Taming the waves:
the foundations of effective data
management
Nigel Turner
DAMA UK Committee Member Principal Information Management Consultant EMEA, Global Data Strategy Marine & Maritime GIS Workshop
Introductions
The data explosion
How ready are organisations to cope? What happens when data is not managed Fixing it – putting Data Governance at the core Planning for and implementing Data Governance Summary
My role & credentials
• 35 years experience in IT & Business Strategy; 26 years in Data
Management
• Initiated and coordinated BT’s enterprise wide information
quality improvement programme
• Subsequently ran a 200 strong Information Management &
CRM practice serving BT’s global business customers
• Since leaving BT in 2010 co-authored Institute of Direct
Marketing online qualification in Data Management
• Also VP of Strategic IM at Trillium Software, Principal Business
Consultant at IPL & Principal IM Consultant at FromHereOn
• Now Principal IM Consultant EMEA at Global Data Strategy
• Data Management Association United Kingdom Committee
Some organisations I have worked with on enterprise data management
About DAMA
DAMA International non-profit, vendor-independent www.dama.org
• Professionals advancing information and data management • Based in the US, chapters on every continent
DAMA UK (the local chapter) www.damauk.org
Mission: To champion the value of Information Management and provide training and support for data and information professionals in the UK
• Three face to face learning sessions per year
• Regular webinars covering best practice, vendor show and tell tool demonstrations, career guidance and industry trends
• Mentoring scheme providing 1-1 support, guidance & career advice
‘Big’ Data – Volumes
90% OF ALL DATA HAS BEEN
CREATED IN THE LAST 2 YEARS
AVERAGE BUSINESS DATA VOLUMES DOUBLE EVERY
1.2 YEARS
2.5 QUINTILLION
GRAINS OF SAND ON EARTH
7.5 QUINTILLION
BYTES OF NEW DATA CREATED EVERY DAY
The Dimensions of Big Data Velocity
Volume
Variety
Key data questions
• What data do we store or have access to? • How and where is it held?
• Who owns the data or is accountable for it? • Who has access to it and why?
• What does it mean - how well is it tagged? • How good is its quality & trustworthiness? • How can I best exploit it?
The industry impact of poor DM – the evidence
In UK in 2013 0.18% of online orders could not be delivered because of poor address data – that’s 1.4 million orders
Millionsof UK National Health Service patient records sold to insurance firms
On average, organizations waste 15-18% of budgets dealing with data
inaccuracies
The US economy loses $3.1 trillion a year
High profile data horror stories (1)
• Major global bank
• Supports community projects in Hong Kong
• Referred its customers to a Community Projects link on its website
High profile data horror stories (3)
SPOT THE DIFFERENCE BETWEEN THESE TWO UK COMPANIES……
• UK Government Companies House confused the two
• Published that Taylor & Sons Ltd. had been shut down
• In fact, Taylor & Son Ltd. had ceased trading
Outcome:
High profile data horror stories (4)
“Millions of NHS records sold to insurance firms”
Poor data management – impact on organisations
ECONOMIC
LAW & REGULATION
Poor data governance: impact on individuals
ANNOYANCE REPUTATIONAL DAMAGE
So how do you fix it?
• Make data a business responsibility, not an IT function
• Implement a data strategy – embrace both improvement & exploitation
• Enforce a data policy to control access and usage rules
• Monitor and measure key data
• Create and run data enhancement projects • Implement Data Governance
The DAMA Data Management Body of Knowledge (DMBOK) wheel
The data problem –
the Data Governance solution
Sales Operations Despatch Finance
CUSTOMER DATA
PRODUCT DATA
FINANCE DATA
Data Governance – a definition
“A process for managing and
improving data for the benefit of
The core principles of data governance
• Key data items & domains are identified and defined – what are they? (Customer, Supplier, Finance etc.), where are they are held? etc.
