Jeff Bezos' annual shareholders letter | Hacker News Reader

Jeff Bezos' annual shareholders letter

sec.gov

254 pointskrelian194 comments

Some of my favorite nuggets from this letter: "[H]igh standards are teachable...High standards are contagious" "You can consider yourself a person of high standards in general and still have debilitating blind spots." "To achieve high standards...you need to form and proactively communicate realistic beliefs about how hard something is going to be..." "[A] culture of high standards is protective of all the 'invisible' but crucial work that goes on in every company...doing that work well is its own reward – it’s part of what it means to be a professional."

If only Amazon as a whole had a culture of high standards...

What you and lots of other people (including me, until very recently) tend to miss is that high standards are one thing. High standards at scale are another.

What precedent do we have for standards at such a large scale? How many examples of organizations that have achieved that size do we even have and what do standards across those orgs look like?

This applies to software quality at unprecedented scale as well.

I don't believe that high standards inhere in small scale. I have also not read or heard anyone articulate a reasonable explanation for why that is necessarily the case. Figuring out how to operate a business with high standards at scale is arguably the primary job of anyone who operates any business at scale.

Let me draw an analogy that might be helpful.

Let's look at manufacturing yield.

It's easy to create a process that hits 99.99% yield when the absolute value of complex goods being produced is 10 items.

Can you maintain 99.99% yield when producing 100,000,000 of those items? Almost certainly not. Does hitting 80% yield mean you have lowered your standards when every other manufacturer of similar scale can only hit 60%? No it doesn't. It means you have maintained high standards, it's just the context around you has changed and so the measurement of high standards has changed along with it (as it should).

That's the core of what I was getting at.

I couldn't disagree more. Standards are not relative. What you're describing in your example is an instance of lowering a standard. The standard does not change because one's tolerance for meeting it is lowered. You're simply accepting that you've failed to meet the standard. You can say whatever you like to excuse it, but it's an excuse nonetheless.

Is stuff like "your workers shouldn't collapse of exhaustion because you don't want to air condition your warehouses" something so difficult to achieve at scale? Is it even a particularly high standard?

Maybe before talking about high standards, we should start with low standards.

People on HN like to talk trash about Amazon and that's fine. All that I ask is that you don't do an AMZN_warehouse_worker == AMZN_SDE / AMZN_SE kind of argument. Warehouse work is boring, it's low paid, at the end of the day you're tired and you're dirty, and you didn't make a lot. During the winter it's cold and during the summer it's going to be hot - logistics companies heat warehouses just enough to keep product from freezing. The closest thing to a computer is a hand-held wireless barcode scanner. Them's the breaks of being a box b for any large logistics / supply chain company. UPS, FedEX, DHL, or UP, none of them are tech companies so we don't talk about them but the reality is supply chain is very price sensitive, the industry is low margin and competitive, it's not romantic, and it's classist. No one turns down Stanford to work the graveyard shift unloading tires. The people there will be randomly drug tested and they will be checked on occasion when they're leaving. Argue there is honor to any job done well and I'll agree with you but stop with this Amazon Logistics goes out of it's way to be cruel narrative.

While I agree with your underlying point, Amazon does deserve special attention because it tries to win hearts and minds, in fact it's the backbone of their branding operation. This letter is a perfect example.

FedEx does not have the hearts and minds of customers, everybody expects their employees to piss in bottles and would switch to the first lower cost provider available. Amazon is turning into a larger and much more ruthless FedEx, yet somehow gets a free pass and is allowed to spout this bullshit with a straight face about how 500.000 bottle-wielding Amazonians are making the world a better place.

> Amazon is turning into a larger and more ruthless FedEx, yet somehow gets a free pass

They're not getting a free pass. I've been reading about Amazon abuse stories non-stop for ten years in the mainstream media, as it pertains to their treatment of labor in logistics.

We're discussing this right now on HN, and every big story about this sort of thing re Amazon shows up on HN, precisely because they have not been getting a free pass. It's closer to universally understood that Amazon doesn't take good care of its logistics workers (and it may be wider than that based on some past stories). Every time something happens in that segment of Amazon's business, it makes the NY Times and the frontpage of all US media (just as the bottle pissing did, or the building air conditioning story did before that, or the employee cancer story).

