I'm one of those people who has a really hard time with the idea of having left money on the table. When I'm in foreign countries, I love to haggle in the markets, but no matter how big a reduction in price I get, I always walk away wondering if they're laughing at me behind my back because I still paid 10x the going rate.
Anyway, working for a large corporation with an opaque salary system was an exercise in frustration. I moved to SF and took a job at a midsize company making what I thought was a very decent salary. I was just out of college and making more than my parents, so I thought I was doing really well. However, after a bit of digging and research, I discovered that the "going rate" for my position was actually 20 - 30% higher. Ugh.
I can totally see the appeal of a transparent salary system, especially when you throw profit-sharing into the mix.
"I always walk away wondering if they're laughing at me behind my back"
They're probably not laughing at you. They are either confused because the foreigner got a decent price out of them, or grateful you didn't hold on for a lower rate. It's the ones that don't haggle that they're laughing at...
My brother gets immense satisfaction about haggling in China. "Don't you give me the foreign devil price! I grew up in Hong Kong, I know how this game is played! This T-shirt is shoddy and only worth 1/6th of what you are charging, I will pay 1/5th and curse your name!"
He can pull it off, despite having grown up in Chicago and being white as the driven snow. Personally I'm glad that Japan is largely a sticker-price society, the stress would wreck me.
I once haggled off a lot of the price on a purse (gift for my mom) by accident: I told the merchant I was checking out the rest of the market first and would come back later.
He just kept going down until I felt no reason not to buy the item.
It's easy to think that things in non-Westernized countries cost peanuts, but it's rarely true. Haggling is almost never over "a few cents" (certainly in no example I've seen - someone who haggles over less than a dollar for a >$10 purchase is probably just a dick).
Answer: "Because we don't call it 'negotiating a raise', we call it 'applying this formula'."
Two immediate problems with the formula and one problem with the concept.
First, "scope", factor 2 in the formula, can be external to the employee. It's in Spolsky's interests to hire people who "could" manage multiple projects if given that responsibility, but the demand for those people inside Fog Creek will tend to be lower than the demand for lower-scoped responsibilities. His formula thus does one of two things: it incentivizes him to hire less talented people --- even when he may want a buffer of talent inside the company --- or it creates a totally subjective ranking system where 10 people are qualified to lead a project, but only the "favorite" does.
Second, "skills", factor 3 in the formula, is also subjective. That he's coded it in a chart doesn't change the fact that the "skills" spectrum is basically the line over which every salary negotiation is fought anyways.
Which leads to the general critique of this, that coding and regimenting salary negotiation does nothing except code and regiment salary negotiation. OK, so Fog Creek is transparent about the behavior they want to incentivize. But so what? The real world still creeps in and subverts this system:
* "I'm capable of leading this project, but I'm never getting a lead role because 5 people with seniority are always going to take the next available slots."
* "I was given this task to complete, but it was poorly specified, and so my output didn't contribute to company success."
* "You'd be ranking me at a higher skills level if you put me on a project that would give me exposure to new technology."
* "You'd be ranking me at a higher scope of responsibility if you'd just recognize that my team lead is a moron."
All the same pathologies seem to remain intact in this system.
This seems spectacularly unlikely given Joel's other writing and the fact that he's running a profitable company, the profits of which substantially depend on the quality of staff he hires. Saving 1% of payroll and giving up any substantial "edge" in development is a bad trade, and regardless of your opinion of Joel, I suspect he's smart enough to know that.
All the same pathologies seem to remain intact in this system.
All of those same pathologies sound to me like pathologies with the people not the system, and the solution is to just recognize that people who can navigate themselves around small obstacles are actually more valuable. People who complain at year-end about why they couldn't acheive their maximum potential need to be steered into a mindset where next time they find a speedbump to drive over or around it, not to stop and complain about the uneven pavement.
(1) I agree that it's unlikely that these problems affect Fog Creek in practice --- although how would we know? But the point is, Spolsky wrote this to advocate for a practice in your company, where this formula probably won't help you.
