The Importance of Good Judgement

When law gets in the way of good
Back in 1988, Mother Teresa (yes, the Peace Nobel Prize winner) was working with Ed Koch, the mayor of New York, to convert two abandoned buildings into a homeless shelter for the people in hard need. Winters in New York are harsh and this would help hundreds, if not thousands, of people over the years. The renovation included establishing a common dining room and a kitchen as well as a small dormitory for the homeless people. Unfortunately, two years into the making, the project failed. The NY building code requires new and renovated multi-story buildings to have an elevator. This would cost an extra $100,000 which they couldn’t afford. The law is the law and even though there are over 1 million buildings in New York without an elevator, exceptions cannot be made. Although homeless people would tremendously benefit from having a roof over their head and warm food in their belly (after all, $100k pays for a lot of meals), the project was killed. All things considered, for sure climbing up 3 flights of stairs was not the biggest problem in their lives.

Process for the sake of process
Many years ago, I was working for a major software company. In order to ensure that software was delivered on time, with quality and with the agreed functionality, a strict process was rolled-out. Requirement specs, design specs, and code, besides a number of other artifacts, had to be produced by certain dates in very specific ways. No deviation was possible either in terms of the artifacts to be produced or the dates by which they had to be delivered. Furthermore, to ensure compliance across product lines, at a world-wide level, every time a milestone was reached, the system automatically verified if the named artifact had been committed into the associated repository. After some initial turmoil things seemed to be working. Teams were checking in all required artifacts, the metrics looked good and management was happy. Nonetheless, on the ground, things were actually not so rosy. Pressured by time and lack of flexibility on what to do and how to do it, teams where checking in specs that were either empty documents or seriously incomplete. The code being put into the repositories, in many cases, either didn’t compile (this was way before CI/CD) or was seriously flawed. After a few months, the reality of the situation caught up with the organization: large number of defects, unhappy clients and complete rebuild of products unfold. All this could had been avoided with a bit of good sense and flexibility.

Guidance protects us from our better selves
A few days ago, I was interviewing a person for a position at Feedzai. One of the questions I asked was: “Give me an example of a mistake that you have done in your career, in a project that failed, which can be directly attributed to you and to no one else.” He told me about being in a security sensitive project and having to move a database with medical information between two different locations. Although there was a process that formally required all the data to be encrypted and moved in a certain way, he was so pressured in terms of time that he decided to simply transfer the data through the network. No encryption used, no process followed. This transfer was readily detected by the client and a huge fall down ensued. This eventually led to compliance problems and the contract being terminated. My follow up question was: “What did you learn from that experience?” His answer was that processes exist for a reason and in many cases they are there to protect us from better selves and to ensure that we do the right thing.

So, what’s the point of these three stories?

As organizations and society scale, they need to put in place processes, guidelines and laws. They help to make sure that activities are done in a consistent and predictable way. They are also there to protect us from mistakes, being it security mistakes, deployment mistakes or even quality problems. Nonetheless, no process protects us from the lack of good judgement. In many cases, they may actually make things worse: people start to rely just on processes and use the processes as an excuse not to really think about what they are doing and inspect the outcomes (“After all, I just followed what the process mandated and what was written down.”) When problems occur, the first instinct of management is to write down even more prescriptive rules. This, in many cases, only makes problems worse. More prescription makes people think even less about what they are doing, how they are doing it, and what is the final outcome to achieve.

If good judgement had prevailed, thousands of homeless people in New York would have benefit from a new shelter; a major software company would not had major software issues; and a project manager would not have seen one of its major clients cancel.

I strongly believe that the only way to scale a successful product organization is by having people that think and always question what they are doing, how they are doing it and, most importantly, what’s the outcome they are trying to achieve. Processes and guidelines are there to help us. But, it’s the responsibility of each one of us, at all levels of the organization, to never lose sight of the objective we are trying to achieve.

Good judgement takes you a long way.

To learn more:

[1] Philip Howard, “The Dead of Common Sense,” Random House Trade Paperbacks, ISBN 0812982746, May 2011.

[2] Charles Perrow, “Normal Accidents,” Princeton University Press, ISBN 0691004129, September 1999.

[3] Atul Gawande, “ The Checklist Manifesto: How to Get Things Right,” Picador, ISBN 9780312430009, January 2011.