HealthTech Magazines

Insight View of Healthcare IT Professionals

Data management Risk and Resilience

Bas van Gils

My studies have led me down some interesting paths lately, and I've written a few posts on LinkedIN with reflections. My exploration of systems theory and brief review of the new Cynefin book are most prominent.

I try to stay on top of data management developments also, of course, and I hope to write more about this exciting field soon. The pile of things to read only keeps growing.

Several readers have kindly suggested new avenues for research which makes me a happy camper. So much to read, so little time!

I'm currently in the process of reading the last two issues of Harvard Business Review (HBR) and Sloan Management Review (SMR) that were still on the pile. In HBR, I stumbled across the following article:

Suarez, F. F. and Montes, J. S. (2020). Building Organizational Resilience. Harvard Business Review, 98(6), 47-52.






This is an interesting topic. A quick peek at Google Scholar shows that there have been many articles on this topic over the years. Two aspects of this specific article caught my eye.

First, in the opening paragraph there is a statement that reads "Researchers have identified three broad approaches to getting work done, and what they’ve learned can help managers respond more effectively to highly changeable environments".

Second, the article presents a toolkit/ framework that shows which approach to try when. For reference, this framework is as follows

These two topics got me thinking: what can I learn from these three approaches/ from this framework in the context of data management challenges that I often work on with customers?

Reflection on the premise of the article

The premise of the article is that there are three broad approaches to getting work done. This made me think of the Cynefin framework, which distinguishes between

'ordered' challenges with either simple or complicated where best practices and good practices are used respectively. Best practices seem to fit on the routines of the current article, whereas good practices seem to fit on the heuristics.


'complex' challenges where exaptive/ emergent practices are used. These seem to fit well with improvisation in the HBR article


'chaotic' challenges where novel practices are used. This, too, seems to fit on improvisation in the HBR article


All-in all, there seems a fair fit between the two. I am inclined to follow the more meaningful/ detailed distinctions of the Cynefin framework but the three approaches listed here (routines, heuristics, improvisation) is a workable distinction that may be of use to teams in the field.

Relation with data management

Data management is a pretty big field that comprises many areas. In another HBR article, two scholars give an interesting overview of what data management is all about. The article is: 

DalleMule, L., & Davenport, T. H. (2017). What’s your data strategy. Harvard Business Review, 95(3), 112-121.

In essence, this article says that there are two perspectives on data management, and the trick to being successful with data is to balance between the two perspectives. These are:

Data management offence: is where you use data to create value. This happens, for example, when you use data in your business processes, or when you use it for analytical/ decision making purposes.


Data defense: is where you get to grips with the complexity of the data landscape. This is all about data quality, metadata, data architecture and such.


The functional areas and capabilities of the DAMA DMBOK - the industry reference for data management - can easily be plotted on these two perspectives.

As a thought experiment, I have tried to apply the three approaches to the two perspectives. Here is what I have come up with: 

Data management offense:

Routines and procedures: this seems to be about all the situations where data is available for a certain goal (helping a customer, updating a dashboard) and all we have to do is to recognize it as "one of those" and follow procedure. The adjectives "mechanical" comes to mind.


Heuristics: this seems to be about situations where you are asked to solve a specific puzzle through data, and you have a pretty good sense of what is expected of you.

You have a fair sense of what is asked and - even though it may take quite a bit of effort to put everything together - you know how to apply certain rules of thumb to come to a solution.


Improvisation: this area puts me in mind of the famous star trek quote, to boldly go where no man has gone before. To me, it seems to be the area of vague question and wicked problems:

we know something is up, but understanding the problem in full is hard, let alone trying to come to a solution. The example listed in the framework is pretty good: if this is the challenge, then it will be quite the puzzle to figure out what variable to look at, how to collect data, and how to come to a meaningful statistic that will help solve the puzzle.

Data management defense:

Routines and procedures: this is about areas where we know exactly what to do to manage data as an asset. We know how to normalize a set of tables. We know how to setup logging. We know how to setup load balancing.


Heuristics: this seems to be the situation where no immediate solution is available, where you have to work to figure out a/ the solution to a problem but there are good practices and heuristics available that will help you do so.

For example, if you want to manage the quality of your data, then there are different ways to do setup a quality framework. By analyzing what you really need, which tools and techniques are available, it is possible to figure out how you will find the norm (ie. How do we determine if data is of high quality), measure the actual quality of the data, as well as how to engage with others to correct potential errors

Improvisation: this may be the trickiest area to deal with. This is the area where tough questions about managing data as an asset are asked, and no apparent solution or heuristics are there to guide you moving forward.

An example would be the challenge: how can we manage the quality of data in a network of players where there is no single dominant player and each player may have different information needs as well as different requirements for/ ideas on how to structure the data.


The first take-away is a personal one. I think the combination of sense-making (what kind of situation am I in, and how does that guide my action moving forward) is an interesting research topic.

I'll have to revisit this when the articles and books about Cynefin are published. A better understanding in this area will also lead to changes to my book (Data management: a gentle introduction) and will find its way into the trainings that we (Strategy Alliance) offer. In this light, I am also interested in thoughts of other scholars in this field.

The conclusion from this brief analysis is for professionals in the field. I am curious to hear if the 2x3 thought experiment from the previous section makes sense. Would it help to think along those lines in your day to day work? Do you have good examples and cases to share?

The final thought is for leaders. Many organizations that I talk to have "something with data" in their strategic ambitions: data-driven decision making, monetizing data, develop sustainable data infrastructure. It would help if leaders are aware of not only the complexity of the challenges of their professionals, but also the fact that different types of challenges require a different approach and therefore also a different leadership style. As before, I am eager to see more examples and hear your perspective. 

Latest Posts