[alantudyk]

Guys who are applying to Gitlab, make sure you checkout the compensation for your experience and geographic location https://about.gitlab.com/jobs/developer/#compensation . This may bite you in the later stages of the interview. If you look at the extremes, the salary you would get in Luanda, Angola is 1500% more than what you get if you were in Valenzuela, Philippines.

[discordianfish]

While I appreciate the transparency, I don't think such large salary differences are reasonable especially with remote work. In the end you compete globally for remote employees and it possible to work remote for SV companies and get somewhat similar salaries than the onsite employees.

...and of course is living much more expensive in SV but I don't see why that should be "subsidized". No harm in people realizing that it's freaking crazy to pay that much rent.

[StavrosK]

Holy hell, you aren't joking. I compared my city with SF in their calculator, and they pay 3,6 times more for someone in SF!

That means I can get a $160k rise just by moving to SF! This seems super duper broken.

[guessmyname]

You will also spend 3,6 times more money while living in San Francisco.

[nothrabannosir]

The price is that high for a reason: sf has a lot to offer. Networking opportunities, job market liquidity (see all the ONSITE SF posts in this very thread), etc. Perhaps not 3.6 more, but it definitely beats a cabin in the Siberian tundra.

By correcting for rent, you give people that advantage for free. That's why it's unfair.

[NonEUCitizen]

The way to deal with their unfairness is to refuse to work for them (unless you're in SF!)

[arihantar]

Trust me on that I am working with GitLab from past 6 months and I have never regretted a day. GitLab is always open for ideas and this is also work in progress. How many companies have you seen declaring their compensation framework :)

[jamie_ca]

On the bright side, it's seriously awesome that they're so transparent about salaries.

[alantudyk]

That really depends on where you live :) Not so bright, If you are living in a country which has a low rent index.

The calculation is really messed up, rent index should be a contributor not a multiplier.

[sashagitlab]

Hey there! Thanks for the feedback. I'm part of the PeopleOps team at GitLab and we are always looking to improve our comp. calculator. We know it isn't perfect yet and we need feedback from everyone to create the most fair and accurate compensation calculator for everyone. If you want to share your experience and suggestions with me, I'd be happy to chat with you directly and share them with our internal compensation committee. Feel free to email me at [e-mail] !

[vilya]

I like that you guys are being open about this, but the way you calculate compensation leaves a pretty bad taste in my mouth. I very much dislike the idea that you, as an employer, are deciding what proportion of their income your employees should be spending on rent. You're also effectively saying that work done by someone who lives in a cheaper location is less valuable to the company than the same work done by someone who lives in a more expensive location. Maybe all employers do this and the only difference is that you guys are being honest about it, but still... yuck.

[nstart]

As someone who works remotely in a company that does take location into account, I can understand their perspective. I think there's a base value for work and then depending on the power of each dollar earned, there are multipliers to that that eventually give you your final salary. So it's "value of work" * location based spending power differences. Value of work stays constant.

That said I do think the factors used to calculate the salary can improve. In general, I love the idea of thinking of quality of life. Can a person in Sri Lanka enjoy a similar quality of life as a person in SF at least in terms of factors that can be controlled (a company can't control the quality of public transport or municipalities for example in a given location but can provide me the opportunity to purchase experiences or work around those matters).

And that matters because although rent in Sri Lanka is lower than Brisbane, buying groceries is actually more expensive. Buying electronics is certainly more expensive because of the enormous markups and taxes. Compared to a location in the US, I pay nearly triple the value of a given electronics item at times just for the cost of shipping it and then paying customs. Even travel becomes more expensive since I have much more lengthy Visa processes to go through. These numbers eventually add up and while I can save huge amounts of money by living frugally, if I wanted to live a good life supporting my wife and child, the number should be ideally 60k USD and above rather than 38k.

Should mention that Software engineers are considered to be some of the lowest level fodder in Sri Lanka and our good salaries can be something like 12-15k per year. Starting salaries would be something ridiculous like 3k USD per year (that was mine). But that's also why so many people are migrating to australia, US, and canada asap if they can.

[vilya]

Doesn't it bother you that you can provide the same value to your employer as someone in Brisbane, but only get paid a fraction as much for it merely because you're in Sri Lanka?

I'm a remote worker too (and I'm very happy with the way my current employer handles it). As long as I'm available at the times and places my employer needs me to be, why should they have any say in where I live or how I spend my money? I want to be able to manage my quality of life myself, not have it decided for me by someone else!

[jchendy]

> I very much dislike the idea that you, as an employer, are deciding what proportion of their income your employees should be spending on rent.

With rent as a multiplier, it's like they're suggesting 100% of your income goes to rent. It seems like a more reasonable way to take housing into account would be something like:

Salary = Base + Avg Rent

Using that formula, salary might be about $20k more for somebody in NYC than for somebody in Tucson. Using the actual calculator, it's $72k more ($117k vs $45k for senior level and average experience).

[sytse]

We found that rent correlates with market rates, see https://about.gitlab.com/handbook/people-operations/global-c... "Perhaps surprisingly, there was a stronger correlation between compensation and rent index than with the more general cost of living index available through Numbeo (or the cost of living with rent index, for that matter); and so we moved ahead with the Rent Index."

