The Keeper test is inspired by Netflix. It’s one of the most popular aspects of their high-performance cultural (see references).
The idea is to focus on managers’ judgment through the “keeper test” for each of their people: If one of the members of the team was thinking of leaving for another firm, would the manager try hard to keep them from leaving?
Those who do not pass the keeper test (i.e. their manager would not fight to keep them) are promptly and respectfully let go so they can find someone for that position that makes them an even better team.
Getting cut from our team is very disappointing, but there is no shame. A job at Dgraph is not meant to be a stable, retirement job.
This is a rocket ship, where every member of the crew brings something special to the team. It is full of risk, adventure and learning. We want to create a culture where every team member is a star performer – a team of humble, talented, high performing equals.
This test can lead to a strong performance based culture quickly. Anyone who is not an A player is (respectfully) let go. Those decisions are made quickly, so a stagnating employee does not hold up the team and the company back.
Having vacated that seat, we can now bring in a star, who can be an A player and be a contributing member to a high-performing team.
I don’t believe in 10x employee. The star players can achieve things which are not achievable by others, period. They’re not 10 times better, they are infinitely better.
10 average painters can’t become Pablo Picasso, Van Gogh, or Leonardo Da Vinci. They would continue to paint average paintings. If you want a masterpiece, you have to hire a master.
Moreover, the star players already know their worth. And they expect the company to continue to add only other star players. Otherwise, they would realize sooner or later that high performance is not a priority for the company, only a nice-to-have. And low performers can build a good career in the company by engaging in politics. Once star players realize this, they would move on, leaving only the low performers behind.
The obvious downsides seem like this would generate a culture of fear. But, having talked to folks at Netflix, it does not do that. When Keeper test is used to let someone go, that generally does not come as a surprise. The open dialogue between the employee and manager allows for issues to be flushed out in advance.
What about under performers who are loyal to the company?
Loyalty is good as a stabilizer. People who have been stars and hit a bad patch, get a near term pass because we think they will likely become stars again.
Similarly, if the company hits a temporary bad patch, we want people to stick with us. But, unlimited loyalty to a shrinking firm, or to an ineffective employee, is not what we are about.
What about under performers who are hard working?
Sustained B-level performance, despite “A for effort”, generates a severance package, with respect.
Sustained A-level performance, despite minimal effort, is rewarded with more responsibility and great pay.
How can the downsides be tamed?
I think this system works when you truly value the people on the team. Their salaries should reflect their worth. And we should ensure that the ones who we fight to keep, are happy at Dgraph, and feeling valued.
The Keeper’s test should only be applied during the perf cycle. It should be done by having a written answer from the manager, which is visible to the employee. So, they know exactly where they stand in every perf cycle.
Also, regular 1:1 is a good way to ensure that the there’s continuous feedback flowing between the employees and managers.
What is the effect of letting someone go?
There is a definitely impact on other team members when someone is let go. But, people leaving or being let go in a startup is unavoidable. As a startup, we have a limited runway and we have a certain goal to achieve.
An employee who is let go can still find a good job in the market. And our severence package should allow that job hunt to be convenient.
However, as a company it would be hard to recover from the ill-effects of an under performer. We would have burnt money in keeping that employee, while sliding on our goals and targets. And the thing is, more often than not, these people know they are not performing well and would ultimately leave anyway, leading us to search for a replacement later down the lane.