Open Source Project – View 689 Salaries Posted on Hacker News, Share Yours
blog.step.com
blog.step.com
It _is_ important to note that the average retention at Amazon (even for excellent people) is ~50% for 1 year, and ~20% or lower for the 2nd year.
Therefore, these comparisons should really take that into account. Amazon MAY look higher than Microsoft (for example) but because you only see 20% of your total vest the first two years, and everyone in the company makes a max of $160k with anything else bringing you up to market rate being a yearly "bonus", you likely won't see the majority of your earnings. </Former Amazon>
Day 1: 1000 employees Day 365: 500 employees Day 730: 200 employees.
It's almost impossible to throw a dart at this list and not hit at least 5-10 - http://www.geekwire.com/geekwire-200/
Your comment on "retention" is factually incorrect. I'll presume your comments are in relation to "tech roles" such as SDE SDM TPM etc.
First the number of Amazon hires in "tech roles" increases by ~50% per year. i.e., in year 1 there are 1,000 SDEs hired, year 2 is 1,500, year 3 is 2,250, year 4 is 3,375 and so on.
"Attrition" represents the number of individuals who leave for any reason. This may be because the employee terminates employment or the company terminates employment. Attrition levels are comparable to the rest of the tech industry, ~15% of the current population per year.
"Tenure" is the length of employment. Because hiring greatly exceeds attrition you should expect a relatively low average and median retention. As I recall Amazon is approximately 12 months median tenure. This is comparable to Google at ~13 months, for example.
Lastly the employees current tenure does not have a substantial impact on the probability of their attrition. Indeed the average tenure at exit is somewhere around 3 years. Which, again, is roughly in line industry standards.
In short, the average tenure at Amazon is low because they hire a lot. And while retention is not "good" it is a far cry from what you've represented.
Edit: The gross growth rate is actually above 50%. I've attempted to simplify the values where it does not make a material difference to the refutation.
I disagree.
> First the number of Amazon hires in "tech roles" increases by ~50% per year. i.e., in year 1 there are 1,000 SDEs hired, year 2 is 1,500, year 3 is 2,250, year 4 is 3,375 and so on.
This could not possibly be true. If it was, they would be hiring ~1.5M technical people a year (since they've been in business since 1997). Perhaps you meant since 2005, but that would still represent 58k people per year hired. According to this article (http://www.geekwire.com/2015/huge-growth-amazon-reaches-2224...), they only have 24k in the state of Washington, where the vast number of people are. So nothing about this statement is true.
> "Attrition" represents the number of individuals who leave for any reason. This may be because the employee terminates employment or the company terminates employment. Attrition levels are comparable to the rest of the tech industry, ~15% of the current population per year.
Can you show me that data? about 15% per year? I've seen vastly different numbers internally, but would prefer not be sued in sharing them.
> "Tenure" is the length of employment. Because hiring greatly exceeds attrition you should expect a relatively low average and median retention. As I recall Amazon is approximately 12 months median tenure. This is comparable to Google at ~13 months, for example.
Correct, but these numbers are heavily obfuscated due to part time hires and vendors.
> Lastly the employees current tenure does not have a substantial impact on the probability of their attrition. Indeed the average tenure at exit is somewhere around 3 years.
Again, show me the data. I've seen the internal stuff. If you'd like not to believe me, feel free; I know I don't believe you.
I'll again state that I am referring to "tech role" workers commonly represented by an "SDE"; not vendors, contractors, hourly, warehouse associates, etc. I am also referring to this population across all of Amazon; Consumer ("Retail"), AWS, and "Digital" aka "Kindle". I make no specific claims to particular internal organizations which may or may not represent the average.
I don't think that anyone is going to repeat or cite a companies internal data. Maybe we can use public data and some simple deductions to arrive at the most plausible explanation.
Your linked "geekwire.com" article is actually an illustration of the SEC 10-K filings. Under that documents definition of "Employees" it specifically terms this to be "full-time and part-time employees." Contractors, vendors, and temp workers are _not_ included in those numbers: "Additionally, we utilize independent contractors and temporary personnel to supplement our workforce." Unfortunately, that does coalesce "tech roles" with distribution ("warehouse") facility workers etc.
