Back in my days as a management consultant, I recall hearing -- culture eats strategy for breakfast. I think this is valid in many cases, including software delivery. That is, culture is a more significant factor in your success than any framework or approach -- agile, kanban, waterfall or anything. An organization must focus on the right culture, rather than just blindly say "we're doing agile".

The obvious next question is how to shift the culture. My answer -- implement agile. Agile is broader than just meetings and backlogs. It is about values, principles, practices, and benefits -- it is about culture.

At first, this might seem somewhat contradictory. To implement a culture change, we need agile, but to get the most from agile, we need the right culture. With this slight chicken and egg argument, the answer is both. You can start off following basic agile methods, like stand-ups, retros. This will kick off a culture shift and force people to collaborate, share and discuss. Over time the culture will change. The formality and rigor of some agile activities, such as stand-ups, meetings, and backlogs, are simply ways of getting people to behave in a certain way. Over time, these behaviors become the norm, the culture shifts, and it will be easier to introduce more agile concepts.

With agile, like many things, sometimes you first build the habits, then the habits build you.

We tend to make predictions by extrapolating what we see today, as opposed to thinking of what could come tomorrow.

The movie Blade Runner, an iconic science fiction film from 1982, was set dystopian future Los Angeles of 2019, in which synthetic humans known as replicants work for humans. In the movie, despite the revolutionary technology to create artificial humans, people continue to make good old fashioned payphones. Now in 2019, we see a very different world. We are still thinking about synthetic humans and the visions of AI. However, the more basics of life, such as phone calls, have been revolutionized with smartphones and digital communications.

It is far easier to imagine technology as replacing something we all now, like AI for humans. It is far harder to imagine technology coming up with something new that we have not seen.

Landing a jet on an aircraft carrier is tough. It's a moving target; there are crosswinds — undershoot and crash into the carrier, overshoot and go off the edge.

Trainee pilots complete their initial landings with an instructor, who is ready to jump in and correct a mistake. There comes a crucial moment — the point when a trainee pilot can land by him or herself. To make this decision, the instructor does not look at the number of successful landings. Instead, the instructor waits until the trainee has made a mistake and recovered. This way, the instructor knows that the trainee can get themselves out of trouble if they ever get in trouble.

Being good at something is not about being error-free; it's about being able to make an error and still recover.