That looks like a whole load of horseshit to attempt to justify the unjustifiable. Reaping record-breaking profits while workers are mistreated and paid a pittance is not a "problem of scale" or whatever it is.

It does. The global amazon experience is imperfect, but amazing compared to what we had before it, unmached by competition and life changing enough that we use it for everything.

You can't do that by being average.

Don't confuse high standards and perfection.

Getting counterfeit merchandise from the factory to your doorstep faster than ever before.

They are amazing In some areas, but have very low standards in others.

> Getting counterfeit merchandise from the factory to your doorstep faster than ever before.

Yep. I get bite at least once a year with that one.

Or if you're in Austria, you are conditioned to accept that delivery dates offered are 1-2 days off. The warning about "purchase within the next xx hours to receive it in 2 days" is absolute garbage. And the fact that I can select whether it's delivered on Saturday is bullshit, because no courier service here ever delivers anything on Saturday. Yet years later, my Prime packages are still being offered to be delivered on Saturday. I guess I shouldn't complain though, their estimated delivery times are still on par with every other eShop.

I find it very hard to communicate about how hard things are going to be (heh). Whether a task is hard depends a lot on the ability of the person doing it. In addition, "hard" can have multiple dimensions: the sheer number of hours, the monotony, or the lack of purpose.

more succinctly: thinking something is easy is the enemy of high-standards. I really like that, and it's something I have to remind myself from.

Along the lines of high standards, something I've realized is that, for people who have high personal standards, working in an environment of low expectations is actually more stressful than working in an environment of high expectations.

> for people who have high personal standards, working in an environment of low expectations is actually more stressful than working in an environment of high expectations.

I can't stress this enough. I was once chided for checking for nulls in places where nulls can clearly show up (ie, not repeated checks to the same value or anything like that) because "You Ain't Gonna Need It".

Three days later, the application is crashing on end users' machines because of null pointer exceptions, generating chaos for a day while I try to wrangle the (lack of) a build/deploy process[0] to deploy an emergency patch.

[0]: You guessed it, YAGNI. Visual Studio has a build option in the menu for a reason, you don't need to overcomplicate things to start the project. Or something like that.

One needs to be careful with high standards.

You go higher and higher and before you know it you're peeing in bottles to reach your productivity goals.

Snarky comments are often amusing and are always self-satisfying, but they degrade the conversation. (When I make comments like this on HN, they always get downvoted. I'm trying to do better.)

I agree with your statement, but I don't see how it applies here.

I did not consider the comment amusing. I do think putting the Bezos' ideas about high standards in the sobering context that Amazons provides for most of its workforce is right on point.

Satire is a perfectly valid form of communication.

Commenting here is increasingly feeling like a policed academic conference, rather than a "tech salon" which seems closer to its ethos.

This comment in no-way degraded the conversation. It added important context in a humorous manner.

How is it even snarky? There are workers peeing in bottles, probably right now, as a direct result of Bezos' obsession with optimizing productivity metrics at all costs.

Now the comment makes more sense. Guessing some here were not aware of that story. http://www.businessinsider.com/amazon-warehouse-workers-have...

I agree that this comment is frivolous based on the assumption that reading HN comments is about gleaning knowledge from the community. I love a great contextual joke, but not in the wrong place.

If you hate Amazon's practices there is more reason to study them because of their success. Every successful company that's done a pereceived evil has also tapped into a market advantage that led them to their success.

If you hate Amazon learn why they succeed and build better. Dismissing everything they do well because of what they've done wrong leaves the advantage to everyone who sees nothing wrong in draconian business practices.

> I love a great contextual joke, but not in the wrong place.

The joke brings to light the quite real and currently newsworthy situation resulting from the attitude expressed above.

These are things that need to be talked about, here is the right place to talk about them, and humor is often an effective way to highlight hypocrisy and abuse.

I want to clarify why I characterized this comment as snarky. The comment implies that high standards at Amazon has created a corporate culture of employee abuse. This makes no sense -- high standards and employee abuse are orthogonal.

Employee abuse is morally wrong and almost certainly illegal. It needs to be exposed and challenged whenever it occurs. But be careful about taking stories published in Business Insider (or any similar online publication) at face value. Consider if this is a reputable source before repeating their claims.

This needs to be tempered hard with the "realistic beliefs".

High standards can lead to heavy anxitety which will kill productivity.

This happened to me.