Additionally by reading these 10-Ks you'll note a section on "Stock-based compensation." As many have noted RSUs are a significant portion of Amazon compensation. From 2008 annually this is $275M, $341M, $424M, $557M, $833M, $1,100M, $1,500M, & $2,100M through 2016 Q1. Regardless of employee count the cost of issuing RSUs increased by 25-40% per year, with an increase in rate during recent years.
As you, and others, have noted the Amazon compensation package weights the initial RSU vesting schedule to years 3 & 4 of employment. I believe the typical structure is that only 5% of grants vest after the first 12 months, and an additional 15% at 24 months.
Let us use the public data which shows a distinct increase in employee count and compensation costs starting around 2011. Suppose again the company might have had 5,000 employees in tech roles at that time. Using my posited net growth of ~50% we could expect ~25,000 employees after 4 years, i.e. in 2015. Which happens to align pretty well to your citation of 24,000 in WA as of mid 2015. Don't forget that Amazon is a global company, so while a majority of tech roles may be WA based, it is certainly not all. Feel free to decrease the rate to 40% net, it merely changes the target dates by ~1 year.
We can also compare the public data on employee count and stock based compensation. Using the hypothetical ~25,000 tech role employees for ~2015 and $2,100M of stock we arrive at an average of ~$84,000 vested RSUs per tech role employee. Of course RSUs arent a uniform distribution, but that seems to map to self reported salary data. Adjusting for the employee count at 2015 Q2 and the stock based comp cost at 2016 Q1 further improves the fit.
From your other comment you appear to posit that employee attrition is approximately exponential decay each year. I believe you have also stated that the average tenure at exit time is 1 year. In that case employees should have only vested ~5% of their total RSU grants at time of exit, forfeiting 95% on average. How do we reconcile that with the public data showing a ~40% annual increase in the cost of stock based compensation?
Further your exponential attrition should lead to asymptotic employee counts unless offset. I do not believe anyone is claiming a net decrease in Amazon tech role employees, most cite a rapid increase in active employees. What hiring rate do you posit to support both your claimed 50% attrition rates and the significant net increased employee counts?
PS: AMZN 10-K filings for your perusal http://phx.corporate-ir.net/phoenix.zhtml?c=97664&p=irol-sec...
I'm saying unless they had 10 employees in 2000 (wrong) or only started the 50% hiring last year (so we have no longitudinal data to go on), there is no number where that level of exponential growth of employees wouldn't result in either a) a company that was bigger than Walmart in employees in just a few years or b) MASSIVE firings. Maybe that's what you're saying? They hire that many people and remove them from the company that fast? I know the removal from the company is certainly correct.
As far as the increase in costs of RSUs, there is a perfectly alternative solutions to your proposal: their stock has increased by 10x in 10 years. So in that way, they're EXACTLY keeping track (https://finance.yahoo.com/echarts?s=amzn+Interactive#{"range...), and not increasing people collecting RSUs at all. You'll note that you list amount of money required for the RSUs, not number of shares. (This is, by the way, another way that Amazon misleads their employees - your yearly cash bonus actually goes down when the value of the stock increases).
So by that logic, they have, in fact, not kept pace with the rise of their stock. New employees are receiving substantially less new grants than older ones (bordering on zero, which can't be true), or they're being removed before they vest.
Further, the 24k in Washington also include many thousands of non-tech roles (Amazon Fresh warehouse in Redmond, Customer Support in Bellingham, some portion of the more than 2500 sales people for AWS, etc etc), so your math does not add up for adding new 25k tech roles, even if we multiply by 40% as a reduction. Doing some approximations using LinkedIn (https://www.linkedin.com/vsearch/p?keywords=amazon&f_CC=1586...) they do appear to have ~24k employees in Washington, but only ~8k in engineering (they appear to add another 4k engineers in India).
Again, these are all public numbers, so they are inaccurate, but they should be good enough to substantially disprove your point. So, without question, your statement about increasing hiring by 50% y/y has no facts to back it up.
