Why we got rid of PIPs at OpenBB

At OpenBB, we removed Performance Improvement Plans (PIPs) in an attempt to increase the company's talent density pool rate.

How did we get here?

We are currently 16 FTE and since the company started 3 years ago, we’ve let go 15 people.

This means we’re letting go of more than 1 person a quarter since the start of OpenBB.

Most people had a 3-week PIP process before their departures. But out of the 15 PIPs done, only one was successful. All the others have resulted in a contract termination.

That’s a success rate of less than 7%, which is extremely low.

Statistics

If we go into the machine learning domain and have a model that predicts that a team member who gets into a PIP is let go every time - this is the classification matrix that we would have.

Which has:

• 93.3% precision - answers: of all the people predicted to be let go (15), how many were let go? (14)

• 100% recall - answers: of all the people that were let go (14), how many were predicted correctly (14)

Now, this isn’t the full story.

This is the equivalent of a physics book treating an object as a point mass, considering the body as perfectly rigid or assuming the system is isolated with no external forces.

So what are the other things to consider? Let’s separate these time-wise:

1. Before a PIP happens

2. During the PIP

3. After the PIP

Before a PIP

Before someone starts a PIP, their performance has already been subpar.

By definition, performance is a lagging indicator, which means you are already late when you catch this person not pulling as much value as others.

Particularly when you consider that the person who would be initiating the PIP is the team lead (TL)/manager and isn’t working as closely with this person as others on a daily basis. Hence, coworkers are likely to see firsthand this suboptimal performance in advance of the team lead or manager.

So, the suboptimal performance from this person over a few days or weeks is likely to go unnoticed and slow down the company.

In addition, individual contributors (ICs) who work closely with this person are likely to notice this before the TL/manager, thus impacting their motivation.

“If this person can get the same compensation as I do for average work, why am I putting in so much time and effort?”

Honestly, if there’s one thing that I’ve learned, it’s that A players get motivated by other A players (“A players attract A players”).

During a PIP

A PIP takes time. A LOT of it.

And that’s the one thing that startups don’t have.

Imagine that you have the following org:

If an IC is underperforming, the TL will discuss it with the IC in advance.

Then the team lead may ask for feedback from other ICs who work with the IC in question.

After that, the TL will talk with the Director about this before initiating the PIP.

Then the Director will mention this to the CEO of the company.

The CEO will likely want to talk with the team lead about this, given that in an organization of 15 people, each team member accounts for more than 5% of the org.

Now, you may be thinking, “But this happens before the PIP”.

This happens before and continues throughout the entire PIP. But, during the PIP, it’s even worse because there are regular meetings for a shorter feedback loop, and there needs to be documentation on the progress.

So yes, this not only takes a lot of time, but it’s also a distraction to the team.

And that’s the other thing that companies need: “focus”.

You can’t fully focus at 100% when you know that someone is “fighting” for their job. And not being able to focus impacts each individual’s performance.

So this inefficiency ends up spreading across the team.

After the PIP

Needs to be let go

Ok, someone was underperforming and needs to be let go.

The company needs to figure out:

• How many options this person has vested and handle the paperwork if they want to buy them

• Whether they have company equipment that needs to be returned

• What the severance package will be

• How to handle the news and how the team will react

Again, this will be a distraction for at least an additional week and will affect other team members, who may be surprised by this.

Particularly because, most of the time, they aren’t aware that the PIP is happening and from their perspective, someone they liked to work with was let go.

Has a successful PIP

Let’s be honest, these cases are very rare.

Not just at OpenBB. I’ve spoken with other founders, and this is the same feedback I’ve received.

But let’s ignore that, we already mentioned it at the start.

Someone on a PIP—almost by definition—isn’t a high performer. They could be a high performer in some parts of the job, but not as a whole. However, this is the exception, not the rule.

The rule, often, is that this person has been doing just enough to be competent at the company—but not excel. Then, over a period of time, due to internal reasons, lack of motivation, etc., they fall below that threshold.

This means that even after a successful PIP, you are putting all of these resources toward getting—not a high performer, but a B+ player.

