«Anything that can go wrong will go wrong». That’s Murphy’s Law. It is also a sentiment that every Product person is too familiar with. And let’s face it, how many times have the best intentions and descriptions from the Product department been carried out with diligence by the developers? And that is not surprising, because each role has its own context, and they see a different piece of the picture. But, how can we improve it?
I have published numerous articles proposing solutions to this problem. Yet most of them depend on techniques that need practice, and fast feedback in order to learn. And precisely that, fast iteration, is often the key point in all my recommendations. However, that is difficult to follow. Most likely, the greatest problem is that we need the whole team to learn and improve. From one point of view, it is good because it is the whole team that improves together becoming a better amalgam, but on the other hand, that is hard, costly, and inevitably slow. At least it’s slow, compared to what I present here.
You should iterate fast in your learning practice, but how to achieve that?
And it turns out, with the incredible tools we have today, there’s a lightning-fast way to learn. Yes, the answer is to use ChatGPT, but more important than that, is how.
In the fastest agile loop iterations, from the idea to the release, there is still a gap of some hours. But in the majority of the developments the delay is between 1 week (as in the faster SCRUM) up to 6 months (as in some frameworks like SAFe).
How much can you learn if you need six months to have feedback on what you tried? Almost nothing. Even if it is one week of delay. You can try something, wait for one week, try to remember why you did what, try something new, and so on… In one month, you would have only four opportunities for improvement. The opportunity to learn increases drastically when the loop is only three hours long. Once the three hours have ended, you have your memory fresh, you know what you did and why, and you can try new things…