Yes, every engineer can achieve more with our latest tools, but we are also expected to do much more. And sometimes, the expectations rise faster than engineers can adapt.
Our job has completely changed in the past 2 years. Instead of helping, many companies just add pressure - by saying engineers are not needed anymore, and by doing massive layoffs.
Almost every manager I’ve talked to in the last 3 months was told some version of: “We need only very high performers today, who adapt very fast. Don’t hesitate to let others go”.
Netflix is very famous for their “Keeper Test”. 2 questions they require their managers to constantly ask (and encourage the employees to ask their own managers):
If this person wanted to leave, would I fight to keep them?
Knowing everything I know today, would I hire this person again?
The goal of course is to keep only ‘top 1% engineering talent’. Those are actually great questions to ask, and I do believe that many managers don’t deal well with low performers.
But the world is not black-and-white.
In one camp, you have those who clearly pass the keeper test. Easy.
In the second camp, you have those who clearly don’t. Those you know are just not a fit for the company, and you’d feel relief if they quit. Harder, but you know the right decision.
But there is a third camp that includes many engineers, where the answer is not very clear. You’ll answer the keeper test with something like: “hmmm, yes, I would try to make them stay, I’d be sad if they left. And I’d probably hire them again”. But it’s not a “hell yeah” answer.
By definition, “hiring only top 1% superstar engineers” can’t work for everyone. You probably don’t pay top 1% salaries, and how do you find them anyway? I’ve worked with maybe 3-4 of those in 15 years.
But I’ve worked with tens of GREAT engineers, who can do GREAT work!
I believe that companies and managers don’t work hard enough to bring them along on the journey. It’s ‘sticks’ everywhere, and a lot of pressure.
I’m a huge believer in the value of long-running teams. The teams that will go through this transition together will become much stronger ones. That’ll become even more valuable as time goes by - the more things you all do in parallel, the more context and friction are created. And the more people are used to working together - the better they deal with it.
My point: there is a lot of value in keeping your existing engineers, and helping them adapt.
How to Maximize the survival chances of your engineers
For me, these 3 steps worked the best:
A very honest discussion
No surprise here - it’s basically the universal advice on good management. What are they good at? What do they struggle with? It’s very hard to build trust and progress if they don’t know what you really think about their work.
Stop with the ‘improvement’ feedback
Common advice is to continue hammering that feedback. Every time you see something they can improve - tell them. Radical Candor, right?
For some people, that might work, but I’ve also seen it kill any motivation to improve. Who wants to finish every day feeling like they are not good enough? Where would you gather the energy for work that way? Constantly receiving feedback is mentally exhausting.
Instead, I recommend taking the opposite approach, and focusing on the good things they do. Discussing together what worked well there, thanking them, and giving praise where due. It shouldn’t be forced or fake, but it can be for even smaller things, not just huge steps forward.
The difference in your feelings at the end of the day, on days you were told you did a good job, versus days where you were told you suck - is huge.
In my opinion, taking that approach for a couple of months is the key part of maximizing success chances. A positive flywheel is created - more energy for work, good results, more improvement. And suddenly, you become EXCITED to work with that person. And then of course you skip the next stage.
The PIP
You gave them direct feedback, and then a couple of months of positivity, and it didn’t work. They are still not meeting your expectations.
Now it’s time for the last chance. Entering a performance improvement plan process has a very clear meaning - the engineer is not meeting expectations for their job, and without improvement you’d have to part ways. Until that moment, that possibility wasn’t usually mentioned out loud.
For some engineers, it’d be a wake-up call, and they’ll be more open to hear where they can improve. For most of them reaching this stage, it’s the way out.
If you want the engineer to have any chance of passing the PIP:
You should FULLY believe that they can improve. A skeptic approach can set them up for failure from the start.
Again, provide very clear feedback, with concrete examples.
Be there for them. Even daily check-ins if needed. Let them choose how to use your time, but make sure to have the availability for it.
The hardest part of creating a good PIP is having concrete but not stupid goals (like increasing the number of PRs or creating fewer bugs). Let’s say the problem is the engineer is not adopting new tools fast enough, and as a result they work very slowly.
In this case, I’d think about a project that needs to be done, and let them lead it, setting a clear timeframe and what you expect at each stage.
Final words
With the current level of stress and uncertainty in the industry, focusing on the positive parts can do wonders. To just for a little while, stop talking about slow adoption, LLMs, and software factories. Recreate excitement for the job by focusing on the good parts.
What I enjoyed reading this week
Management in the age of AI. On why engineering management is becoming more important as AI changes how businesses work (and not less).
There is no such thing as technical debt. When the metaphor is doing more harm than good
Broken cross-team collaboration . When you optimised each team's own velocity so hard that you have killed the company's.
