bank


Manpower Planning at Starbucks


Manpower Planning at Starbucks

No of Staff No of Days

5 4

6 10

7 16

8 50

9 62

10 38

11 12

12 8

Documents

HR Analytics Nov 2012.PDF