And ultimately, this is why we are getting rid of the PIP at OpenBB.

Being “good enough” isn’t the culture we want for OpenBB and doesn’t represent our team today. If you let the bar slip, you won’t even realize it until it’s too late.

Again, performance is a lagging indicator and can have both positive and negative effects on the team—so it’s important to protect the team from poor performers.

There are two exceptions to this:

1. Imagine that this person can turn their output into 4x, imagine they had a wake-up call.

Several questions need to be asked:

• If this person can perform at this level, why weren’t they doing it before?

• How long will they maintain this level of performance?

• Will we need to have another serious conversation to get this person to reach this level of competency again at a later stage?

• Will they always resent the company because of the PIP?

It all boils down to this: if this person isn’t motivated by what we’re building, regardless of their skill set, they weren’t a good fit in the first place.

We’re fortunate to have a pipeline of people applying for positions at OpenBB, not just for the money but for the product and the mission of the company.

2. The person is a high performer but has been performing poorly in some areas of the job (e.g. communications, testing, documentation, …)

This person had likely received feedback multiple times, but the PIP made it more real: “This is what we are looking for in a person for your role; you have 2-3 weeks to prove that you can double down on your weaknesses and reach the level the team needs you to be at.”

This is what happened to us, and the person improved significantly, so much so that they are now a core part of who OpenBB is today.

This success story was one of the main reasons we continued doing PIPs.

But the likelihood of it happening again is so low that it’s not worth keeping PIP to look for another success story like this one.

So what’s next?

How we think about talent level at OpenBB

Let’s say you define company’s talent value as the sum of the talent of each individual divided by the total number of team members.

There are two ways to increase this value:

• Hire people who are above OpenBB’s talent level

• Let go of people who fall below the talent level

Or, ideally, do both.

The problem is that for the first option, you often need a LOT of capital.

For the second, you don’t. Not only that but letting go of low performers will accomplish two things simultaneously:

• Increase OpenBB’s talent level immediately.

• Free up resources that can be invested in someone above OpenBB’s current talent value (assuming that companies should always seek high performers and avoid settling for underperformers).

And that’s why removing PIPs leads to an increase in the company’s talent level. You’re not just increasing the talent level once, but likely twice.

Here’s an example:

Imagine we have 5 people at OpenBB with talent scores of 2, 7, 7, 7, and 9. Then OpenBB’s talent level is:

(2+7+7+7+9)/5 = 6.4

If we let go of the employee with a talent score of 2, our talent level becomes 7.4. Then, if we bring in someone with a score of 8 using the same resources, that talent level increases to 7.6.

You get the idea.

What the team can expect?

Full transparency.

We want to build a culture where feedback is an ever-present element, and we don’t need to wait for performance reviews to give feedback that can substantially improve team performance and push the company forward.

In fact, not sharing this feedback puts the company in a worse position, and it is your duty to share it. But do so with candor, in a constructive manner that keeps the team member motivated.

However, each team member must care. This means you can’t rely solely on your team lead to give you feedback every day—you need to ask for it regularly. That’s the best way for you to grow.

Final notes

We made this decision after reading No Rules Rules: Netflix and the Culture of Reinvention, where they also removed PIPs.

Unlike Netflix, we don’t have the resources to:

• Pay top of the market

• Offer a generous severance

We still pay good salaries, just not enough to compete with public companies. This means we need to spend much more time finding diamonds in the rough.

And that’s why we have a higher turnover; finding diamonds in the rough is much riskier.

In any case, I think optimizing to pay top of the market is misguided—at least for startups—as it incentivizes the wrong type of talent.

It incentivizes mercenaries instead of missionaries.

At an early stage, you need people who want a lot of ownership and autonomy, who are excited to work with a team and on a product they believe in, and who have a chip on their shoulders.

Regardless of the startup, I have yet to see someone with this mentality who doesn’t end up being successful.

Note: Most of the people who were let go would be considered good employees in most companies today, and they had strong referrals. But companies have different types of needs that evolve over time, and as founders, it’s our role to look at the company as a whole and understand what it needs at the moment and, more importantly, what it will need in the coming months and years.