Big Data LDN 2021 - Data Governance and the essential CDO

Big Data LDN has been running 22 -23 September. There have been some really interesting sessions covering data governance and the CDO role.

Robin Sutara spoke on the analytics paradox - balancing competing demands for agility and governance. Think big, one technology is not a magic bullet and learn to fail fast. 

Start small with one business problem and build upon that. The data quality with adapt and improve to meet needs of each problem being addressed. It is people and culture that change a business. We need to think beyond the technology. 

 

Great session and an important thought for the future to stop talking architecture and focus on business value. It goes beyond economics and needs sustainability.

Data is a hot mess, so lets cook session from Ben Schein mentioned a key point Data Governance is never done and always be open to new ideas. Data governance is a team sport. You need to find a balance between innovation and experimentation, data teams and consistency, usability and scale. He finished with some important areas to consider

• keep data definitions consistent
• don't wait for  perfect data set to arrive
• data quality is essential for good insights
• let people go shopping for the data they needs
• there is no data magic wand
• data governance is a team sport

Another very insightful session from Cindi Howson on data-driven culture: is your organisation a laggard or leader, gave perspectives on the biggest challenges to being data driven. 

Perspectives on the biggest challenge to being data driven is culture (67%) followed by talent and people (22%).

 

To disrupt your culture bring in a change agent, identify relevance (WIIFM) and organise for collaboration.

She talked about CDOs having a short tenure in organizations being an average of 2.5 years. There is an interesting article about that , Why Do Chief Data Officers Have Such Short Tenures? by Tom Davenport, Randy Bean, and Josh King 

Why WIIFM (what's in it for me) for communication, incentives, skills and tribes and role models.

To move forward into this new world there needs to be a transition in organizational design, from that traditional BI centre of  excellence to this embedded design .

Centralize for economies of scale for common data, infrastructure and specialist talent but decentralize when business domain experts are essential to analytics workflow. Ensure that cross boundary communication is ongoing for best practice, synergies and career management. We are looking at a new hybrid model that is transparent and optimized.

She finished with sharing there is a Data Chief Community (TheDataChief.com) . It has podcasts, blog,  roundtables and community newsletter. 

What we learned from 400 data leaders in CDO summer school  session by Carruthers and Jackson highlighted

• Data strategy is multi dimensional not linear
• Governance is being revised but you need to tell the story. It is still a problem but it needs communicating the purpose through strategy.
• They had Scott Taylor talk about data storytelling - keep simple
• Risk - listen, listen and listen more 
• Soft skills are important
• Data Literacy is a spectrum.
• Nurture your community
• Play to your strengths and culture is the biggest hurdle
• Keep policy simple
• There should be a call to arms on Ethics
• Methods can be disrupted, innovative and evoke change
• It is not about technology outcomes

In the summer school they discussed the DIKW Pyramid which represents the relationships between data, information, knowledge and wisdom

Exasol have published the journey to the CDO. 

https://www.thedatadreamer.com/insights/the-journey-to-CDO/

Data Governance Podcast

 

It was amazing to take part in my first podcast about the benefits of Data Governance and how to get started in Coeo Conversations with Justin Langford.  

The data field is such an exciting place to be at the moment. Data Governance is more than just compliance, it is about managing the whole ecosystem. I thought I would run out of things to say talking for 30 minutes in the podcast on governance, however that was not the case. I hope you find the podcast interesting, informative and fun.

Looking forward there is an exciting digital event coming up Maximize the Value of Your Data in the Cloud: Achieve unified data governance with Azure Purview . The event is Tuesday, September 28, 2021 | 9:00 AM-10:00 AM Pacific Time (UTC-7) Register here

Data Governance: An Introduction

Initially published on the Coeo blog.  

Data Governance is a core area that businesses need to adopt in the data-driven world. Data has been around since the earliest of times, from the first libraries in the ancient world that started to collect and store information.

The collection of scientific research information, from census information about human populations, weather and spatial data to DNA genetic data, have all been contributing to the need to store data for analysis. The breadth of the information that is available for analysis covers our entire planet and beyond, and the population as well as different species. With our life and environment becoming documented to the finest degree the need for categorisation, data labelling and data management has become engrained into our society. Where research led the way for documentation of classification for data, business is now at a crucial time of growth and expansion to enable innovation.

With all data there becomes a continual need for its management and a core starting place is data governance. The DAMA Dictionary of Data Management defines Data Governance as “The exercise of authority, control and shared decision making (planning, monitoring and enforcement) over the management of data assets".