(2) If "people" didn't have irrational issues with salary negotiation, it wouldn't matter how you handled salary negotiation. The purpose of a system like Spolsky's is to neutralize those problems, and it succeeds or fails based on how well it accomplishes that.
It's in Spolsky's interests to hire people who "could" manage multiple projects if given that responsibility
I'm not sure that's true.
I've never been a project manager, but I have had a job where I would sit idle for weeks at a time because nobody had anything for me to work on. I eventually quit.
So I would expect that someone who could manage multiple projects at Fog Creek, but is only managing one, would not be happy there, either, even if he/she were paid as much as the people who actually were managing multiple projects.
And, yes it's good for there to be a buffer of talent, but that buffer would ideally be spread across the whole company--every person who is managing one project should be able to manage one project and contribute to another--not concentrated in a few stars.
If every person in the company can manage a project, then Spolsky's formula is artificially restraining compensation for the majority of his employees, and the selection of who gets to lead FogBugz and Copilot and Project Antelope or whatever is simply a proxy for the same subjective political BS that happens in every salary negotiation. Which is kind of my point.
Again: no doubt at all this system works for Fog Creek. But it was written up in Inc. because someone believes it could work elsewhere. Not so sure about that.
I went to a Fog Creek open house year before last with this pay-scale transparency in mind (Joel had talked about it in an article about 6 months beforehand) and nobody I spoke to had any idea what I was talking about.
Trying to run a company by formulas like this rarely works. And a 'take it or leave it' approach to negotiating compensation (yes, you are negotiating, even if your stance is not to) would seem to encourage hiring of employees who either don't know their own value in the marketplace or, worse, are looking for a place where there is less pressure to perform.
It might make sense for a non-profit, but if the company is for-profit, it makes sense for the employee contracts to reflect that fact. Of course it's incumbent on you as manager to pay people what they really are worth, and not play favorites for emotional or other random reasons.
"Take it or leave it" is the best price that you're going to get from a given party. It's unclear why one would prefer a lesser price.
It's also unclear why taking a "take it or leave it" offer vs some other metric implies "don't know value" or "looking to coast".
Then again, I'm pretty sure that a significant number of employees at non-prifits aren't volunteering, so if a non-profit wants to keep them, it has to be competitive wrt their compensation.
While some/many employees at a non-profit may be volunteering,
Years of experience is only sometimes correlated to employee value. So, including it in an inflexible formula to determine level will cause mis-levelling. Exceptionally talented staff with few years under their belt will be levelled too low and more experienced staff who aren't so capable will be levelled too high.
Supplemental point: IANAL but I think it would also be illegal to factor years of experience into salary calculation in the UK.
Read the article, particularly the table at the end. After 1 year of experience, the years of experience make no difference to your rank, except for people with an "average of scope and skill" of 1 (out of 6).
What a silly rule. This 'equality rule' is why union shops have some of the worst productivity levels in the industry and why union workers have zero incentives to innovate and become more productive. Under this plan, people have no incentives to give more and work harder. Inevitably, the equilibrium for this type of a system is the productivity of the weakest worker.
Theoretically, across the board decreases. Given the profit sharing though, I think by the time he did that, it would be pretty evident that there was a major hardship to be distributed amongst the staff.
Once a year, my management team sits down, reviews every employee's work, and recalculates every employee's level.
There's your incentive, besides profit sharing[1], to buckle down and work hard. Besides, if Joel & co. have an eye on market wages through use of glassdoor, they're not going to get very far out of line in the course of a single year.
[1] This works at the size Fog Creek is now, but somewhere around 100-120 employees it breaks, because you nobody has a good feel for how individual productivity feeds into company productivity, and it turns into politics.
Software is far more a group activity than building widgets. With an assembly line you can slow others down but it's hard to boost the groups productivity so if you are competent enough to keep your job you are worth X. With software each individual can have a large positive or negative impact on everyone's performance while still being individually productive. Do you write reusable code or break the build?
I would say software is far more of an individual activity than building widgets. On an assembly line one amazing worker can't boost the overall performance by much, but in software you can.