As far as active hiring, that is absolutely correct. They are extremely active in hiring - but they have to be because people leave so fast that unless they do, they'd be out of people.
EDIT:
After writing this, I was curious how the BLS calculates wages. Here is the answer from their FAQ: http://www.bls.gov/oes/oes_ques.htm#overview
The following are excluded from the collection of OES wage data:
Attendance bonuses
Back pay
Clothing allowances
Discount
Draw
Holiday bonus
Holiday premium pay
Jury duty pay
Meal and lodging payments
Merchandise discounts
Non-production bonuses
On-call pay
Overtime pay
Perquisites
Profit-sharing payments
Relocation allowances
Severance pay
Shift differentials
Stock bonuses
Tool/equipment allowances
Tuition repayment
Uniform allowance
Weekend premium pay
Year-end bonuses
I would argue that RSUs are rarely something the employee wants, compared to equivalent cash. At a big public company, you aren't likely to be able to affect the stock price with your personal performance, so it's not an incentive. You also likely can't predict whether the stock is likely to go up or down -- if you can, you're in the wrong job.
I would speculate that the appeal of RSUs to Google is as a hedge -- if Google's stock price goes way down, at least they'll get a break in employee compensation. Otherwise I honestly don't know what the point is. Just give people cash.
Startup equity is a completely different beast. You're getting a grant that is (in theory) worth $0, but you may actually be in a position to predict whether that startup is going to succeed, and even to influence the outcome. Maybe. That said, if the company doesn't tell you what percentage of the company you're getting, then it's a scam and you should treat it as $0.
(I'm a former Google employee and current startup founder.)
The other advantage is you can incentivize people to stay to at least the RSU cliff, which improves your retention and institutional knowledge.
eg Software Engineer from Atlantis earns 100 000 USD but pays 55% of tax so takes home only 45000 Software Engineer from Middle-Earth earns 60 000 USD but pays 10% tax and takes home 54000
etc
I mean iphone costs about the same anywhere in the world ( okay it's sometimes more usually not less (Brazil?)) so there's that...
In Taiwan, for example, I got a 3G phone plan for $40 month-to-month, unlimited data, and consistently ~20mbps download speeds.
In developing countries like Thailand or Central America, I can get a 3G card for 60 days and 1-2GB for less than $10. Speeds obviously aren't as good as Taiwan, but the basic access to Internet is way cheaper.
You might want to take a look at the Big Mac Index, which tries to give some measurement of how much your money is worth in another currency or country.
[0]http://www.numbeo.com/cost-of-living/compare_cities.jsp?coun...
That tax is buying us stuff.
Comparing pre tax income is more useful, by and large.
And how the buggery are you managing to spend £2300 a year on health insurance in the UK? That's crazy expensive. Ok, it's less than people in the US pay, but it's still mad for UK.
The health care expenditure in the US is the biggest in the world but that's including all the prices that insurance companies pay for treatment and services. If the US stops charging 3000$ for an MRI scan it's expenditure will go down drastically (an MRI costs about 300$ in most European countries) but it might not have any affect on what the individual ends up paying.
I suppose prices could have increased a fair bit since then. Also, I note you're in London. I'm in Scotland, so we actually have a different NHS, and it does generally get rated higher than NHS England.
I can compare public healthcare spending as that's the money that comes out of your taxes. Americans are taxed more for their public healthcare than we are in the UK for the NHS. They also pay more for private healthcare than we do in the UK. So your claim that it's more expensive here is simply wrong.
(EDIT: I'm certainly not saying this is a bad thing.)
Which makes it worth it, imho.
Except it isn't, anything you want to do there is ridiculously expensive
I think it's not true that the ACA caused that reduction (the recession is probably a much bigger factor), but it also didn't cause prices to increase faster than they were prior to the ACA.
That's great! I wouldn't way to be one of the few people able to afford healthcare in a world full of sick people who can't afford to get well...
Just want to point out that there is a good reason for the higher taxes.
