In my previous posts on Indeed’s engineering culture, I wrote about keeping teams independent, making better mistakes, dealing with consolidation debt, encouraging autonomy and initiative, and the importance of not stack ranking individuals. These are my closing thoughts on how we work here at Indeed.

Perhaps one of Indeed’s traits that has struck me the most is how individuals, including top engineers and managers, have stayed humble, although they have built this company from the ground up. Bragging and boasting are not behaviors I have seen at Indeed. As I met my colleagues, it seems like being right or wrong is overrated. Instead, we are all focused on doing the right thing. I was surprised how veterans had time and patience. At first I wondered what I did to deserve it. They were curious to hear my perspectives although I had not demonstrated anything yet.

It took me some time to understand why. Then I figured it out. Being data-driven and expecting initiative means there is always a practical and immediate way to settle a burning argument. If you really believe that feature X is what the business needs, you don’t need to convince an army of managers and directors. You don’t need to be resentful for failing to do so, or not even trying. Your credentials (almost) do not matter. The only thing you need to do is to design the experiment that will prove or disprove your theory, and ship it. And if you are not doing so, why are you talking?

Editor’s note: This post concludes our 5-year anniversary series highlighting some of the most important aspects of Indeed’s engineering culture. We want to thank James Dingle for sharing his perspective and reminding us why we love coming to work here every day. Want to join us? Learn more at http://indeed.jobs.

IndeedEng Culture: Be Ambitious, Stay Humble cross-posted on Medium.

In this post, I stand unabashedly against stack ranking. Evaluate how employees perform, not how they compare. Is this really controversial?

If you assign to teams or divisions a fixed budget for bonuses, promotions or stocks, you create de facto a zero-sum game. You need losers to have winners. From that point, you can make lengthy speeches about the importance of being on a team, you will see people nod at you silently, but immediately after your all-hands meeting, they will go on competing with each other.

“But competition is good, right?”

Not when it turns into sabotaging each other’s work, by action or inaction, consciously or not. Teams accomplish much more when they combine their efforts. It won’t happen if peers do not have a genuine interest in helping each other, but have more interest in watching each other fail.

“Everything in the company is budgeted, why not compensation?”

Every budget item has unknowns, including compensation. In a small company, the absolute numbers do not really matter. As the company gets bigger, statistics kick in, and compensation becomes predictable enough without requiring comparing individuals.

“If there is no fixed budget, managers will be too generous.”

After all, it is not their money, so why not give it away. But that would quickly lower the productivity and incentive.

At Indeed, we don’t stack rank or fit to an expected distribution. We have a process that aligns performance reviews across teams and offices, so the company keeps the same definition of what “good”, “better”, “best”, and “not enough” means. No manager decides alone an individual’s evaluation, and there is no competition between teams. This review process is probably the fairest I have ever seen.

What’s next?

In the next post, I conclude my series on Indeed’s culture with thoughts on the most striking trait of individuals at Indeed.

 IndeedEng Culture: Don’t Stack Rank cross-posted on Medium.

In the previous post, I wrote about innovation and consolidation and the importance of balancing the two. In this post, I write about how Indeed fosters initiative through autonomy.

Most compensation systems reward for business impact

Engineers get better bonuses when their impact on the business is greater. Fair, right? No, it is not. Engineers are not salespeople. There are probably not two of them in your company who are executing the same tasks (and when it happens, they try to write a library or something, because they just can’t help it…).

Rewarding by impact cannot be a fair system, can it?

Rewarding by impact cannot be a fair system if, in reality, individuals are assigned tasks. If you still reward for business impact, you will create and fuel a system where the well-intended will do their work and hope for the best, while the less-so-well-intended can train their skills into political back-stabbing games, trying to grab the juiciest projects, as they already have figured out that is how you have defined they will move up the ladder. Guess who you’ll find at the top in a few years, and what it does to your enterprise culture.

The other option, which Indeed fosters, is to let individuals pick their own tasks. They can’t complain about their project, the product or their boss, because they chose what they wanted to work on. You would think that this does not make sense at all. A company is not a democracy or an entertainment park. If we let people decide their tasks, they’ll pick the easiest or safest, not the most important or most urgent, and most importantly it’s going to be completely disorganized and the product won’t make sense.

Employees in a company, as it turns out, are people with common sense. They can understand that there are times for high-business impact work, and times for washing dishes. They need complex tasks and great challenges, but they also need to rest from them and execute less challenging, yet still productive work. I observed that giving more autonomy does not change much what eventually gets done; but it greatly changes the energy and passion put to the task.

At Indeed, we strive to measure business impact and recognize initiative as key components of the quarterly performance review process. By doing so, we give our engineers the right incentives to balance their work in ways that are best for Indeed’s mission: to help people get jobs.

What’s next?

In the next post in the series, I’ll describe why I think stack ranking is a bad idea.

 IndeedEng Culture: Encourage Autonomy, Measure Impact cross-posted on Medium.