09 4 15 data driven hr impacts of talent management bus case slideshare

Don’t Be Left Behind… As HR

Shifts To A Data-Driven, High

Business Impact Approach

© Dr John Sullivan

43www.drjohnsullivan.com

• The most dramatic shift in HR in decades is occurring

right now… and it will soon make your current HR

approach obsolete.

• You can welcome it or fear it, but this shift is inevitable

3

I’m from the Silicon Valley…

They asked me to highlight the most advanced

data-driven talent practices

• Obviously you can’t adopt all of them

• So pick & choose and adapt whatever practices

that you find are best for your organization

4

A definition of

data-driven HR (ddHR)

5

A definition of data-driven HR (ddHR)

A strategic proactive approach to talent mgmt.

It replaces the currently common… opinions,

beliefs, intuition, hunches, speculation, trial & error

learning & automatically continuing with past practices.

It instead relies on… data, facts, analysis, charts,

metrics, algorithms, science and predicted trends…

to improve the accuracy, speed and business

impacts of major people management decisions…

that are made by managers and HR professionals.

And finally, it reveals “why” programs work.

6

2 quick examples of:

How data-driven HR can

improve retention and

recruiting

Traditional HR would guess / speculate for retention purposes… when do new hires quit?

Source: entelo.com using 1 million resumesWaiting period

ddHR can prove what causes these “turnover spikes”

Employees that quit

Years at the firm

8

During what month do most salespeople quit?

I

Source: Entelo 2015

Turnover seems even… until you add December

9

Do you see how using data

can change you from “a guesser” to an expert?

10

Reasons why HR must shift

to a new data-driven model

11

Almost everyone agrees that HR…must become more strategic

When CEO’s and board-level executives rank

business functions… which one is listed as the most

strategic?

Sales

Where was HR ranked on the list?

“the least strategic function” Source: DDI

11

12

Almost everyone agrees that HR…must increase its business impacts

Of the 18 business factors that contributed the most

to business outcomes…

Reducing operational cost structures was ranked #1

“Talent was dead last” (#18)

(source: KPMG / HfS research)12

13

Almost everyone agrees that HR… is not very good at analytics

% of advanced users % of non-users

1.Finance 58% 7%

2.Executive team 51% 11%

3.Operations 48% 9%

4.R&D 44% 23%

5.Marketing 41% 16%

6.Sales 34% 20%

7.HR (last) 27% 23%

Overall – only 26% of organizations said they had

the ability to meet their analytics needs Source: AMA/i4cp 2013

14

CEO’s do not have faith in our metrics

Only 12% of CEO’s are confident on the quality

of Human Capital metrics

14AICPA survey 2012

15

Do you argue we should keep the black phone?

It’s time to drop the black phone approach to HR

Std. excuse… “our HR model has served for years &

it seems to us that it works fine just the way it is”

Answer… the old ways must be abandoned

because HR’s capabilities have not kept pace with…

the higher expectations of executives

and the expanded capabilities of technology and

big data 15

16

Additional reasons why HR must adapt to a changing world

1.Every other business function has long ago shifted

and managers have learned to expect/ demand data

2.A volatile world history doesn’t repeat itself very

often anymore (HR has 100% historical metrics)

3.The speed of change requires fast but accurate

decision making (VUCA = volatility, uncertainty, complexity and ambiguity)

4.The mobile phone makes data instantly accessible

5.You can’t fix / improve programs if you don’t

know “why” they work and fail

17

Moving from an art to a science

Years of assuming that

“what we have always done” was right…

may be proven completely wrong with data

Why we must shift

This Google data… shows they were wrong about what predicts new hire success

“GPA’s

“Test scores

“Brainteasers

Interviews – “many managers, recruiters, and HR staffers think they have a special ability to sniff

out talent. They’re wrong” “it’s a complete random mess”… “we found a

zero relationship” (between interview scores and on-the-job performance)

No value is added “after 4 interviews”

College –“the proportion of people at Google without any college education… has increased over time”

What predicts? – “capability & learning ability”18

are worthless as a criteria for hiring”

are worthless”are a complete waste of time”

Laszlo Bock, Senior VP of people operations at Google The New York Times

19

Let’s look at a quick case study

A snapshot of

how Google does ddHR

20

Learn from Google

Laszlo Bock VP at Google

21

Google is the benchmark data-driven firm to follow

“All people decisions are based on data & analytics”

“We apply the same level of rigor, analysis and

experimentation on people as we do the tech side”

“At Google, almost every benefit is broken down

into crunchable, poll-able or graphicable data”

22

ddHR goal # 1Data-driven experimentation

Google has an R&D team in Talent Management

People & Innovation Lab (PiLab) - it runs dozens

of experiments on employees in an effort to

answer questions about the best way to manage

a large firm (on-boarding, diversity, pay,

retirement savings & employee weight)

23

ddHR goal # 1An example of a data-driven experiment

1 email causes a 25% improvement in productivity

Each new hire’s manager is sent a JIT on-boarding email reminding them to do these 5 things

Have a discussion on their role and responsibilities

Match your new hire with a peer buddy

Help your new hire build a social network

Set up onboarding check-ins once a month for

their first six months on the job

Encourage an open dialogue

Result – time to minimum productive is 25% faster

24

ddHR goal # 2Identify the highest impact actions

Project Oxygen showed actions make good managers

1. Be a good coach – have regular one on ones, and provide constructive feedback

2. Empower your team and don’t micromanage –give employees freedom and provide them with stretch assignments

3. Express interest in team members success and personal well-being – get to know employees as people, with outside lives