For example, my first job was 25 days + 8 days of bank holidays and then you could earn up to 5 extra days if you stayed with the company long enough. Other jobs were all 22 + 8.
[1] https://www.gov.uk/holiday-entitlement-rights/entitlement
As a proportion of GDP, Government spending in the US is only a couple of percentage points lower than the UK. The US actually spends slightly more on public healthcare than the UK, 7.9% vs 7.3%.
Of course healthcare here is so horrifically inefficient that another 8.5% of GDP is spent privately...
We may pay more tax, but I'm glad we have the backup system that keeps you going when you're sick or unemployed.
American also has Social Security & 401k (pensions) and health care for the old (medicare) / poor (mediCAL, medicaid), and some social services (public school, various state programs in california)
I am less compassionate than the other folks in Europe and would really prefer a more capitalistic system at least in my country. But we Europeans are....ehhhh.
And expect corruption in the private system, too. There's no other explanation for $500 aspirins :-)
If you can get in the healthcare industry in the US you'll be a genius. Or probably a criminal, there's no legal way to get out of the current status quo without government intervention.
I like the free market, but you're oversimplifying reality.
Personally, I rather pay a 10% tax to cover healthcare rather than 20% of my salary to insurance premiums.
https://docs.google.com/a/hackbinary.com/spreadsheets/d/1aMx...
http://www.theguardian.com/news/datablog/2012/jun/30/healthc...
That said, one of them is the kind of person that I would expect to do really well in an Amazonian environment, and the other now works at Nintendo.
"Amazon's vesting schedule is 5%, 15%, 40%, and 40% over the 4 years"
This is a significant deviation from the industry standard, which is to have a one year cliff, and have stock vest month-to-month afterwards. What happens if you leave in the middle of year 3?
If you leave in year 3, you lose 80% of your signing grants.
What you call "signing bonus" I call "guaranteed raise if you stay"
and there's also refreshes that vest at (probably) a different time of half-year.
When I started, I expected that I'd be there a couple years then move on to Google, but I'm not even excited about talking to Google, Facebook, Uber, or the others at this point.
P.S. We're hiring on DynamoDB. We need for software developers and SRE-type folks. My e-mail is in my profile if you're interested in finding out more.
SDE3 is a quite senior role at Amazon. I don't know very many of them, and almost all of the ones I know are extremely bright. Seems like they're getting ripped off.
Additionally, last time I interviewed there they did not have remote work in any form. If you get a 2am page, you have to drive in to the office. (And you have to commute.)
Amazon generally has a pretty strong remote work culture - I can even get my ipad onto the VPN and SSH around to get my work done.
It's a truly horrible place to work and I would never ask my worst enemy to go work there. Engenders the worst in humanity (people review, anyone?), terrible technical architecture, many systems based solely on tribal knowledge that leaves the door every day, comp structure that is nearly outright lying... I could go on.
Some people say that it was just the group I worked in but I worked in retail (1 year), new businesses (1 year) and AWS (1.5 years). There's virtually no difference. I still have PTSD (or at least what feels like PTSD) about by 3.5 years there, and I'd give anything to get that time back.
BTW, I'm now at another 20k+ person company which is doing very well, and making 30% more than my total comp when I was there.
> I am a former Amazon employee. I worked there for exactly a year. I assume I was doing well, I was offered increase bonus and base after my initial year. https://news.ycombinator.com/item?id=10413966
Just to clarify, did you work at Amazon for 1 year or 3.5 years? It seems like these messages are contradictory, but perhaps I've misunderstood. In the message I'm replying to now, you wrote:
> I worked in retail (1 year), new businesses (1 year) and AWS (1.5 years)
In the previous comment, I was worried that people would know who I was if I had said where I worked (what groups and how long). The truth is I had worked in the group I moved to for exactly a year, but I had been at Amazon for 3.5 years.
The level of retribution at that company is substantial, and I was quiet worried they would hurt my ability to get a new job.
I have a nice new job now (I was still searching when I posted that), and am less worried. It's really not hard to figure out who people are on HN.
