I've been noodling over this lately.
If you want to ask someone to change how they work, change the tool they use, change their workflow -- it needs to be an order of magnitude better to justify the cognitive and opportunity cost.
Not 50% better, not twice as good. 10x better.
If you want to ask someone to change how they work, change the tool they use, change their workflow -- it needs to be an order of magnitude better to justify the cognitive and opportunity cost.
Not 50% better, not twice as good. 10x better.
And until fairly recently, I think that most developer tools didn't reach that bar. Yes, you could improve the rate at which you shipped, the speed of your recovery, etc, but the benefits weren't pronounced enough to be unmistakably worth it for many teams.
I believe it's just in the last two, maybe three years that we've crossed that threshold.
The DORA reports have some fascinating clues here. Consider the 2017 report, which compared high performers from 2016 and 2017. Not much change, and mostly convergence.
The DORA reports have some fascinating clues here. Consider the 2017 report, which compared high performers from 2016 and 2017. Not much change, and mostly convergence.
In the 2018 report, we saw the emergence of an Elite category...
and in the 2019 report, the Elite category has tripled in size, is accelerating, and basically achieving fucking escape velocity relative to the other performers (who are still basically where they were in 2016).
and in the 2019 report, the Elite category has tripled in size, is accelerating, and basically achieving fucking escape velocity relative to the other performers (who are still basically where they were in 2016).
These are the teams who are leaning hard into not just observability, but feature flags, automated delivery, swift automated deployments within minutes after code is committed, shipping one merge at a time, empowering engineers to own their code in prod -- the whole banana.
The elite teams are not just orders of magnitude faster and better than the low performing teams, they are an order of magnitude better than the *high-performing teams*. It is worth it.
You cannot afford *not* to invest in this shit now. Your business's life will depend on it.
You cannot afford *not* to invest in this shit now. Your business's life will depend on it.
(I am now going to grumble again over how gutted I am that there will be no more DORA reports. Fuuuuuuuucking hell, first Reeder, now this?)
