Data Governance Interview with Dr Gerald J Wong
/
I was lucky enough to work with Dr Gerald J Wong earlier this year. He is the Data Strategy and Governance Lead at the UK Hydrographic Office (UKHO), which is a world-leading centre for hydrography and an executive agency of the Ministry of Defence (MoD). The UKHO specialises in marine geospatial data that helps others to unlock a deeper understanding of the world’s oceans. This data is shared with governments, defence users and academia, as well as available through their portfolio of ADMIRALTY Maritime Data Solutions.
Originally specialising in Nuclear Physics and Optical Engineering whilst in academia, Gerald joined BAE Systems Avionics (now Leonardo SpA) to invent and patent sensor technologies. After diversifying with an MBA from the Edinburgh Business School, Gerald then moved to the International Defence arm of the UK Meteorological Office. There he supported weather impact predictions for remote sensors and Big Data issues for UKMO partners such as NATO SHAPE (Supreme HQ Allied Powers Europe) and several national Air Forces throughout Western Europe. Following five years at the Met Office, Gerald transferred into the UKHO to support their transformation from paper charting towards modern on-demand digital services, including Marine Spatial Data Infrastructure (MSDI).
How long have you been working in Data Governance?
I have formally been evolving Data Governance (DG) at the UKHO for over two years. Prior to that at the Met Office I handled ‘Big Data’ and associated Governance issues for five years. As a result, I have been operating within the DG space for far longer than my present role title suggests, and I suspect that many readers would be also able to credit themselves with much more DG experience than traditional or conventional “job histories” usually imply.
Some people view Data Governance as an unusual career choice, would you mind sharing how you got into this area of work?
My journey into Data Governance was a gradual evolution from starting as an end-user of “simple data” during my early Physics and Engineering roles in a closed-loop environment (experiments with clear start, end and/or reset points). This evolved via “richer data” forming a crucial input into decision-making analyses around weather impacts for sophisticated, but well-defined static scenarios, which started to include the need for Data Governance. The final step was moving upwards to formal Data Governance within a dynamic ecosystem of complex real-world dependencies and feedback loops, namely the oceans and human activity above and below the waves, which is dependent upon the physical environment, yet also can affect the physical environment, leading to future changes in human decision-making, and so forth.
This natural evolution tracked my career development from roles with constrained remits – laboratory experiments – to roles that included increasing needs to consider human (mis)behaviour around data and technology, which also includes how to practically integrate data and information to support real-world, socio-economic decision-making.
This evolution closely mirrors the typical hierarchies of corporations and institutions, from the end-user Tactical level of ‘how’ to do something with data, the middle-management operational level of ‘what’ to do with data, and finally the Thought Leadership level of ‘why’ to adopt a certain business strategy for data in the first place. Hence in today’s information economy with increasing adoption of Artificial Intelligence, there is a rapidly growing need for competency and experience in Data Governance – whether that be within marine geospatial data, cyber technologies, green manufacturing, logistical supply chains or retail customer sales patterns.
What characteristics do you have that make you successful at Data Governance and why?
One crucial characteristic is a healthy scepticism and a drive to improve ineffective practices, especially where they’ve become entrenched as tradition, convention or the “way it’s always been done here”. I like to counter such perceptions within organisations, particularly those that genuinely want to evolve, with the view that “if you always do what you’ve always done, you’ll always get what you’ve always gotten”. Long-term existing practices evolved in the past to meet some requirement at that time in that environment and may have once satisfied a need very effectively, but the problem is stagnation while the market and competitors have moved forward.
Another important trait is avoiding unwarranted change for its own sake, as the mirror opposite of static tradition or convention, but this time as the modern trend of “continuous disruptive change without strategy”. This type of “burn it all down” or wrecking-ball approach to Data Governance omits that many long-term practices can still be effective and that change needs to be incremental, integrated and monitored – not only with corporate structures but also human behaviour, means, motivation and opportunity (often the true critical factor). Adapting, modifying, and repurposing established policies or existing processes can help preserve “change capital” for those changes that are genuinely novel or necessarily disruptive. It can also mitigate frictions with those invested in existing practices, such as their users, instigators, designers, and owners; instead bringing them onboard and engaging them with the repurposing and updating.
The third characteristic in a triangle of ideal traits with the other two, is to have a keen applied interest in human behaviour around the use (and misuse) of data or information. Traditional or conventional “Hard Governance” centres around the assumption that people only make the wrong decisions because they have the wrong information or not enough of it. Hence the traditional view of Data Governance coalesces onto hard compliance measures and management surveillance, which includes formal audits, regular in-depth reporting, restrictive checklists, with a focus on top-down, non-negotiable command and control. This approach was suited to traditional mass manufacturing of standardised products but is insufficient by itself for modern data services that are digital-first by design and characterised by near real-time changes.
