Implementing Data Analytics
for Business Solutions
Nestlé
Office for National Statistics
Layout
Nestle Case Study
Office for National Statistics Case Study
Comparison: Similarities & Differences
What are our individual problems?
Nestlé
HR Analytics in Practice
Who am I?
Business Intelligence Specialist for Nestlé UK&I Analytics Team
History & Archaeology Degree
Worked in Nestlé for almost 2 Years
Dad and Spurs fan
Background in Sales Analytics
Get in touch!
: linkedin.com/in/harry-metcalf
Nestlé at a glance
CHF 89.5 billion in sales in 2016
328,000 employees in over 150 countries
Over 2000 brands
1 billion Nestlé products sold every day
418 factories in 86 countries
Nestlé Story
2011
2012 2010 1990s 1970s 1947 1929
2000s 1960s 1980s
1867
Henri Nestlé
1905
1866
George H. Page Anglo-Swiss Condensed
Milk Company
2014 1938
Simple Complexity
Nestlé
Globally Managed
Businesses
Zones
Markets
Nestlé UK & Ireland
Employing over 8000 people across UK&I
Employees based in 21 different sites
14 Factories and 7 Offices plus additional boutique
stores spanning around 400 miles
Our Analytical Roadmap
“Fostering a culture where data drives decisions and actionable insights to future proof the business”
Business Intelligence
• Focus: Reports & KPIs • Process: Static, Comparative • Data: Pre-planned, added slowly • Transform: up front, planned • Data model: Schema on load • Analysis: Retrospective, descriptive
Data Science
• Focus: Patterns, correlations, models • Process: Exploratory, experimental, visual • Data: On the fly, as needed • Transform: on demand • Data model: Schema on query • Analysis: Predictive, prescriptive
Raw Data
& basic
graphs
Interactivity
&
Self Service
Segmentation
Descriptive
Analysis
Diagnostic
Analysis
Predictive
Analysis
Prescriptive
Analysis
What should we be measuring?
Start with a Business Problem
Start with a Business Problem
Data Sources
Action Insight Business Problem
Data Sources
Data Sources
Data Sources
Action
Action
Action
Action Insight
Insight
Business Problem
Business Problem
Business Problem
When we start with the Data…
Start with a Business Problem
Action Insight Data Hypothesis / Question
Business Problem
What If…?
You could predict, and subsequently prevent,
your top performers from leaving and reduce
attrition costs?
‘Attrition typically costs 1
½ times annual salary’
‘For an average sized
organisation, every 1%
decrease in attrition saves £6m
for every 10,000 employees’
How do you unlock growth potential through people analytics?
Deloitte
Employee Turnover Dashboard
Employee Turnover Dashboard
What If…?
You could predict absence patterns and
determine which health and wellness
interventions would be most effective in each
area, job, demographic? ‘The total direct cost of employee
paid time off, overtime costs and
replacement workers was 29.4% as a
percentage of payroll’
The Total Financial Impact of Employee Absence in Europe
Kronos, 2014
Absence Story Telling
Flu Stress Operations
Long Term
Short Term
Stress Broken/Fractured Bones
12 Days 15 Days
Time to Return to Work
15 Lost FTE
Start with a Business Problem
Business Case
Approval
Consult & Contract
Data Investigation
Analysis
Delivery
Follow Up
• Assessment of business case/problem • Hypothesis/question creation
• Data Investigation, consolidation & preparation
• Decide on Technology/Visualisation Tool • Create Visualisations and Story
• Recount results back to stakeholder group • Create actions based on our insights
• Post-Delivery actions review • Did we fix the Business Problem?
• Consultation with stakeholder group • Assignment of roles
• Review of existing analyses • Approval for valid business case
Thanks for Listening
Implementing the Use of data Analytics for
Business Solutions
Alison Pattimore
Head of Organisational Design and development
Find me on LinkedIn
21
ONS
4300
3110 in our main offices
1190 across the nation
Economists
Statisticians
Researchers
Data Scientists
Administrators
Enabling functions
People Analytics Team
Insight out People Analytics
The
psychology
of people at
work
Business
and data
knowledge
and context
Statistical analysis &
data science
techniques
Intel in
HR Business Partners &
Business Leaders
What we did
Mature
Modernise
Automate
23
Identify the need for change
Our starting point
24
Identify the need for change
Assess the gaps
Strategise
Workforce
Reporting
People
Analytics
Prescriptive
Predictive
Descriptive
Diagnostic
Operational
Feb
18
Feb
17
Challenges
25
Buy-in
Time to deliver
Capability
Numerical refute
Financial commitment
Build buy-in Identify the need
for change Assess the gaps
Strategise
26
Build buy-in Identify the need
for change Assess the gaps Automate
Strategise
A technique to achieve buy-in
Offering the current position – workforce profiles
Explaining what’s happened in the past – trend reporting
Predominately: Business as Usual
Workforce Reporting
Investigating why something has happened – root cause
Taking a data-drive view of what may happen next
Suggesting how the business can mitigate risk – or maximise opportunity
Predominately: Project Based
People Analytics & Insights
Metrics Our metrics had to centre on measuring our strategic
objectives:
• Improve professionalism
• Increase capability
• Recruitment of highly skilled individuals
• Enhancing well-being
• Creating an innovative culture
• Workforce transformation
Underpinned by a statistical framework and standardised taxonomy
28
300 100
Are we recruiting and retaining the right calibre of people?
Planning for the workforce of the future
Impact
30
Build buy-in Identify the need
for change Assess the gaps Automate Modernise Mature
Strategise
Lessons
Differences
• How long our teams have been running
• Customers
• Metrics
• Starting Point: Regular Reporting vs. Projects
• How we got buy in:
Statistical vs. Non-Statistical Organisation
Similarities
• Roadmaps
• Business as Usual vs. Project Work
• Start with a Business Problem
• Build up buy in
• Use Interactive Dashboards
• Team Skill set
• Growing Capability vs. Increasing Demand
Challenges
• Data Quality & Trust
• Credibility & Ability to Challenge
• Automation
• Avoidable Contacts
• Compliance
• Prioritisation
• Analytically Educating Customer Base