Ask HN: How does your company do performance reviews?
Share your company's performance review methods (or even link to your performance review forms), and your thoughts on them.
1. We have an in-house developed tool that tracks git and mercurial commits and calculates the test coverage and code quality for each individual developer.
2. We use jira to track the number of points each developer burns (points are shared between developer, reviewer and QA). We also track the number of bugs each developer introduces because closing a bugfix is not possible without first assigning the developer who introduced the bug to the jira task.
3. We use peer reviews where each team member rates each other team member on a scale from 1 to 5 (3 being considered sufficient) on aspects like availability, communication, reliability and result orientation.
4. Team leaders offer subjective input on each team member.
Of course, many of the metrics will be skewed or won't reflect the reality, which is why the team lead has the right to make adjustments to the final mark.
This being said, one of my colleagues and friend was "forced" to quit the company because he scored really low on the jira burn rate metric. He was given a tremendously huge task which nobody cared to properly estimate and break into smaller tasks. As a result he spent 6 monts working on the equivalent of 2 weeks of estimation points. The management (including the team lead which is otherwise a great leader and an awesome person) didn't want to assume any blame for this.
This sounds totally insane. I Cannot believe this happens in real world.
Managers don't say, "So let's see who's the worst performer... it's Tom! Weird, he seems to be one of my best developers. Oh, well! The metrics say he's the worst. Off to fire Tom!"
"Best to Worst Ranking" is the most toxic and destructive environments.
Manu Cornet famous organization chart is the best example of this. https://upload.wikimedia.org/wikipedia/commons/e/e1/%22Org_c...
If you have to let go 20% of your staff, how do you do it? You have to rank people...
It sounds like justification for a decision made for less quantitative reasons.
This happens depressingly often.
"6 monts working on the equivalent of 2 weeks of estimation points"
It's really the fault of the manager for not changing something after the first 3 weeks of this.
Sometimes it's tough to be that ass and say "hey, these weren't estimated properly PM!"...but if it's your job or theirs...
Software development in the Bay Area and other tech hubs with similar (or aspiring to be similar) company profiles can be depressing.
Aggregating them for an individual developer over a long enough period of time might work, but I wouldn't trust it without some statistical hypothesis testing backing the idea up.
But using them to evaluate productivity on a small scale violates the collaborative culture (people over processes) that agile is supposed to foster. I've also seen them misused by:
* mapping them explicitly to hours or days
* thinking they are estimates of value produced instead of work required
* having project managers point and pull in things for developers
But, no, I've never heard of agile mistakes coming up in a performance evaluation or a 180 review for a manager.
The Gitlab performance review process is described on https://about.gitlab.com/handbook/people-operations/performa... (it is over 100 lines so I won't paste it here).
This is why the person doing the work must be involved in the estimation process, that way they have ownership of the estimate and are personally accountable for it.
Not assuming any blame is a trait neither of a great leader or an awesome person. But given the hell hole of a review process, I am not surprised the company has most likely imposter great leaders!
The problem is that the feedback from superiors seems to be weighted much higher so the peer feedback is almost useless so don't bother mentoring or helping someone get back in front of their tasking. This leads to a lot of I'm-working-so-hard theatrics instead of actually being efficient/effective and delivering product.
Combine this with "suggested targets" (quotas) for the number of people in each 1-5 bin, you end up with the usual stack ranking problem that getting on the shittiest project (and keeping it that way to ward off competition) and making it damn clear to your manager that they can't live without you is the best strategy.
I just started at a company that kind of use this system. My first impression is that it is a good system and the weighting seems fair. But I would like to be aware of any potential pitfalls with it (not just for me, but for the company also).
Not really because the stack ranking would still turn everything into a knife fight. It might even be worse because people would blame their coworkers for getting another 3 rating instead of everyone being unified in our hatred for upper management and the performance system as a whole.
We recently had a big Pres/VP management shakeup and the new crew is supposedly looking into changing the performance reviews but who knows how far that'll go. But they did sort out our crazy hiring requirements almost immediately so fingers crossed that they'll set some more fires before things stabilize.
>But I would like to be aware of any potential pitfalls with it (not just for me, but for the company also).