The goal of data governance is to help an organisation to manage data as an asset efficiently and effectively. It provides the principles, policy, processes, framework, metrics and oversight that are required to drive the most business value. Data governance programs have a goal of creating sustainable data management, good data quality that is measured and defining policies and practices. A much-needed area that needs to be considered is that of culture and embedding that culture of data management into the business.

We start with understanding what data assets a business has from the core known data and dark data; data that is collected but not used. The proliferation of duplicate data around a business is key to document. Often the first thing that comes to mind with data governance these days is compliance with all the data breaches that keep occurring. The areas one thinks of here are:  

• Policies
• Transparency
• Governance
• Regulations, such as GDPR
• Standards
• Rules
• Law

These require data inventories and audits to understand what personal data your organisation collects, where it is stored, how it is protected and who may have access to it.​ This is part of the picture that needs to be considered.

DAMA-DMBOK is an international guiding framework for the management of data. The framework includes areas such as:

• Data Strategy – defining, communicating and driving execution​.
• Policy – metadata management, access, usage, security, quality​
• Standards and quality – data architecture and data quality standards​
• Oversight/audit/stewardship​
• Compliance​
• Data issue management – compliance, ownership, policy, terminology, data quality, data access​
• Data management improvement projects ​
• Data asset valuation constantly define business value of data assets.

Consideration for the allocation of roles and responsibilities within an operating model helps guide the adoption of best practices.

In conclusion, managing data assets within a business requires it to be embedded in the culture of an organisation. Having high quality data leads to better business decisions. Having a core oversight function that is provided by a Chief Data Officer helps with keeping the day to day running of data in the fore front of everyone’s minds and you never know where the next innovation will come from.

More Information

• DAMA-DMBOK
• Non-Invasive Data Governance: The Path of Least Resistance and Greatest Success - Book by Robert S. Seiner A Guide to everything you need to know about dark data
• A brief history of big data everyone should read
• A very short history of big data

The Chief Data Officer and Data Innovation

The chief data officer (CDO) is a new role that has been emerging over the last few years. It sprung into life with businesses realising that every business needs to utilise data to succeed in the age of digital transformation. We have spent the last few years thinking about data culture in organisations and how to enable change to utilise organisational data assets and dark data. There is difficulty in the business change process and the CDO role has seen various iterations, adapting to find its place.

As we have seen, data is core to every business and there is opportunity within business to grow and introduce efficiencies by nurturing data as an asset. The CDO is the champion for data within the business and the voice that always thinks about using data as a strategic asset. Data can be utilised intelligently in advanced analytics to create new business opportunity and increase efficiency.

Cross business team utilisation of data requires a change in culture from hording data to making it assessable and advertising in the business what assets there are. CDOs require budget to be successful and must be able to influence enterprise change. There are restrictions of regulation and compliance that must be adhered to, but the CDO needs to be disruptive and drive innovation through the business. 

The role is still evolving and a blueprint is becoming more established. Having a passion for data and storytelling with data is such an envisioning experience for growth. Erudite CDOs have a passion for exploration of data, just as data scientists do, to find that truth and evangelize the possibilities of data value and find that business opportunity. To enable data within the business to reach its potential use there are some foundations that need to be in place, such as data governance.

Thus a CDO should:

• Understand the data context
• Collaborate with business and IT
• Create a data strategy with best practices on managing data, standards for data sharing, policies and procedures across the entire data lifecycle and identify that data value
• Establish data governance operating models and sit on the data governance council
• Oversee architecture best practices to review the impact of infrastructure change
• Reduce barriers to data accessibility
• Drive a data quality vision
• Support operational efforts with strategic oversight
• Be aware of legislation relating to data oversight
• Data never stops being created and the ongoing role requires evangelizing, finding business value and adapting to changes of use with continuous small steps
• Be a data driven culture leader
• Lead data literacy improvement to help with the data culture adoption.
• Work with IT and the business
• Always know the current state of data maturity
• A CDO needs technical and business acumen
• Lead data transformation projects
• Operationalise data usage

It is not only large corporations that require CDOs. Hiring a CDO for SMEs is a challenge, but will be instrumental in enhancing business value and making sure the data that is used provides that much needed business innovation.

More Information

The Chief Data Officer’s Playbook, Caroline Carruthers and Peter Jackson, second edition (2021)

The CDO Journey, Insights and advice for data leaders, Peter Aiken, Todd Harbour, Kathy Walter, Ed Kelly, and Burt Walsh (2020)

World Economic Forum: 6 data policy issues experts are tracking right now  

Published on the Coeo Blog 26 May 2021