(above, scribbles of a machine trying to scribble before it sets out to write an entire life's story. Or whatever a machine wants to write, anyway)
For the past 8 years I have been working on understanding why organizations such as a large, international financial institution should invest significant efforts in developing a number of initiatives around new uses of data. Our journey started thinking about the role of banking in future cities scenarios, and from there we traced the most significant role to that of the descriptive power that lies in the data banks have on, essentially, how life flows in cities.
Understanding how we could best use that capacity posed no trivial challenge. Data is a multi edged sword, one that serves in many an honourable battle, but one that may burn in anxious, impatient hands. Hence since the beginning we navigated the question of how should we as an institution approach the role of data in a transformation path that had already started, in such a way that it would unlock the promised troves of value we all hear about while preserving trust, balancing operational excellence with the introduction of new data based products, etc. Many things to ponder come your way, which presses towards a desire for developing a strategy that one can use as a guide to make the proper choices along the way.
Thankfully though, we resisted that temptation.
Things to note when dealing with the unknown: do we even know the territory we need to cover? We did not. No matter how expert and complete our knowledge of our business was, we were very conscious of the fact that keying in data and its various means of application would most probably change our business in ways we couldn't possibly foresee back then. With no map in uncharted land, we chose to perform small, targeted experiments to start discovering the space around us, sizing opportunities and above all learning along the way. There is no substitute for what you learn from experience: very much in the same way as when our brains learn, creating new connections among neurons and effectively yielding a slightly different brain structure, institutions need to learn from experience to prepare for the big changes ahead, forcing their internal organizational schemes and processes to adapt to the new situation. Which is something that an acquired strategy, or one designed from scratch and not from your own experience, would very rarely achieve.
As we walked that path, the experiments grew more ambitious. Broader in scope. More complex to coordinate, involving more stakeholders, reaching further into actual production stages, getting into the hands of the customers. With each increment in scale comes an increment in learning, and this is a road that every person that has departed into increasing uncertainty recognizes easily. There comes a time when the institution sees this new capacity coming of age, and out of maturity rises again the question of a data strategy to lead the complex efforts into tangible results.
My take on this is that there is no such thing as a data strategy. There is no such thing as a digital strategy. There is an institutional strategy, and this is pretty clear from the standpoint of the transformational capabilities that data has.
Whichever efforts a data strategy has to organize, they must be intimately tied from the very beginning to the efforts that the company strategy prioritises. Data and its various considerations must be there at the beginning of time for any initiative that any institution may approach. Data and what you do with it, the processes around it, the governance, the algorithms, etc. is one of the forces that shapes what can and can't be done. Hence, you cannot separate what to do with data from what your company wants to do. Once you have a clear strategy for your company, for your institution, then you may organize what pertains exclusively to the data realm, and you may spell out a specific strategy for that part. It is a simple and quite obvious observation, but there are more than one or two institutions and companies out there, or more than one or two thousand of them, that are trying to write, buy, import or inspire a data strategy that exists in a separate form of the main corporate strategy. Not to mention that there are more than one or two companies out there without a clear strategy to start with, but we won't go into that here.
I believe this is much easier to understand when the institution has made the effort to make honest questions and honest experiments about how to approach this issue. Designing and executing experiments that can really add new knowledge around how a business changes when it considers data from a foundational standpoint requires a bold attitude and a good deal of stakeholders' patience and understanding. Which is something that also requires work and experimentation, of course. Reflecting on what those experiments mean for the company itself is the very act of planting the seed for a successful data strategy, or rather just a successful strategy for the company.
Along the way, questions such as how does the company create value for their customers, which data sets are most important, how to organize your teams to collaborate better and many more will pop up. Organizations need to have the right questions at the right time, that is, they need to understand the previous steps well to understand what the next question actually means. Again, very much like our brains need time to settle down the new learnings and create new connections between newly acquired and previously stored knowledge, hence incorporating the new knowledge in a useful, actionable way (that is, related to our previous experience and to our contexts), organizations need to consolidate the steps on their evolution before they tackle the expansion of their data efforts. Like I've mentioned, this is something that will shape the strategy, and strategy will need to shape this learning process as well. A strategy laid out too early in the process won't be wise enough to do this. It comes with time and reflection on the experiences, experiments, the learnings.
The path of understanding how/when/where to use data to improve/expand a company's practice and service to their customers is a neverending learning process. Don't buy into the first bible that tries to establish rigid, canonical patterns. Let your company learn and become its own guiding force. For there is no truth in obeying without understanding.