My only tip is to find a project lead/manager who didn't drink the kool aid.
Fairly big (>1000 devs) company. Performance reviews are twice a year.
In theory: before each cycle, an employee and a manager meet to discuss manager's expectations. If an employee "exceeds expectations" he or she gets a good rating. The rating is then approved and bonus and compensation is determined.
In practice: the rating is kind of determined by how employee's work contributed towards "important goals" of the company. This contributes to a deluge of half-assed work pushed into production just before the evaluation begins. Also it of course depends on performance of the other devs. The existence of the curve was never officially acknowledged but it is an open secret that it indeed exists.
The evaluation process is extremely opaque and shrouded in mystery. An employee (and his or her manager too) can never be certain that enough work was done for a particular rating as they are subject to correction at the highest levels of hierarchy (rumor has it that the CTO himself approves the final ratings of all developers above certain level). The process is also extremely long (easily exceeds 1.5 months) and taxing for line managers who have to defend their subordinates' ratings against cuts.
The meetings between an employee and a manager are very awkward. In theory the manager should discuss career prospects and deliver valuable feedback, but what is the point of delivering feedback on something that happened 5 months ago? Why would you wait for so long to do it? So everyone just goes through the motions during these meetings as quickly as possible.
The whole process is very inefficient. Frankly, it stinks. (Personally, I've fared fairly well so it is not an instance of sour grapes). My opinion it that the only reason it exists is because it provides the potential of almost authoritarian control and ample micromanagement opportunities.
I wonder, are my experience and feelings somehow special or is it a common thing in our industry?
What more, it seems managers always set the expectation high, or vague, that way everyone always get 3/5 (meets expectations). By now I'm almost 100% convinced that there is an official HR guideline that they are not allowed to give higher than 3/5 overall rating.
Systems like these create strong disincentives to focus on job performance above and beyond a certain baseline, and strong incentives to focus on self-promotion, politics, and lobbying for sexy projects and allocations. Especially true when there is a forced curve or stack ranking involved. (Spoiler alert: there is almost always a curve, whether formally mandated by process, or informally expected by leadership.)
"You can't all get 5's!"
Result : does not meet expectations
Promotions happened entirely on the visibility of your project. One team reported directly to the VP, so EVERYONE in the team got a raise/promotion. Among the rest, around 5 people were promoted. That team, for reference sake, had 7 people if I remember correctly.
I think the lack of pressure is the reason why this company is the market leader in its niche, has very low fluctuation and a very high quality code base. I used to work in a sweat shop where we had to justify every 15 minutes spent (via time tracking) ... never. Again.
Performance reviews at the large company always involve a lot of paperwork, self reviews, and hoping that your manager gives you good feedback. Reviews are rated from 1-5 where 3 is the midpoint and is considered 'acceptable'. Tbh, I started as an intern there, and I can recall three or four times someone scored 4+ off the top of my head. Getting a 3 was so trivially easy it was almost insulting. Getting a 5 once required a lot of work outside of the office and usually meant you wouldn't be able to get it again the following cycle.
Most of these incentive systems are just focused around being the best personal brand manager anyway, because they incentivize employees to sign up for the shittiest projects and then make a lot of noise and throw around a lot of money and bullshit to make sure everyone knows that they're singlehandedly saving the company.
At the small company, we mostly do small one on ones, and you set some yearly goals for each fiscal year that you get evaluated against. There aren't any 'metrics' involved, and it seems mostly like it's aggregation of subjective reviews from superiors and coworkers. This approach seems to work well for everyone. We have a really technical developer who doesn't really do the social game (and he's remote) and he does excellent every year. I'm more of a mix (I like to communicate with endusers and so on) and I did well this year too.
Frankly, I think the biggest thing is just getting out of big corporate environments. They're tough for mental health, and they're tough to get ahead in.
A colleague of mine always said: you get what you measure.
When I brought it up to various PMs and execs, I got the same answer - "that's the way it's supposed to work."
And they used these point counts and burndowns to analyze performance of each dev in comparison with the rest of the team as a whole!
So if you estimated 1 point for your story, but some other dev estimated 5 for their story, and they are essentially the same tasks just on different projects - that developer is 5x as productive as you are by their record.
