This post previously appeared on my blog, 'Absolutely No Machete Juggling'.

I’ve been a professional software engineer for nearly twenty years. Almost every place I’ve worked has had some kind of peer review or peer feedback system, and yet I’ve rarely, if ever, heard someone say something like the following:

“I really appreciate the constructive criticism I got on a peer review. I made an effort to take the feedback to heart, and as a result, improved myself.”

Your mileage my vary obviously, but in two decades I personally have never known anyone to truly appreciate their suggestions for improvement from a peer, at least not within the confines of an official peer feedback system.

Virtually every time I’ve heard peer reviews mentioned in any kind of conversation, it’s been within one of three contexts:

  1. “Ugh, I have a bunch of peer reviews to write and I really don’t want to.”
  2. Someone revealing they don’t like someone else on the team, and a big part of the reason why is that the other person once gave them negative feedback on a peer review that they consider unfair or otherwise simply disagree with.
  3. A peer review comment being used to deny someone a raise, promotion, or transfer to another team.

At the engineering level, based on how I’ve seen peer reviews discussed, the only functions they serve are 1) annoying you 2) sowing seeds of discontent within the team and 3) fucking you over.

None of these are good things for your engineering organization, and you’re virtually guaranteed to be better off with no peer reviews whatsoever. Unfortunately, companies seem to have a total inability to discard peer reviews for the morale-obliterators they are.

Ostensibly, one of the value propositions of peer reviews is the negative feedback they solicit. Constructive criticism, after all, can be very valuable, and knowing where one can improve is immensely helpful to growing within one’s career. Yet it seems to me that valuable criticism can be expressed entirely on a private basis; involving someone’s manager in any way fundamentally alters how negative feedback is perceived. It’s virtually guaranteed to strip the feedback of its value.

Often official peer review systems are purported to help draw out constructive criticism from people who are shy or struggle with confrontation. They can leave anonymous feedback for peers without having to speak to them directly. In practice, I’ve never seen this actually play out as designed: 100% of the time, someone can deduce who left them the “anonymous” feedback if it’s anything more detailed than a numeric rating. The confrontation still happens, and any negative feelings that would result from a direct communication still occur. “Anonymous” feedback doesn’t avoid confrontation, it just means every confrontation now involves three people instead of just two. If you can’t find a way to phrase your constructive criticism so that it wouldn’t offend the recipient, the absolute last person on the planet you should share your poorly-worded feedback with is the person who signs their paychecks.

This is why, years ago, I adopted a philosophy of “Strengths Only” peer reviews. I’m happy to review peers and provide feedback to them, but I will only ever focus on my peer’s positive qualities. I’ll zero in on the skills and values that a peer brings to the table above most other engineers. This helps my peer’s manager better understand why they are valued on the team, and it helps my peer know exactly where they are excelling.

Knowing one’s strengths is far more valuable than focusing on one’s weaknesses. Energy and time are finite resources, someone can focus intensely on improving a weak skill and likely will still fall short of someone for whom that skill is a natural strength. It’s a vastly more efficient use of resources to focus on playing to one’s strengths than improving one’s weaknesses.

If there is an area where I believe someone can improve, I will give them my constructive criticisms privately. This can sometimes be an uncomfortable process, but if the feedback isn’t valuable enough to push through that discomfort, it’s not valuable enough to express at all. If the other person feels like my suggestion is indeed something they’d like to focus on, they can take it up with their own manager on their own terms, set their own goals, and track their own progress.

The above approach avoids an extremely common problem I see which is that when someone in the middle (like a manager) tries to put the feedback in their own terms (often to anonymize it), they accidentally alter the intent of the feedback provider. It’s basically a game of telephone, but with managers accidentally inserting their own biases and feedback into the rewording attempt. In extreme cases, I’ve caught managers hijacking someone else’s feedback to “reword” it to align with their own feedback for someone, which allows them to provide their own feedback without owning it, because it’s supposedly from peers.

I’ve found that almost every time I’ve gotten feedback from someone via a manager, if I was able to figure out who provided it and talk to them, it turned out the manager’s version of their feedback was not what they quite meant. It’s less important that feedback be anonymous than it is be accurately preserved.

So this is my promise to my fellow engineers and other peers: I will never type anything negative into a text box that’s going to be seen by your manager or potential manager. I will speak only of your strengths in an honest way to them (and you) and give you any constructive criticism privately.

If the peer review system will not let me proceed with the review unless I conjure up some kind of negative feedback, I will input negatively-worded positive traits like “should have more confidence” or “works too hard”.

I have taken some flack for this position from people on occasion. I’ve been told this makes peer review schemes “useless,” a critique I find bizarre. If someone believes there is literally “no use” for positive feedback in a peer review system, then it must be the case that they only want negative feedback. That’s not a peer review, that’s a peer critique. And if you think your engineering organization would be healthier if negative feedback between peers were maximized, you’re building a dysfunctional group.

The worst dysfunction I’ve seen is that some companies will factor in the skill of “giving constructive criticism well” when evaluating someone for a raise or promotion. In other words, someone has a financial and career development incentive to find negative things to say about their peers in order to further themselves within the organization. Many organizations will even adopt tools for performing peer-feedback that require peers fill in a box with suggestions for improvement. This has to be just about the most bafflingly destructive way to turn a healthy engineering team into a group of competitive, infighting, political, individualistic jerks as I can imagine. If you make someone’s bank account balance depend on their ability to find fault in their peers, they will.

I’m not going to sit here and recommend everyone adopt this philosophy toward peer reviews (though I do think things would improve if everyone did). I know that some people think so highly of their own opinions that they cannot imagine a world where their constructive criticisms aren’t seen for the life-changing gems of wisdom that they clearly are, but I would like to assure those people that probably not once in their lives has anyone appreciated anything negative they had to say. People generally don’t take kindly to criticism, they get defensive and are much more likely to simply dismiss the critic than they are to take the feedback to heart. Even people who believe they are actively seeking criticism will usually find the criticisms they recieve to be invalid for some reason or another.

Most negative criticism, no matter how constructive, will have no positive impact on the recipient. Our lizard brains see criticism as a threat to our survival. Most of it will be dismissed, and the primary result is much more likely to be a damaged working relationship than a positive improvement. Packaging that criticism into an official system that becomes a permanent record shared by someone’s manager is a virtual guarantee that the feedback will be reacted to poorly. Mandatory peer review systems are a blight on the corporate world, eating away at employee morale and breeding animosity and dysfunction. There’s only one way to transform them into something that might actually help your team succeed: focusing on Strengths Only.

UPDATE: A few months after I posted this, Harvard Business Review published a very similar article saying a lot of the same things - I strongly recommend you read this article, it’s basically what I wrote above but backed up by scientific studies instead of just my personal observations.

comments powered by Disqus