• Individual business people are made accountable for these data items and domains– often called Data Stewards
• All critical data is defined, indexed, measured regularly & reported on by Stewards
• As problems are uncovered, data improvement initiatives are launched to address them
Traps for the unwary –
why Data Governance can fail
Lack of business leadership and commitment
Failure to link DQ / DG to organizational goals and benefits
Failure to focus on the data that really matters
Giving people data responsibility but not equipping them to succeed
Placing too much emphasis on data monitoring and not
data improvement
Thinking new technology alone will solve the problems Forgetting DQ / DG must embrace all who use data
across an organization
Data Governance barriers: one approach
OPTION 1
ADDRESS BARRIERS REACTIVELY
Data Governance barriers: a better approach
OPTION 2
ANTICIPATE BARRIERS PROACTIVELY
Applying a structured Data Governance Framework
DG VISION & STRATEGY
BUSINESS GOALS & OBJECTIVES
TOOLS & TECHNOLOGY
ORGANISATION & PEOPLE PROCESSES & WORKFLOWS DATA MANAGEMENT & MEASURES CULTURE & COMMUNICATIONS KNOWN / SUSPECTED DATA CHALLENGES
Data Governance framework:
Vision & Strategy: Example Business Motivation Model
32
Mission Vision
Goals & Objectives To provide a full service online retail
experience for art supplies and craft products
To be the respected source of art products worldwide, creating an online community of art enthusiasts
Artful Art Supplies ArtfulArt
C
External Drivers
Digital Self-Service Increasing Regulation Pressures Online Community &
Social Media
Customer Demand for Instant Provision
Internal
Cost Reduction Targeted Marketing 360 View of
Customer Brand Reputation Community Building
Revenue Growth
C
Accountability
• Create a Data Governance Framework
• Define clear roles & responsibilities for both business & IT staff • Publish a corporate
information policy • Document data standards • Train all staff in data
accountability
C
Quality
• Define measures & KPIs for key data items • Report & monitor on data
quality improvements • Develop repeatable
processes for data quality improvement
• Implement data quality checks as BAU business activities
C
Culture
• Ensure that all roles understand their contribution to data quality • Promote business benefits of
better data
• Engage in innovative ways to use data for strategic advantage
• Create data-centric communities of interest
• Corporate-level Mission & Vision • May already be created or may
need to create as part of project • Project-level, Data-Centric
Drivers
• External Drivers are what you’re facing in the industry
• Internal Drivers reflect internal corporate initiatives
• Project-level, Data-Centric Goals & Objectives
• Clear direction for the project • Use marketing-style headings
Organisation & People:
create a CDO led virtual organisation
Chief Data Officer (CDO) Data Consumers Business Data Owner Data Steward(s) Business Data Owner Business Data
Owner CIO or IT Lead
Privacy & Security Experts Data Steward(s) Data Steward(s) IT Subject matter Experts(s) Strategy setting / Data Governance steering group
Processes & workflows: apply ‘Lean’ methods
..identify the “hidden factories” in your organisation
Data Management & Measures: Example Enterprise Data Model
Implementing enterprise DG – applying the Framework
Maturity Assessment Current Status Vision & Strategy Org. & People DM & Measures Processes & W/flows Culture & Comms Tools & Tech. Activity Roadmap Overall Strategy Business Justification DG Vi si on Bus in e ss Dr iv e rs Desired State
Example Maturity Assessment
Description + - RAG
Vision & Strategy
Strong recognition of the need for DG No clear alignment between DG and the goals of the organisation
Organisation & People
Widespread recognition that ownership of data is required
DG is not seen as business as usual therefore there is a lack of awareness
Culture & Communications
Access to shared platforms to help communicate DG messages
No communications plan or ownership of DG communications
Processes & Workflows Elements of DG methodology in place in parts of the business No overarching and consistent approach to DG Data Management &
Metrics
Some validation of data formats Insufficient focus on verification of data
Example DQ / DG Framework Output: Summary Heat Map
Vision & Strategy Organisation & People
Culture & Communications
Processes & Workflows
Data Management & Metrics
Tools & Technology
Priority Level Description 1 – High
Structure or strategy required to realise Data Governance capabilities are not yet in place so requires high priority action to develop them to enable the Framework to meet the requirements
2 – Medium
The foundations or part of the required structure or strategy are partly in place but require further development to enable the Framework to meet the requirements
3 – Low
The capability is already in place and only requires minor actions to enable the Framework to meet the requirements
The Roadmap DATA GOVERNANCE DATA IMPROVEMENT IMPROVEMENT CYCLES DG DRIVERS & DATA PROBLEMS
IMPROVED DATA EVOLVING BAU ENTERPRISE
DATA GOVERNANCE LAUNCH THE DG
In summary: taming the waves
• More & more organisations are transforming the way they do business through data…
– More efficient ways of operating (e.g. marketing, issue resolution)
– New business models (e.g. customer self-service) • To create an effective data strategy ensure you:
– Align data strategy with business strategy – Implement Data Governance
– Focus on improving and not measuring data
– Remember that all in your organisation share the responsibility for good data management
– Deliver benefits early and regularly
Contact Info
• Email: [email protected] • Twitter: @NigelTurner8
• Website: www.globaldatastrategy.com
• Linkedin: uk.linkedin.com/in/nigelturnerdataman
• DAMA UK: www.damauk.org and use Contact Us section