Boosting the group's activity does not make it a group activity, it shows the power of an individual programmer.
Also, look at the one employee that made Fog Creek over a million dollars from his job board concept. Some individual employees can have spectacular results.
As a software developer it's easy to think a million dollar cash flow is the result of a good idea and some code. But a tester to discovers nasty bug could prevent a million dollar problem from showing up. My point is when a small team that's working together on a measurable task you can reward the team but it's easy to overlook the tester or junior developer that was vital to the teams success.
As to your example the reason it seemed so obvious that one employee had a million dollar idea was how little effort it took to get there. However, his idea would not have made as much money if his boss had not decided to up the cost of advertising. And if the traffic was not there to create the market etc. So while the idea might have generated a million in revenue he was far from the only person involved.
If you look at the chart on the second page, it's clearly not "communism". There's a "scope and skill" component to salary level. If I want to make more money there, I increase my skill, and push to get more responsibility. It's just saying that people with the same skill and responsibility get paid the same.
"Under this plan, people have no incentives to give more and work harder."
There are some people ("salesy" types particularly) whose performance is motivated almost exclusively by money, but I find this to be the exception.
As was mentioned in the article, hard work tends to come from high job satisfaction. Compensation is only one component of that, and arguably not the most important, provided it's not unfairly low.
Finally, if you're right and Joel's wrong, how do you explain Fog Creek's success? Do you consider them an outlier? If so, can you give a counter-example of when you've seen this system fail? Fog Creek is strong evidence of the efficacy of this approach, while you seem to be relying on nothing more than your opinion.
Well, unlike a union shop, people can be fired. And we're talking about (50? 100?) people that are chosen very selectively from a nationwide (worldwide?) pool. It's not a 20,000 worker steel plant.
I think you need much more flexibility than this. I've worked for several companies with this kind of scheme and had to fight and juggle with the system and the scales to be able to reward my staff appropriately. It is very easy to end up with a middle-of-the-road performer, who has a broad skill set and much experience and is therefore very highly valued by a formula. It is extremely frustrating to have to pay them more than someone with less experience and fewer skills on paper, but who is perhaps making a tenfold greater contribution to the company.
Rigidity in the formula itself is not the only problem. I've encountered rigidity in how to apply the formula. I was responsible for a team of engineers in such an organization a few years back. They were a fantastic bunch altogether, but we had a real star who transferred in. She'd been barely raised above the starting graduate scale, despite making a pretty significant contribution to the company as a whole. When I discovered this, I set about re-assessing her ranking based on the company's formula and basically worked out that her salary needed to double - it had to jump about 3 pay bands. My request was turned down - it was apparently too big a jump! I was left wondering if I should ask her to stop doing some of the things she was doing - like mentoring new employees, since the company wasn't prepared to pay her for it. I didn't stay there, and neither did the star engineer.
I think in a smaller, less bureaucratic organization perhaps these schemes have a better chance of working, as long as the formula leaves you enough flexibility.
Is employee A working late nights because he or she is working hard, or to fix bugs introduced by careless coding the previous afternoon, or because for every hour physically at the office, 45 minutes is spent reading HN? It's hard for the manager determining salaries to tell the difference.
More likely is that in negotiation system, employee B gets a pay rise because he is good negotiator and employee A loves his job but is less confident so doesn't even ask for a pay rise, so gets nothing.
I makes sense as long as you are not paying less than $90k/year senior level web developers. Does anyone know how much is a Fog Creek Software senior developer salary?
This system can only work if there are non-monetary benefits to working for the company, like a stimulating work environment or increasing independence with increased demonstrated skill...
As someone that runs a profitable company I can tell you right now that my company would a) not be able to hire really awesome people and b) lose all the better guys as a result of this policy.
yes, that's exactly what he's saying. The dirty secret is that the variation in employee 'worth' to a company varies much more than a tight range of salaries based on statistical averages. It's not just talent; some employees become inordinately important due to their knowledge and acquired experience in a narrow area. Losing them might mean shipping a product 6 months late, which could be fatal.