Example I witnessed during people review (obviously anonymized):
A: I think X is one of the best members of the team, he took a bunch of customer requirements and put out something super fast that addressed some customer needs. (Invent and Simplify, Bias for Action)
B: I disagree. His product didn't think about scenarios a, b and c [ed: these would be things that caused the product to slip a year, and would leave customers in pain during that time] and he did not investigate g, h and i [which would have taken 3 months to figure out, still with customers in pain]. I think he needs to be put on a PIP. (Dive Deep; Insist on the highest standards)
Yes, I'm highly biased here - this person was on my team, and I endorsed his plan, as did person B, until we got into People Review. One of the most brutal and subtle things about the entire process is the fact that you're consistently asked for negative feedback about EVERYONE... even if you don't have it.
I moved from that group (AWS) to new businesses shortly thereafter.
I don't see them as arrows to knock people down. I see them rather as, well, principles to use when reasoning about a situation. Giving a name to a concept is a good way to reason about it and discuss it. They are used when reasoning about performance, it's true, but it's used across the board: when justifying a promotion, when praising a high-performer who excels within their level, when identifying areas of growth, as well as while discussing job performance with people who are doing less well.
Performance and leadership in a creative job like software engineering cannot be perfectly objective (it's not like an assembly line where you process N units per hour), but the LPs provide some objective concepts and terminology to apply while discussing performance; they break down performance into dimensions that can be discussed individually in the context of examples, which is an advantage over just vaguely discussing things at a high level. For example, "Bob found a creative way to pack more software onto existing servers without harming performance (Frugality)" or "Sally spent two hours on the phone helping our biggest customer (Customer Obsession)" or "After causing an outage, Joe wasn't able to identify anything he'd do differently next time. I'm concerned that he's not vocally self-critical." I'd much rather have a discussion of my performance or leadership in concrete terms like these, rather than vague ones.
You could also try comparing with visadoor.com or another source of H1B data for the same companies to try to see where they fit in. There may be a systematic difference between H1B salaries and the larger employee base at these companies, but at least you're not limited by self-reporting.
There is a market for doing user submitted comp analysis, but the real value is building trust, which takes a very long time and is expensive.
Surely scalping a ton of mid level engineers at major tech firms for sharing their salaries online will fix that pesky problem of soaring profits and thick margins...
IMO sharing salary data is the perfect first step for some sort of software engineers association/guild/professional body.
And perhaps more relevant to the thread, I know two types of people who talk about their salary, people who make a lot and people who make a little. Their is just so much bias present to me to give any credibility to these lists. Take the google spreadsheet where the employee was fired, do you think the engineers who thought they did better then their peers in negotiations listed their wages?
At it's simplest level I think something like the SAG or profession sports players associations. Kept out of the way, stars can rise, and there if you need them. It'd be useful to be able to have an organization that I can say "Hey, I'm interviewing at Github, based on your verified compensation data that you keep, what can I expect?"
The fact is, our employers share and buy each others compensation data similarly for use in negotiations and when setting compensation. It only makes sense that the parties on the other side of the table do the same thing.
The taxes on equity are a huge PITA when equity is a large fraction of your salary. This is because they withhold at only 25%, even if your tax bracket is higher. This means you have to file quarterly estimated tax payments or over-withhold from your salary.
I much prefer Netflix, where I just get salary. I might even be able to do my own taxes again!
But it's also a tax hassle...though it's hard to complain about it.
Or a hybrid (pay a reasonable amount in estimated taxes but don't put any serious effort into getting it exact).
Sheesh!
I had been working remotely for a long time for a small company, and making what I considered to be good money, considering that I lived in a fly-over state. When I was hired at Google, it more than doubled my salary after a while, when stock grants began to vest.
I now work for Netflix, and I am making even more than I was at Google. To this day, I cannot believe I am paid so much.
All-in-all, I work less, and have a far better work-life balance than I did a few years ago, when working for the small company. In that company, I had sole responsibility for an entire product area, and often worked over 80 hours per week.
As an afterthought, I actually worked far less at Google. Probably about 45 hours a week. It was like a vacation. Being a cog in a machine and not on-call 24/7 was refreshing and enormously freeing after years of constant pressure.