[ernstvn]

We're working on a global compensation framework, to be open and fair about compensation for everyone that works at GitLab. It's described in more detail on https://about.gitlab.com/handbook/people-operations/global-c... . The local rent index (+ a fixed 0.25), NYC benchmark, level, and experience all play in to the compensation. Having the calculator has allowed us to make offers to people in lots of new locations. I'm always looking for ways to keep it robust (i.e. as simple as possible) while being fair as well. If you have specific ideas on how to improve it, please send me an email on [e-mail]

[alantudyk]

If you are really serious about being fair. Ask yourself this question: Why would a guy who can easily get 6K+ USD per month by working on Upwork, Toptal work for Gitlab for 1/3rd of that money and that too Full time?

A quick google search will show you that people use 30% of their salary towards their rent: https://www.google.co.in/search?q=rent+as+percentage+of+sala... What does that tell you about computing salaries based off of rent?

You can see a lot of people discussing Gitlab's salary in a negative way on HN: https://hn.algolia.com/?query=gitlab%20salary&sort=byPopular... Most of them have fluff responses talking about being open and fair.

[ernstvn]

We are serious about being fair :-)

Regarding using the rent index; that was a data-driven decision as described on https://about.gitlab.com/handbook/people-operations/global-c... but as I mentioned, it is a work-in-progress just like everything else at GitLab always is, and I'm open to alternatives / ideas.

[canyon289]

Why are we bashing GitLab when they're at least taking steps to make their pay public? Every company has some version of this formula, you just can't see how off it is.

[mikekchar]

If I ever apply for Gitlabs (which I'm often tempted to do) I'm afraid that my history of conversations on HN will betray my tendency to bring up uncomfortable issues... Oh well... ;-)

Maintaining the idea of fairness is often important in the eyes of employees, but I wonder if it's actually a good idea in practice. Really, you want to hire the best people you can for the money that you've got. So with the system you have in place, if you have 2 equally skilled people, then there is a pretty big incentive to hire the cheaper one.

The end result is likely to be a bit of a skewed culture. Very skilled people are more rare than less skilled people. They are hard to hire, so you will tend to hire whoever you can find. Less skilled people are much easier to hire, so you will find them in almost any geographical location.

The end result will be a company where only the best people will be hired in the expensive region, while the inexpensive regions will have a mixed bag. Because very skilled people are rare, you will end up having inexpensive regions being overwhelmingly represented by lower skill levels (low skill -> easy to hire -> available in any geographic location -> cheaper geographic locations will be hired first)

This will create a power imbalance in the company because the highest density of high skilled workers will be geographically close and therefore in the same timezone. High skilled workers in lower paid regions may have a stigma attached to them because they come from a lower paid region -- and hence are associated with the higher occurrence of lower skilled workers. This may result in considerable friction over time.

I think you can mitigate this problem by creating a second tier pay system for your most skilled workers. This should be a harmonised pay scale and you should pay attention to trying to evenly distribute positions in this pay scale across geographic boundaries. To make it obvious which pay scale people are attached to, you can create new titles for the positions.

You will still have an "Us vs. Them" problem, but at least it will be people you have consciously decided that you want to promote in the company. It is explicitly not fair (in that not everybody is equal), but it makes a clear message of how you want the leadership to work.

It also makes salary negotiations a bit easier. Often people aspire to the highest level of compensation, even if their contributions do not warrant it. When people ask to be promoted to the special pay tier, it creates an opportunity for having a frank discussion about the person's performance. This can clear the air and set proper expectations -- or possibly indicate clearly to the employee that they aren't as valued as they wish to be. Even if someone leaves in this circumstance, it can often be to the benefit of all parties.

Hope you find this interesting/useful. It's always a tricky balancing act, so I wish you luck :-)

[sytse]

We value directness https://about.gitlab.com/handbook/#values so I expect us not to hold this against you.

I think you make a great point. I do want to add some nuance:

1. We currently have great people all around the world. Many of them used to be in expensive locations but moved.

2. Because we're remote only the problem of concentration would be time zone only. But for practical purposes South America is a similar time zone as North America.

3. We have one career path (junior/intermediate/senior/staff/etc.) that is available to everyone based on performance.

Thanks for the comment, it is interesting and I'll share it with our Sr. Director of People Ops.

[zoobab]

Just run a simulation for Belgium, you are way below the average.

[sytse]

We have recently hired great people in Belgium based on the calculator rates. Of course there is no objective market rate, there are likely people getting paid more and less.

[user5994461]

Most people in far away cheap(-ish) countries (eastern Europe and Asia) can't easily go to upwork/toptal and make 6k+ a month.

[dominotw]

not to mention the stress of doing work on upwork.

[jhhn]

yeap! You face people ghosting all the time and some hoaxes...

[codesnik]

toptal AFAIR tries to assign rates based on location, too.

[JshWright]

In my opinion, "fair" would be paying people doing the same job the same amount of money...

[StavrosK]

That's crazy talk. I like Ferraris, should I get paid the same as a Honda-liking person? How am I going to pay for my Ferrari then?

[guessmyname]

I live in Valenzuela, Philippines and can attest that the cost of life is very cheap here, I am not surprised that the compensation is lower. What I wonder is what happens if you get hired while working from the Philippines and then move to a more expensive city like New York, or vice-versa. Will GitLab change the salary automatically? Because one could easily create a bank account in the US and live in a cheap country and there is no need to disclose that information.

[ernstvn]

If you relocate, then yes, that affects compensation. This is described in more detail on https://about.gitlab.com/handbook/people-operations/#relocat...

[colechristensen]

Wow, for exactly this reason I now have no interest in applying. GitLab went from meh to awesome over the last few years but for my city the compensation is just wrong. I'm not interested in applying to have that fight later (or being paid 20-40k under market)

[ernstvn]

Can you please email me at [e-mail] to let me know what city and role you're looking at?