Soft Governance works with the grain of human behaviour to achieve better results by enablement and empowerment, rather than by command and control alone – principles take precedent over prescription, thus allowing an organisation to leverage the deep insights and frontline experiences of their entire workforce. Shortcut thinking, lack of active engagement and wrong assumptions are some of the key targets for a Soft Governance approach, which still always requires the ultimate backstop of Hard Governance – but meaningfully targeted and monitored using a risk-based approach. Combining the two approaches can yield outsized and transformative results.
Finally, some supporting characteristics to boost the Big Three above include being able to transcend organisational hierarchies, stovepipes and functional siloes. It is crucial to not bury Data Governance within your Data, Digital or Technology domain but to reach out, persuade, influence, and engage far wider afield – especially with customer-facing or revenue-generating areas. The mission is to demonstrate that Data Governance is not merely a cost centre to meet a required need at a minimum level, which is the traditional, outdated viewpoint, but is a key investment in an external marketable strength that can grow business opportunities. Governmental, private and industry users of digital information services are increasingly keen to partner only with trusted providers whose Governance they can have evidenced confidence in for the assured data they consume.
Are there any particular books or resources that you would recommend as useful support for those starting out in Data Governance?
When starting a journey within Data Governance, the main problem with resources is the sheer proliferation of information! The key step for any aspiring learner is to self-govern their own reading by always keeping in mind that “bigger picture” Data Governance is commonly conflated with the technical details of Data Management. Though these fields are clearly interdependent to some extent, this conflation can happen even within respectable publications, so critical thinking is needed by those starting out in DG.
The following three books are my recommendations for building a firm foundation in Data Governance, supplemented by the insights and experience from whichever business sector they operate within. Both the second and third recommendations may be surprising to those expecting technical tomes or lengthy academic textbooks. They are both inspiring reads and essential prompts for thinking differently about DG to unlock progress that is not shackled by outdated assumptions, mainly that people are automatons of a sort and behave in entirely predictable, logical ways around information.
“Non-Invasive Data Governance: The Path of Least Resistance and Greatest Success” by Robert S Seiner is my first recommendation and is a compact, accessible book when compared to more formal textbooks, which can be intimidating and hard to apply for some. Using clear language, memorable quotes and supportive graphics, the book gives an excellent grounding in modern Data Governance, emphasising the value in a low-resistance approach by repurposing existing corporate structures and artefacts.
“Thinking Fast, Thinking Slow” by Professor Daniel Kahneman is renowned within its field with the author’s underlying research into Behavioural Psychology earning him the 2002 Noble Prize in Economics, by evidencing the existence of cognitive biases within people’s behaviour. Cognitive biases are systematic deviations from rational behaviour that might have served humanity in the past (“Thinking Fast”), but now can interfere with rational decision-making in the modern world (“Thinking Slow”). Confirmation bias is one of the best-known examples, but there many more that can subtly exert their influence, even over professionals and experts. These can all cause real-world effects, including injury and loss of life, especially in safety-critical ‘outlier’ situations under time pressure and uncertainty. It is a relatively long and engaging read, but each chapter is self-contained to an extent with excellent opening quotes and memorable takeaways to encourage recall.
“Inside the Nudge Unit” by (now) Professor David Halpern is an excellent follow-on from the previous suggestion, however this time showing the application of Behavioural Governance within a real-world Governmental setting. Halpern is the CEO of the Behavioural Insights Team that was instituted in 2010 by UK Cabinet Office to directly support Government efforts to create outsized effects with relatively small changes of the right type. By giving case studies and real-world examples with their outcomes, this book can inspire readers to begin considering what nudges they can instigate to encourage their existing Data Practitioners to become active and engaged “Data Citizens”. This is needed for modern DG as risk-adverse Hard Governance is akin to “The Law” that commands people what to do or not under specified circumstances. It cannot detail every possible set of circumstances and doesn’t inform how to go above and beyond to create a “Data Community”, which exploits opportunity in new circumstances and requires risk-informed value-judgements. This is ideally achieved by Soft Governance to empower those on the frontline with their wealth of both experience and insight via principles and guidelines, with the backstop of traditional Hard Governance to formally manage the most common and significant risks.
What is the biggest challenge you have ever faced in a Data Governance implementation?
The biggest challenge I’ve encountered is the institutionalisation of Hard Governance as the sole way to carry out effective Data Governance, where DG is seen only as a “Cost Centre”, with a need to have minimally acceptable Governance at the lowest possible outlay – normally for meeting auditing needs or an externally-imposed requirement. This naturally focusses upon documented check and balances, rigid procedures, detailed checklists, all supplemented by top-down command and control that is enabled by management surveillance. This was overcome by explaining that modern and holistic Data Governance places Soft Governance as a first step, which seeks to unlock both the experience and expertise of frontline Data Practitioners, by getting them involved and engaged with DG via principles of Best Practice or other channels for bottom-up Governance.