Why would that be? I suspect that's more because the pool of people I know is biased than because the pool that posted is biased -- some orgs in Microsoft will match offers from other companies even though average compensation is relatively low, so if you're new and you know a lot other people who are new, the sample of people you know is probably biased high.
To be clear, I think that's bad both for Microsoft and for Microsoft employees. I'm just trying to figure why the Microsoft data is anomalous relative to the numbers people have personally told me. The reason it's bad for employees is obvious. The reason it's bad for the company is that it's pretty common for people who have been here a long time to shop around, find out that they can make much more elsewhere, and then get a matching offer from MS. Once word gets around that the best way to get a fair raise is to interview elsewhere, people start interviewing elsewhere, and some of the people who interview are going to leave, even if that wasn't their original intent.
Isn't that true pretty much everywhere?
No, $172,000 is the median for year 1. Even $200k+ total comp at year 1 is typical for Google in the bay.
The significant increases in bonus pay for higher levels reflect stock vesting for year 2/3/4/5 employees.
Edit: "Google" in the bay
At any rate, I've received more than my fair share of $200k+ 1st year total comp offers for _start ups_ in SF for positions that only wanted 2-3 years experience. So, I wouldn't be so surprised.
Myself, I make $160k base salary at a small company in Palo Alto (was offered that at 2.5 years of experience), and know I could've gotten a lot more elsewhere.
I'm going to say that's a huge stretch.
With insane variances between the teams offcourse.
It's harder to see who will be amongst the best engineers early on in someone's career. Each person develops their abilities at different speeds & times in their lives.
I'd also like to see a breakdown by market, as I have been led to believe that my salary is only slightly below market for Dallas, but the median compensation numbers posted in this article for my experience level are three times what I make now, and it's depressing.
For anecdote, approximately 66% of my total compensation is in base salary. When I first started working 5 years ago, it was 92%.
These numbers are probably skewed somewhat by Bayarea salaries, which have to compensate a little for the high cost of living (compare for example Facebook/Google to Microsoft/Amazon, which are primarily Seattle-area employers). In general, I've found the Bayarea salary bump doesn't overcome the Bayarea cost-of-living bump. So these salaries are only higher on paper.
In NYC and SF, there's high demand for and short supply of developers, which drives up compensation. Companies in these cities will recruit from across the country to get talent. Cities like Dallas though also have a much lower cost of living, so that should be taken into account. There was a Hacker News thread about this at one point.
We don't yet have enough data on each market, but we plan on breaking out the analysis by city once we get that data.
For example, if you knew four years ago that you would receive 30 GOOG stock in 2016, you would assume in 2012 that would mean a $10k bonus. But now the stock has gone from $325 to $754, the bonus is actually $22k.
If you work for a company with an increasing stock price, then stock based compensation can also create the illusion that your "salary" is increasing each year, even if you get no promotion.
With that said, Google did have this odd thing where you could do an in-plan Roth conversion, and put post-tax dollars into your 401K. I did this in 2013 or 2014, and I think Google was one of the few companies to offer this at the time.
One interesting area of differentiation is health care. When I left Google in 2015, they were cutting health benefits to avoid Cadillac plan taxes. Lots of people were saying that Facebook had better plans.
Frugal FTW
In reality, Google's hiring process selects for people who can follow instructions and are happy trading self-sufficiency and independence for the social validation that comes with working for such an "elite" company. Google engineers are some of the most sheep-like creatures on the planet.
We detached this subthread from https://news.ycombinator.com/item?id=11535180 and marked it off-topic.
I don't think ego is really the only thing that drives pay distinction. I think that many people consider that different individuals contribute differently and that those who contribute more value should get more pay. Isn't that realization independent of ego?
Even for a married couple earning 500k in high-income-tax California, effective tax rate (as opposed to marginal tax rate) is only 30-35%, and that includes the employee portion of Medicare/social security (the latter of which cuts out around 110k).
The exact effective tax rate will vary mostly with your mortgage interest deduction.
Massive assumption you are making regarding home ownership.
5% higher for $500k