Of course there's a potential here for employee blackmail, but hey, live by the market, die by it. It's a constant struggle to be 'fair' yet maximize your return on investment. These two goals don't automatically align.
I have worked for a university which, under budget pressure, allocated for < 1% average annual raise, but STILL put all the effort into review to spread it out so some got .7%, some got .8% and so on. What I remarked from this was: all the pain and anxiety with none of the benefit.
More recently, the company I work for seems to have moved away from an emphasis on raises and towards an emphasis on bonuses. Somehow for me this is psychologically infinitely superior. I actually can't remember the time series of my bonuses. I suppose if the company was stupid, it would report my bonus as "your bonus is 20% lower than last time" or "your bonus is 30% higher than last time." But they just report the amount. Contrast that with salary changes, which in my experience have always been reported to the employee as a percent change, just in case the employee was not thinking that way anyway.
What these guys are doing appeals to me very much. The real changes in salary come from moving up and down the ranks, which is a public process even though not transparetly linke to salary changes generally. A super simple salary system is very appealing to me. Give me my feedback in significant (>1% effects) chunks, and keep me from doing the equivalent of staring at 16 indistinguishable brands of soap wondering how they picked Tide for me.
In an Inc column a while ago, Joel mentioned that his offer for new grads is $75k/year. Not sure how that scales with experience, but Glassdoor might give you some idea about going rates in the NYC area.
As a sidebar, I've discovered that the cultural reluctance to discuss wages is not shared with other countries. In China, for example, it's quite common to compare salaries.
Maybe this works for Joel. I find it funny though how performance does not get included in his calculation.
Some employees are more productive that other employees and get more done therefore they should be payed more. An employee with less skills can still be more productive that an emloyee with more skills.
The reasons CEOs make millions is because their performance is directly measurable in how well their company did that year.
"The reasons CEOs make millions is because their performance is directly measurable in how well their company did that year."
They make millions because they negotiate a contract that says so, irregardless of performance. They're able to get that contract due to the perception that with them leading, the company will make more money than if they weren't. It's essentially a bet the company is making on the CEO, because even if the CEO drives the company into the ground the contract is still in place.
The point I want to make is that I think it's rediculous to not negotiate for raises and salaries. That is what a free market does.
The link captures what Paul said 5 years ago and it's the point I was trying to make. That some people are more productive than other people and that they should be payed accordingly.
In the United States, the CEO of a large public company makes about 100 times as much as the average person. [3] Basketball players make about 128 times as much, and baseball players 72 times as much. Editorials quote this kind of statistic with horror. But I have no trouble imagining that one person could be 100 times as productive as another.
They make millions because the people determining their pay are CEOs of other corporations with a vested interest in keeping the average CEO salary as high as possible.
Agreed. I think there are multiple factors but I'm positive that is the largest single factor. Basically they get that much because they can.
I actually have a more complex and subtle theory to explain it involving a system to keep feudal aristocracy alive in the modern age, but I'm afraid it is too involved to describe in the tiny page margin space I have remaining.
You have to consider what being more productive translates to. In this case, since we are talking about software, lets assume that being more productive means that you can implement more features of a similar complexity at the same level of quality compared to the guy sitting next to you. Assuming everything else equal between you two, you would expect that you would be paid more. In Joel's model, the only way for this to be accounted for is scope. Since you can produce more features you are now responsible for more features. Assuming that this increase of scope is enough to move you up to the next grade, you would have a higher salary. Otherwise, it could be accounted for with your bonus.
But what if you could implement twice as much features, but you required additional over site to manage those features. You have just created additional work for someone else. Before your additional productivity can be utilized, someone else has to help you convert it to something more usable. Depending how needed that other person is, your increase in productivity could result in a net lose of productivity for the team as a whole.
When talking about productivity, you need to be specific in what you mean, because productivity is hard to measure and quantify.
Very good article. I'd also like to mention that as a libertarian I strongly desire to see efficient free markets and those require INFORMATION.
Attempts to hide information are almost always market manipulation.