Gaining traction with the wider workforce takes patience and consistent effort, who are naturally suspicious that DG represents yet more traditional hard measures and controls upon them. By giving “quick wins” via the simplification or removal of outdated procedures that currently hinder them the most, it helps develop trust and the momentum that is needed for more involved changes later. I consider such an approach as stockpiling a notional resource of “Change Capital”; that is built by trust, common understanding, open conversation, and evidence of success. Change Capital is a perishable resource that can be wasted, expended, or will fade over time, so ‘investing’ it wisely in further DG change efforts that will grow it can lead to accumulating DG benefits.
Another challenge associated with the established practice of traditional Data Governance is to neglect that different communities of internal stakeholders have different measures of DG value. Drawing upon the analogy of Change Capital, it is as if these difference communities from frontline Data Practitioners to Strategic Leaders are using different “currencies” when they measure the value of DG activities. It is crucial to be aware of and accommodate such differences, to balance the Change Capital between them.
As an example, without sufficient traction with Data Practitioners, any attempted change will not be sustainable and/or will be undertaken “to rule” with the least possible compliance. On the other hand, lack of traction with Strategic Leaders will result in under-resourcing, lack of management support, and limited room to manoeuvre around any deeper changes.
The ideal solution I’ve implemented a few times is to encourage a common vocabulary around DG, which can “speak value” to as many stakeholder communities as possible, including externally. Using and explaining terms like Soft Governance (with Hard Governance in support) to show how DG can unlock and retain workforce talent, whilst also being able to show market partners the quality of internal DG can be an efficient way of leverage DG changes as a marketable strength and not just a Cost Centre of old.
Is there an industry you would particularly like to help implement Data Governance for and why?
I’ve always had an affinity for the geospatial, ever since completing my Doctorate in remote sensing while at BAE Systems Avionics and Heriot-Watt University. It was a natural shift to considering weather impacts on military operations during my half-decade at the UK Meteorological Office, with a limited side dabble in Space Weather. My current role at the UK Hydrographic Office to embed Best Practice in modern Data Governance is the next step in a career chain from data creator to data analysis on its impacts and then finally to its governance within an organisation.
My current focus in developing the Data Governance needed for an MSDI (Marine Spatial Data Infrastructure) is helping to bring together all my past insights via a vision for an inclusive socio-economic “ecosystem” of marine geospatial information. Someday I would like to progress to the grand vision of an NSDI (National Spatial Data Structure) and its Governance to unify the domains of air, land, sea, space, and cyberspace into a single coherent ecosystem of policies, people, processes, technologies, ethical Best Practice, and inclusive socio-economic outcomes.
What single piece of advice would you give someone just starting out in Data Governance?
The most important piece of advice I would give relates to professional strengths to make oneself - and the practice of Data Governance - indispensable to any organisation. The October 2011 article in Harvard Business Review on Leadership Development under the title “Making Yourself Indispensable” was a key milestone in my own professional development. Although it is important to improve on weaknesses, the necessary step of further developing a personal strength is much less clear for many. Increasing how much you do of something you’re already good at will yield only incremental improvements – beyond a certain point, being even more of a technical expert won’t transform someone into an outstanding leader. The authors suggest that “nonlinear development”, using the analogy of athletic cross-training, can yield exponential results greater than the sum of its parts.
An example given is developing the capability to explain technical problems both more broadly and more effectively, that when coupled with existing technical expertise can work together even more than alone. The leadership of major Big Tech companies exemplify this characteristic as an example. Overall, it is not enough to be a “pure specialist” in Data Governance, but the skillset to persuade and influence different stakeholder communities, along with the ability to demonstrate interdependency and common interests between them, via the common language of DG, is paramount for longer-term career progression.
Finally, I wondered if you could share a memorable Data Governance experience (either humorous or challenging)?
Sometimes effective changes around existing practices that personally connect with people daily can be more than just policies, procedures, technology and data or information. Many years ago, an organisation that I worked for ran a ‘Change and Innovation Scheme’ that invited submission for changes which might make a large improvement over enough time or repeated instances. That organisation had a cafeteria with glass doors and closers to keep them shut normally, but which always seemed to result in dropped lunch trays and spilled soup on a regular basis. The owners of the change scheme were probably expecting a selection of technical and business submissions, but one that got the most votes was to permanently keep the cafeteria doors open, thus leading to no more stained shirts and soup puddles on the tiled floor! The moral of the story is not to prejudge the changes that can make a real daily difference, but to embrace them and support them, thus showing that no suggestions are too small or trivial.
When transforming the practice of DG at an organisation, insights via frontline Data Practitioners is crucial throughout, so one of the tasks of a DG team is to cultivate DG innovators at frontline, not just merely innovations themselves. If you lose an innovator because they felt that their suggested changes were trivialised, then you lose all the insights that they would potentially share in future, or even worse the innovator themselves to another business that will value their lived experience and insights. The message is that Data Governance also must include the people element and the improvements that they can bring to any DG journey for an organisation.