(1) Scare quotes around "conversations" because I've never seen management approach an underperforming employee and then realize that they had a big part in that. Conversations tend to be more two-way than that.
I don't think having it around as way to fire someone is the right approach.
At the end of every year, I picked 5-10 people I wanted feedback from- teammates, people I've worked closely with, or people I genuinely wanted to get opinions from. They'd each receive an email that I'd requested their feedback and I'd receive around 10 requests for feedback from everyone who asked for mine. The only mandatory one: everyone reviews their boss.
As well, at any time throughout the year, "Anytime" feedback of an identical nature could be sent in. Usually, I sent these when someone had done something amazing and I worried I'd forget it before review time, or I don't work with them often.
Important to note: I can't ever read feedback given to me. I don't even know if anyone I requested feedback from actually gave it or not.
Each review itself took a lot of time to write, 20 to 60 minutes was normal for me. Everything was asked to be phrased in this way: Situation (what's the background on what happened?), Behavior (what did the person do?), Impact (what was the outcome, effect, etc?). You were expected to give a few good examples and a few bad examples.
At the end of all of this review writing, my manager would read through all of my feedbacks, and compile them into a single overall review with the common themes from many reviews.
Downside: bad managers exist. Here's one person who makes the entire decision and he's the only one reading your reviews. That's a lot of power and you need to trust your manager not to abuse it. Also, that feedback is anonymous to you. If someone is trying to sabotage you, you'd never know- but your manager would, as they can read who sent it.
Upside: if you're into personal growth, an annual review is candy. Here's an aggregate view of where you need to do better. With a good manager filtering out bullshit and finding the real patterns that matter, you got so much out of this.
On the other hand, it took like 2 days to write reviews, and you'd lose your manager for a week or two while he read and compiled.
Of course, I say all this and should be using past tense: due in large part to the media complaining that Amazonians are encouraged to "rat each other out", and that the feedback system was actually a toxic part of the culture, the company got rid of it all. Now you can only say short, nice things about each other and your manager has even more power that he'll never have to justify.
It was good while it lasted.
I know someone who works at Amazon just out of college.... out of 50 that started in June 17, they have 7 left... every week, people get rated and given a score out of 100%....every so often people are let go
Like, I want to believe you but if you're talking about normal everyday developers it would be more reasonable for me to believe you are making this up or misunderstanding something. It just sounds that weird to me.
Over a period of time we started using Impraise (impraise.com), although it's not the best tool - it helps people to force to take the time and leave there current mindset to focus on something like giving feedback (on an equal base, because some people are just very good at giving feedback and keep on talking, while others are struggling and need to write it down). Every quarter, you review 2 other people. People read it, interpret it and carry on. However, I sit down with them after everyone has filled in the questions on Impraise (which are not the default ones, but they're pretty straightforward, like: "how good is the quality of your colleague's work?"). This information is used as a base to allow me to ask even more questions, like: "do you agree with this statement?" or "what could you do to improve this situation this colleague is describing?”. The result of this conversation are 4 things:
1. self reflection,
2. setting personal goals (which is mostly one of the things from the self reflection, getting more focus/attention than other things, which is not shared within the team),
3. create and/or help out with team goals, something we should improve as a team and shared within the team,
4. feedback on the process itself.
These results are written down as notes and then reflected on the next time we sit down.
We have some other tools, methods and things in place to have more of a continuous feedback loop - but it remains a living thing, rather than a set in stone method. If you have questions, you're always welcome to contact me :-)
I've been at my current position (C# Dev) for almost a year and a half full time and I worked part time for almost a half a year.
This seems to be far from normal, but since this is my first job out of school I really just want to see:
- If I'm performing to standard (Adding value) - which I think I am. - If raises are a thing at this company or in general with reviews.
We seem to be shifting towards a more I guess you could call it an "agile" approach (due to our PM). Daily standup meetings talking about what we will focus on for that day, allocate time for it, and report back the next day.
Also, I believe you should ask for a raise every year. Even if you've completely stagnated professionally, your salary is decreasing continuously with inflation.