In the case of secret salaries, the market is being manipulated for the benefit of employers and employees who are very good at salary negotions in addition to their other skills.
One thing I liked about the military is that we were paid based on time in grade and rank. I at least knew I wasn't being cheated out of something someone else was getting.
I agree, I work in construction where there are clear, visible differences between two employees. If this kind of thing happened at one of the companies I've worked for or with then everyone would quit.
The only people this benefits are the poor employees. I have some natural advantages over the next guy, namely that I can carry about 140 lbs of equipment, while a lot of the guys I've worked with can only carry probably 70-80 lbs. Basically I spend half the time with the petty crap and end up finishing the job sooner. I don't believe I should be getting paid the same as the next guy, I also don't think I should be paid as much as a guy who can carry 200 lbs and can get things done faster than me.
If someone does 20% more than me, they should be paid 20% more. I don't believe in pay brackets and crap like what Unions use, because I've seen what unions do to people. I knew a guy who'd been in a dry waller's union for 20 years, he hadn't had a single pay rise since he got put into a foreman role, and he'd seen a lot of guys his age be canned because their wage was ridiculously high and they were too old to do the heavy lifting.
I suspect (from nothing more than your writing skills) that you are quite a skilled tradesman.* Why don't construction companies hire strong but unskilled guys to haul stuff and let you be the guy who does the more technical stuff?
* People who write well generally think well. Construction, beyond brute strength, requires mechanical thought. Therefore,...
It all depends on the construction site. The companies that do building houses and small business offices generally do have a few really strong guys doing the heavy hauling, back in the UK where I worked you'd get a guy carrying like 200 lbs of brick up a ladder.
However, I specialized in the Handy-Man kind of work so there's too many different job sites or the companies too small to actually have a specific guy hauling stuff. So in the jobs I worked you had to be both smart and strong. Electrical, plumbing and gas can be extremely complicated to put in, especially when you're modifying a house that's over 50 years old.
(At 17 I started working as a reviewer and the editor considered me the best they had. So I believe I have a high aptitude for writing. I've had a couple short stories published, and right now I'm working on a novel.)
You're just using word play to try to validate your point. Any salary discussion is going to go straight to the employee's performance.
The problem with this system is it takes the focus off that individual employees performance and puts it on the rest of the office's performance. Suddenly you aren't talking about how good that employee is you're talking about why Joe in the next cubicle over rates higher than him.
That's exactly where you don't want salary negotiations to go.
It not word play; there's a difference between "You aren't paying me enough for the work I'm doing" and "I don't belong at the level you say I do."
The former is open to much more interpretation than the latter, being that the latter has a documented set of predefined expectations, such as years of experience and daily duties. It's hard to be obscure about whether or not you lead a project, or whether or not you have 5 years experience. Joe in the next cubicle rates higher because he is either your - or your colleagues' - senior. This is self evident.
What you are left doing is saying "I lead my project better than Joe leads his project." This isn't a salary conversation, it's an annual bonus conversation. Annual bonuses are designed to measure performance in most cases.
I still don’t agree in that I simply don’t think the real world works that way.
On the Joel point: Joel runs a small-ish company and it’s become obvious in several HN threads that his employees are people who essentially worship the ground he walks on. That’s great for them, for him, and for that company and I’m happy for them all. But what that distinction means is that sometimes Joel’s advice doesn’t apply to the rest of the world outside his little utopia.
In my experience living in the rest of the world people don’t respond well to "your salary has nothing to do with your performance but you MIGHT get more 9 months from now." A bonus is great but it’s just that: a bonus. Salary is where employees are paid for their services and that’s where they expect to be rewarded for their performance.
Again, hopefully your employees are well off enough that they couldn’t care less about waiting 9 months. But in my experience the "wait for your annual bonus line" usually leads people to look for companies that pay their employees based on their performance.
P.S. You still didn't address my central point which is the fact that this method puts the focus on other people's performance in relation to the employee rather than the employee's performance alone
I have no comment whatsoever on Joel, his company, or his products. I don't know him, anyone in his company and don't use his products. I've taken issue with him previously on this forum.