4. Don’t be a sissy: be productive and results-oriented – help the team prioritize work and remove roadblocks

26

Examples covering each of the

remaining ddHR goals

27

Replace them with…

I know

The data proves that…

This chart / graph proves

Our algorithm shows it works

A predictive trendline shows

An A/B test shows B is better

It’s a science

ddHR goal # 4 Use the language of business ($ and #’s)

Outdated phrases

I think/I believe/ I feel

My opinion is…

My gut tells me

We tried that once…

Historical metrics show

We always do it this way

It’s an art

ddHR goal # 5 ID the functions with the highest business impact

28Source: BCG/WFPMA - From Capability to Profitability: Realizing the Value of People Management, 2012

Which TM functions have the highest impact on profit?

Mid and lower impact HR functions

29

30

ddHR goal # 6 HR quantifies its business impacts in $

A revised data-driven recruiting program was

implemented that focused exclusively on

salespeople

After acclimation, newly hired salespeople now

sell nearly 15% more each month (+ $15,100)

With over 100 hires per year, the CFO

calculated that the revenue impact exceeded

$1.5 million

Which was nearly double the overall recruiting

budget

What happened here?

Hockey stick approach

31

Turnover %

TimeOctober 2014

Steady turnover of 9 %

Personalized retention plans were implemented

4 % turnover

ddHR goal # 7Use charts to prove that things work

32

ddHR goal # 8Use data to influence (vs relationships)

Relationships are slow to build and are quickly

lost when one party leaves… so instead use data to

influence managers

“The best thing about using data to influence

managers… is that it’s hard for them to contest

it”

“For most… just knowing that information…

causes them to change their conduct”

Source: Laszlo Bock

33Source: Workday Insights Retention Analytics

ddHR goal # 9 - influence managers with databy distributing ranked reports (retention flight risk)

34

This is what data-driven retention looks like (Workday)

Source: Workday Insights Retention Analytics

Software makes decision-making easy

ddHR goal # 10Improve decision-making speed (at GSK)

Open-office design improves decision-making speed

by 45% and cuts the cost of delays / “doing nothing”

35Source: Forbes 2012

36

ddHR goal # 11Calculate the added value of top performers

Prioritize your top performers

The top 1% of your workforce produces what %

of your total output ?

The top 5% produce

The bottom 5%...

A top performer produces how much more than

the average employee in the same job?

- 10 times the average

- 25 times more than average employee

- 300 times more than the average

- 1000 times more than the average

GE, Netflix & Yahoo

5%

Apple

Google

Microsoft

26% (5X) - U of Indiana study

may actually lose you $

37

ddHR goal # 12Calculate the cost of weak performers

A weak employee may cause errors and disruption

each year up to 2 ¼ times their annual salary

(O’Boyle and Aguinis)

They take up a manager’s time because their

managers must spend nearly one day a week

(17%) dealing with them (Source: Robert Half)

Toxic employees make their teammates 54% more

likely to quit Source: Cornerstone Selection survey

Bad ones stay forever… weak hires may stay 20

years, multiplying their negative impact

38

Conclusion – best practice sharing increases revenue by $250,000

per year, at a cost of $50,000, with an ROI of 4 to 1

Monthly sales

Best

practice

sharing

in sales

started

Sales up

+ 22.4% after 1 yr.

of B. Practice sharing

1st Quarter

$1.5 mil

$1.4 mil

$1.3 mil

$1.2 mil

$1.0 mil

.02% change in sales in control group

2nd Quarter 3rd Quarter 4th Quarter

Control group

ddHR goal # 13Use split samples to show programs work

39

ddHR goal # 14 Integrated data allows you to “connect the dots”

Don’t miss the costs of unintended consequences

Pocket #1 – Froze safety training (Saved $50,000)

But look at “other pocket” costs after one year…

Pocket #2 – Accident rates doubled (+$400,000)

Pocket #3 – Insurance rates up 23% (+$187,000)

P. #4 – Turnover of safety ee’s +15% (+$89,000)

Other pocket costs = - $676,000 >

40

ddHR goal # 15 ddHR reveals the cost of slow

Last year they had only 110 vacancy days in loan officer positions at this Midwestern bank, before the CFO cut the recruiting staff by 20% (The root cause)

This year there were 210 position vacancy days

$5000 is the lost revenue for each day that a loan officer position is vacant

X the 100 additional excess vacancy days equals a $500,000 loss per year

The savings from the 20% recruiter cut was only $200,000 (a minimum $300,000 net loss)

OMG!

ddHR goal # 16Predictive metrics alert you of upcoming problems

41

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Employee retirement projection alert

(% of all employees likely to retire)

2015 2016 2017 2018

Alert: Retirements will triple by 2018

at a cost of $12 mil. in lost productivity

42

Benchmark Google (read Work Rules)

Put together an HR analytics team

Hire HR pro’s with data/analytics skills

Work with exec’s to prioritize HR areasShift from historical to real-time and

predictive metrics to alert

Start with retention –predict who will quit

Next rec. – ID factors that predict success

Run a small split sample/ control group

Use stats to ID what correlates with perf.

Initial action steps to consider under ddHR

Do we have time for questions?

43

Did I make you think?

JohnS@ sfsu.edu or www.drjohnsullivan.com