3 months, 14 days ago

Belief #4 Best architectures are based on learning

Link: http://architecture-therapy.com/belief-4-best-architectures-are-based-on-learning/?utm_source=rss&utm_medium=rss&utm_campaign=belief-4-best-architectures-are-based-on-learning

Have you ever found your organisation repeating its mistakes?

Albert Einstein defined insanity as doing the same thing over and over again and expecting different results. Yet this is exactly what we often see happening in our work life. One reason for this is the so called “zero defect culture” in some organisations where it is simply not encouraged to experiment, fail fast and learn from it. In such organisations any transparency on bad previous performance will be perceived as failed leadership. Another example is the “fairy tale company”, one that only tells rosy stories about itself, and one where all bad news is ignored or twisted into a positive angle. We can unnoticed slide into a starry-eyed “yes culture” where seeing real problems becomes a matter of not having the right attitude or mindset. Because there are no problems, right? There are only challenges.

A system architect, that we know, assigned to an agile release train in her
company had to fight hard to change a flag in the management report from green
to yellow, even though the situation was critical to the success of a major delivery.
Leaders of fairy tale companies can very well end up being blind drivers at the
steering wheel.

Yet another common observation is that people with previous experience are
deliberately not involved in a new change initiative. Presumably because they
are burdened by too much detailed knowledge that will limit the solution space,
or they are perceived to have a negative attitude based on “we tried that before,
and it didn’t work”.

It is also common to disregard the business knowledge people have and
only focus on the fact that they work with a given technology. For example, experienced
developers have been solving the business problems with mainframe technology
for years, and now they are not involved in the cloud-based modernisation as their
knowledge is deemed irrelevant. This is truly a dangerous approach leading to a
kind of “corporate amnesia” that probably only very large organisations can
afford – for some period at least.

No matter what, this is perhaps the single most expensive tendency we
see in many organisations. As the Spanish philosopher George Santayana said,
“Those who cannot remember the past are condemned to repeat it”.

So best architectures stand on a culture and practice of looking back and
learning from past implementations to activate your corporate memory.

Examples of
valuable learnings,

  • what were the important features when business stakeholders started to understand the solution?
  • is it desirable to develop operational processes in separation from the related reporting-, analysis- and model development processes in that same business area?
  • what should we do different next time if a pilot or MVP became the final solution?
  • is a specific technology working as expected?
  • does the implementation effort (cost and time) of a packaged standard-system meet the estimates in the business case when we include the actual total cost of ownership and change flexibility?
  • what can we learn from implementing new common solutions while not decommissioning existing ones?
  • where do we see the “interest” curves on technical debt not being linear, or have an impact on other areas?

Do you work in fairy tale organisation? And what is your experience with establishing a culture around learning from past transformations?