However, I think you're misunderstanding a Base Salary with a Performance Bonus - which may or may not be annual - and the notion of Total Compensation.
In most companies, including mine, the idea is a that a salary is what you earn in exchange for doing your job to the minimum expected standard. The only reason these salaries aren't equivalent to legislated minimum wage across the board is that there are market forces involved that companies must work within. Joel is simply stating that in most organizations, there is a drive to push this salary as low as possible, and that this creates problems. I see nothing in mine or other's experience that is contrary to that point.
His solution is to basically simplify this equation, and publish it. That's the exact same thing that every union workforce has. The only issue he has with this is that if he miscalculates the market value of a certain level, he'll basically create the conditions for a strike. He recognizes this and explains his workarounds in the article. He also retains the notion of a bonus to help him recognize those he chooses to.
Salary is where employees are paid for their services and that’s where they expect to be rewarded for their performance.
This is clearly where your opinion differs, but this attitude creates the entire situation that Joel is commenting on.
I'm quite sure that the bonus at fog creek is not based on performance... It is based on profit (he does say that he has a profit sharing plan) but joel is against giving bonuses as incentive...
see http://www.joelonsoftware.com/articles/fog0000000070.html
Compared to the situation at my old company where salary was determined by how well you negociated, I'm quite sure I prefer joel's version. And when reading the ladder http://joelonsoftware.com/articles/ladder.html, it's clear enough so that any disagreement are probably going to be minor (since he's using the average of skill and scope in a lot of cases a one point difference in the skill or scope won't make any difference to the level)
If the performance bonus isn't annual than it's no different than a salary. In terms of an employee asking for a raise, how is asking for a larger performance bonus any different than asking for a higher salary?
You've just categorized the money differently in your head at that point.
If the performance bonus isn't annual than it's no different than a salary.
Of course it is, in that it's the result of an outcome and isn't guaranteed.
There is no company in the world that would mind you asking for a larger bonus. After a certain level however, you need to qualify the value you bring to earn that bonus. If I said to my employer: "If I bring in 200 new users this year, will you give me X more (where X is a percentage of the revenue generated by 200 users) in bonus?"; No competent manager would deny you this.
I suppose I can see the distinction in that this allows a company to tailor how much each person makes dependent on their performance in a given time frame.
That said I certainly wouldn't want to work for that company. Basically what they're saying is "we don't trust you the employee to continue to perform optimally so we're going to make you prove yourself each and every (whatever the time period is) to get a bonus" while the employee has to wonder how much their pay check will be every month (or whatever)
I'm sorry Tom, but in 95% of for-profit enterprise, you are continually measured against the bottom line, and your performance is constantly evaluated.
Most companies do this via your salary in a secretive ranking ladder - which is exactly what Joel is speaking against.
I think you agree with him much more than you you think you do. :-)
Yeah, I agree - the negotiation comes in when someone says they are rated too low, asks to talk to the review comittee about it, and leaves when they don't get it.
That said, I think this system is probably superior to most.
If they think they should be rated higher, then they should give the reasons why.
If those reasons are valid, then the employer can change the rating. If the employer is unwilling to change it for irrational reasons, that means that the rating system is not objective and the system is undermined and has lost its meaning. This becomes a separate problem entirely.
If the reasons for valuing themselves more are not valid or applicable to the company, then the employee is over valuing themselves in this context. If the employee is not willing to accept this, then they have the option of declining the offer and moving on.
Exactly - I do not think the coding Mark Zuckerburg or Larry Page/Sergey Brin did while they were students was trivial + it lead to two "game changing" companies, so to speak.
Anyway, working for a large corporation with an opaque salary system was an exercise in frustration. I moved to SF and took a job at a midsize company making what I thought was a very decent salary. I was just out of college and making more than my parents, so I thought I was doing really well. However, after a bit of digging and research, I discovered that the "going rate" for my position was actually 20 - 30% higher. Ugh.
I can totally see the appeal of a transparent salary system, especially when you throw profit-sharing into the mix.