‘Twas the Week before Christmas, and all through the Zoom, not a vendor was stirring, except the one pitching to a prospective client in an online meeting room…

Everyone’s inbox is going to be full of twee festive themed emails and blog posts torturing the Holiday Season metaphors. My colleague Eoin has shared his end of year thoughts here. I try to avoid them as it can be a challenge twisting the structure and messaging of festive favourites to make a tortured point. However, 2020 has given me a Christmas gift. Sit back now and listen to the tale of a Data Christmas Carol.

The Ghost(s) of Christmas Past

2020 started for Castlebridge with us looking at some research into data management trends. We had been hoping to scope and execute some work to help inform client strategies. But then we had the “Great Disruption” and our immediate plans had to be put on hold as we put our efforts into researching and advising on the various data aspects of pandemic response with clients. But as we dug into the details of contact tracing apps, thermal scanners, and other ‘new tech’ panaceas, we saw that the fundamentals of data management and data quality were as relevant to these problems as to any other challenge we’ve worked with clients on in the past.

For example, a contact tracing app using bluetooth-based exposure notification sounds like a great idea as it can support a much more privacy-protective approach to public health monitoring. But the laws of physics mean that the quality and accuracy of the data underpinning the process isn’t necessarily 100%. But neither is the accuracy of other, more invasive, methods of tracking the proximity of people to each other. Therefore, it was really important that there was clarity on the purpose for Exposure Notification systems as a support for existing public health interventions and not a replacement for them. And then governments needed to determine if the quality (accuracy) of the data was ‘fit for purpose’.

New use of an existing technology in a novel way threw up an age old data governance and data quality problem, with a side order of data protection and privacy.

In May, however, my erstwhile friends and mentors Dr Tom Redman and John Ladley rolled a hand grenade under the data management profession. “Data Management has failed” they proclaimed. This provoked some discussion within the global data management community, but the reality is that I’ve been hearing the same things and seeing the same things for over twenty years, and Ladley and Redman have been at this a lot longer than me. Indeed, one of the first things I ever read about data quality was in 1995 in a Sloan Management Review article (I was that kind of nerdy undergrad who read the SMR). It could have been written yesterday.

Spirit, what lessons should we learn?

So, if organisations have been approaching data management in the wrong way for the last 35 years, what lessons does the Ghost of Christmas Past have to teach us?

Redman and Ladley identified three issues. We have:

  1. Confused “management” and “technology.”
  2. Failed to build political capital (essential to sustain focus)
  3. Accepted half-hearted measures as reasonable Data Management practice.

Scott Taylor added a fourth lesson. We have been telling our stories all wrong and we haven’t crafted our messages with enough attention to our Vision, our Vocabulary, and our Voice.

In this Data Christmas Carol, Scrooge McData has to learn and accept that, while we have had some successes and built some momentum, we haven’t sustained the change and need to learn our lessons.

The Ghost of Christmas Present

In “A Christmas Carol”, Scrooge is visited by the Ghost of Christmas Present to show him the bigger picture of the world he lives in and influences. As we worked through the last few months with clients we have taken a long hard look at the data world we are living in. We were catapulted in March into a world of Connected Working, governments and businesses scrambled for technology solutions to help alleviate and mitigate the impact of the pandemic on lives and economies.

Data became a key part of our daily conversation as governments reported on arcane things like R numbers and “14 day infection rate averages” and the statistical trends impacted whether we could visit grandma or not. But the Ghost of Christmas Present also showed us how the gaps and issues in our historic data management practices impacted on things like the effectiveness of connected working and the productivity of staff. The issue wasn’t one of technology but one of management. And we managed. But the gaps were exposed.

And against that backdrop we have the rise of Data Ethics (aka AI Ethics, aka ‘Whatever we are doing today Ethics’). The Ghost of Christmas Present has show that while this is an important area, the pearl clutching and navel gazing hasn’t moved the dial. In part, this is because the hype cycle around this issue is following the exact same path as the one that we have seen around data quality, data governance, metadata management, master data management, and Customer Relationship Management in previous cycles. And the same mistakes are being made. And the same conclusions are not being drawn early enough. This is not helped by Regulators who want to distort their mandates by getting caught up in the debate about what is ethical or not in data management.

And Data Literacy has struggled to evolve from the conceptual strait-jacket that it has been put in by vendors who tied the idea of being ‘data literate’ to the ability to use and operate data analytics technologies and self-service BI. Which is akin to an industrial revolution definition of literacy as “being able to work a loom” (the corollary being that good literacy like that helped numeracy as you could count to 10 without taking off a shoe).