Companies seem to take advantage of introverted people who are too afraid to speak up about perfectly reasonable requests, like clarifying your performance and goals. It definitely sounds like you're in that boat.
I'm not terribly sure how to go about forcing the issue when managers are unable to fit time into their schedule for me.
I wouldn't consider myself completely introverted, but I don't want to seem needy (ie: cutting into other people's busy days/schedules)
Hopefully I hear something back soon, but I'm not holding my breath.
Rating 1 - 5 (1 best, 5 worst).
Management required to restrict to rating quotas. These kick-in at a span-of-control level of around FTE>50, and below that at the span-of-control manager to manage. For some departments there's increased flexibility. Some managers who really don't see the reason for filling their quota of 'good' (3) sometimes trade with other managers. 1: ~3-5% 2: ~23-25% 3: ~50-60% 4-5: rest.
Formal performance reviews done annually, with interim (6 month) review. Now is 6 month stage.
1-1s on a bi-weekly basis, always documented.
Formal performance review an opportunity to highlight and push to one's manager what they may have forgotten or fell from their consciousness.
Had an associate of mine at the same company two weeks ago get a performance review (more of a contract since he had to sign it), that said he had to improve performance in a month or he'd be fired. The reasons? Too slow/too low quality, but no specific measurable metrics to meet that he'd be able to use to keep his job when re-evaluation comes up in a month.
Needless to say, performance evaluations aren't given a lot of thought here. Wish they were.
I find that even with all these formalities, actions tend to be totally up to the subjective judgment of management, just like any other company. Companies are not legally bound to abide by the rules of their own HR department -- HR is mainly there to provide an illusion of objectivity. In a conflict between leadership and rank-and-file workers, HR will always side with leadership. "Whose food I eat, his song I sing."
I've never heard anyone get cool points at our company for being online more than 8 hours a day, but being online less than that (without reason) is definitely frowned upon.
Sure, communication is important, but there are relatively few things that can't wait half a day (and for those, there are alternative channels if you're not on Slack)
I've had a manager in the recent past try to call me out on it, even though I was always in contact within seconds/minutes..."yeah but Slack doesn't show you as online..."
Any thoughts would be greatly appreciated. We're typically asked to discuss negative feedback with managers beforehand and it's important to note that these reviews are not anonymous.
Good team players that will get along with people and keep things running are very worthwhile in my opinion. Not everyone need to be able to invent their own crypto currency from scratch.
If the problem is more severe, like an inability to code at all, then they might just be in the wrong line of work.
Employees do self-assessments, then managers do assessments, then you meet for your review. The review is more about being on the same page + communicating future expectations.
Internally, we run quarterly 360 degree reviews based on 3 questions: what should this person, continue, stop, and start doing. Each person comes up with their own goals and projects which they’re evaluated on. In between reviews we share real-time feedback so that people always know where they stand and what they can do to improve - in this way review results are never a surprise. Team leads also hold regular 1-on-1s and check-ins with each team member (at least twice a month).
A few things we’ve learned:
- Simply prompting people to share more feedback isn’t enough. Education on how to formulate actionable feedback and how to take and internalize received feedback is essential for it to actually be effective
- Strengths based feedback[1] is highly motivating
- Reviews should always be followed up with 1-on-1 conversations during which a development plan is formed
- Continuous feedback throughout the year is essential to keep everyone on track and ensure that people aren’t blindsided during reviews
- Continuous and 360 degree feedback can help combat bias[2] in performance reviews
- If reviews are results based, goal-setting should always include equal input from the individual and their team lead
There’s not a one size fits all approach to performance management, yet so many companies try to fit their organization into a standard model. Many companies are now customizing their own processes, picking and choosing the elements that work best for their size, industry, culture and the specific pains they want to solve. For example:
- Ratings vs ratingless reviews
- Anonymity vs non-anonymous reviews
- Reviews linked to compensation decisions vs reviews focused solely on development
- Companies allowing people to choose their own reviewers
- Results vs competency based reviews
For advice on the different performance management trends we’ve seen working, feel free to reach out to us.
[1] https://blog.impraise.com/360-feedback/constructive-feedback...
[2] https://blog.impraise.com/360-feedback/4-factors-hurting-you...