The knot is an enmeshed jumble of many small, niggling issues. Any single one, in and of itself, wouldn’t be that big a problem. But combined, they begin to interfere and combine to cause significant distress to a team, the individuals, and the company.
Indeed, precisely because each one seems like it’s ‘no big deal’, they tend to accrue quickly and without people noticing.
The standups that keep overrunning. The unclear feature delivery timelines. The meetings where it’s only ever the same people contributing. The long pipeline builds. The painful merge conflicts. The multiple-day-long code reviews. The nervy releases.
By the time we notice the pain, it can be hard to see how we arrived there.
So how do we solve this?
Two approaches.
One is that of Alexander the Great, who simply cut the Gordian knot in two with a sword.
Fire people. Reorg the teams. Introduce new processes. Deliver some harsh truth bluntly.
If you know exactly where to cut, and the sword is sharp enough, this can be an extremely effective way of going about it. If there’s not much time, it might be the only viable way. But it’s risky. There will be fallout. And it will be painful.
The other is unpicking the knot, thread by thread. Understanding the threads, seeing how they overlap and stitch together, takes time and requires skill. Sometimes we don’t have enough of either to do this.
But when we do, we can start having an impact on day one. It might not be big, but it is tangible.
By addressing the obvious issues we can change with little friction, we deliver value and build trust. Slowly at first perhaps, but inevitably and incrementally, we gain momentum.
As we free more and more time and energy from being trapped in these draining loops and whorls, we can reinvest them in unpicking more threads, loosening the knot further, in a virtuous cycle.
Then, at long last, and suddenly all at once, the knot falls apart.
I’d love to see examples of specific situations from practice for each of the approaches. This would help me understand the trade offs.