Forbes CIO Central wrote in January 2020 (when word of a novel respiratory illness in a province in China was beginning to bubble into the mainstream) that “firms must overcome human barriers if they are to enable data-driven transformation“. This article highlighted that the dial has barely moved on data-driven change in almost a decade. This reflects, in the words of Thomas H. Davenport  “a field that is struggling to succeed despite massive investments in technology and applications”.

What now, oh data spirit?

By March, organisations of all sizes were doing in weeks things that might previously have taken years or been considered impossible. But these all largely related to the fundamentals of connectivity and access. The technical and data debt in organisations in respect of governance and management of data and its quality was exposed in many organisations as they put staff working remotely. Workarounds were found, but human barriers remain. And those human barriers haven been compounded by barriers to humans that have been thrown up through the need to work in a remote manner. Good data management practices paid dividends for organisations who were able to capitalize on their investments and implement Connected Work instead.

But amidst all that, the Ghost of Christmas Present showed us how the lack of vision and the confusion of vocabulary remained a challenge for data management practitioners. For example, 2020 is when the Data Protection community seems to have discovered the term “Data Governance” and co-opted it to mean “Compliance”. We’ve had to add that to the pile where “Data Strategy” as “Doing Machine Learning” and “Doing CRM” as “Buying a customer database software tool” have lain for a few years. The clamour for a technology panaceas to fix things has also served at times to drown out the voices of data management professionals (and I include public health professionals in that where their role is the gathering and analysis of data to inform public health responses).

 Not all heroes wear capes (or buy shiny technology)

But while the perfect may be the enemy of the good, the half-cocked is the nemesis of trust. Where investments were made to supplement and build on existing proven processes and practices we have seen successes. And there is a lesson to be learned there about continuous improvement rather than radical disruption. The Ghost of Christmas Present has shown us that where there is investment in trust, transparency, and good management practices for data, there is acceptance of change. But the Ghost of Christmas Present has also starkly highlighted how it is often the simple and unglamorous things that can have the most profound effects. Whether it is wearing a mask and washing your hands or documenting the meaning and purpose of data in your part of your organisation, these are unlikely to win medals. But they are none the less heroic.

The Ghost of Christmas Future

The last Ghost to visit us in this tale is the Ghost of Christmas Future.

Organisations have a stark choice to make in how they manage their data and information to build resilience against future external shocks that make the unthinkable doable and the undesirable essential. And we have a stark need to address the fundamental issues that were identified by Ladley, Redman, and Taylor earlier this year. If, as a society, not just a profession or as organisations, we can get this right we have a chance, a slim chance that we can improve things. But history is not on our side.

  1. It is not about technology, it’s about the information and it is about the OUTCOMES we deliver through the effective management of and application of data.
  2. We need to build on the political capital there is now in place around the importance of data to societal issues and societal impacts. That capital may lie in a dissatisfaction with the status quo in your organisation having struggled through the last few months. Or it may be found in the emerging strategic need to make the changes of 2020 sustainable.
  3. We need to up our game on the application of data management practices whether that is data governance, data strategy, data protection, data quality, or even data modelling. Half-arsed measures should not be accepted.
  4. We need a clear vision, both at a societal level and (more importantly) internally in organisations for what these things are and how they will add value. What is the elevator pitch for data protection or data quality in your organisation?
  5. We need to be rigorous in our vocabulary. Internally, this means nailing down terminology that is used (good metadata practice), but it also means being very clear what we mean when we say things like “Data Protection” or “Data Governance” or “Data Ethics”.

All of these need to be underpinned by some long term skills development thinking around data literacy. This needs to be at a social/societal level so that people can recognise and deal with misinformation on social media. But also so they can understand the information-enabled world we operate in and how their work practices, and indeed their jobs, will be impacted and how to adapt. It is not so long ago that organisations had filing clerks and typing pools. We expected management to learn new technologies as we modernised the workplace. While they can drive the technology, we forgot about the managing of the data.

The Ghost of Christmas Future is warning us that if we don’t learn the lessons of the past, and use them to sustain the progress made in the present, we will have a difficult future.

But if, like Ebeneezer himself, we wake on Christmas Day and embrace the lessons of the past, we might just save Tiny Tim.

How can Castlebridge help?

Castlebridge can help your organisation embrace the opportunities of 2021 through a range of services and supports. From helping you frame your data strategy, tackle your data literacy through training and coaching, or resource up your data protection function, to helping you identify the key critical to quality data in your organisation, we can get you on target in 2021.

Get in touch!

And Merry Christmas and a Happy New Year to you all!!

Daragh O Brien

Daragh O Brien

Daragh is the founder and Managing Director of Castlebridge. He brings over twenty years of experience in data strategy and regulatory operations to the table for clients. He lectures in the School of Law in UCD and in the Law Society of Ireland on Data Protection and Data Governance. He is a Fellow of the Irish Computer Society and holds CIPP/E and CIPM certifications from the IAPP and other data management qualifications.