Bezos addresses this in the letter, mentioning that leaders have to accurately teach the "scope" of the high standards, where scope is a measure of the time required to achieve the set standards.

High standards? For what? I look at amazon and I see a sleazy money-making machine. Nothing about it screams high standards.

Amazon's performance implies high standards internally perhaps?

High standards making money. That may be the reason Bezos is the world's richest person.

"We established long-term relationships with many important strategic partners, including America Online, Yahoo!, Excite, Netscape, GeoCities, AltaVista, @Home, and Prodigy."

From the year one letter. It's incredible what Amazon has done when you realize who their peers were when they started.

What's really fascinating is hearing about people that work there now that used to work with them in the early stages. I talked to a dude who was in charge of a database tool called Tango back in the day who was worried they wouldn't be able to make the yearly license cost and now he works for them.

Technology only grows more fascinating as the titans rise and fall.

"Alexa – Customer embrace of Alexa continues, with Alexa-enabled devices among the best-selling items across all of Amazon. We’re seeing extremely strong adoption by other companies and developers that want to create their own experiences with Alexa.

There are now more than 30,000 skills for Alexa from outside developers, and customers can control more than 4,000 smart home devices from 1,200 unique brands with Alexa."

Meanwhile, Siri recently was able to tell me cricket scores :/

This also had me shaking my head. Siri can’t even play Spotify. Oof.

Amazon can't even sell Chromecast dongles. Oof. Its as if both companies have parts of their ecosystem they wish to protect.

The difference is that Amazon could sell Chromecast dongles if they wanted; Siri can't do anything

I'd love it if a colossal waste of resources on shitty voice search was the downfall of giants.

I don't get the infatuation with voice skills and the associated tradeoffs.

Despite the issues, I can definitely argue that using alexa has made my life more convenient. A lot of the skills are still awkward to use but I can 100% say that voice control has taken a lot of friction out of certain parts of my life.

i.e. music, news, the time, etc while getting ready in the morning or taking a shower.

Though it may not be as useful to you or the demographic to which you belong - there are plenty of households who enjoy the device multiple times a day, and multiple people in the same household.

This is the first time in awhile Amazon has disclosed the number of Amazon Prime users. ("exceeded" 100 million paid Prime members globally)

for scale, there are estimated to be 326 million US citizens and 126 million US households.

Amazon never discloses numbers - the exact quote is "over 100 million".

As a shareholder, shouldn't you be entitled to information like that?

You're entitled to know how much they make from Prime, but not the business specifics (you can't extract information from the profit, either, as not all prime costs the same - there is student membership, month-to-month, yearly, etc)

> You're entitled to know how much they make from Prime

Do you? I don’t think they disclose that.

No, as a random shareholder you're not automatically entitled to all the details.

You are, however, entitled to the following three things:

1) Whatever information they provide has to be actually true to the best of their knowledge; so they can be vague or secretive, but "more than 100 million" has to literally mean that;

2) There's a set of core information that you are entitled to, but it doesn't go into as much detail;

3) Shareholders as a whole are entitled to whatever information they want; they can elect a board that will ensure that as much detail as you want gets publicized. But it's not likely that most shareholders in a public company will consider it a worthwhile decision.

Also, material information must be disclosed to all shareholders at the same time.

Look at Apple as another example. They disclose the number of iPhones, iPads and Macs they sell, but they don't disclose:

- how much they make from Beats even though it was their largest acquisition.

- the number of watches or AppleTVs they sell. They are both assumed to be 1 billion dollar businesses.

- they say they have over 38 million subscribers to Apple Music but never give an exact number. Even though Apple Music is a $4 billion+ a year business.

Etc.

Good catch on the exact quote: fixed.

Based on the price of prime, they’re pulling in like $6-9 billion per year on that. That’s crazy. Of course that goes back into shipping costs but still.

Has the number of things eligible for free prime shipping been dropping or increasing?

There is a little known feature of prime: you can read a book (from a selection) for free each 30 days. Theses books cost usually about 4€.

12 * 4€ = 48€. This little offer (which is usually not even advertised as a prime feature) alone almost justifies the cost of a prime membership. Prime is a cost-center for amazon. It's not meant to make money. It is a customer loyality thing.

I've been a prime member for years and didn't know this.... Thank you very much, ill be sure to nickle and dime Amazon for all the value it offers.

"One thing I love about customers is that they are divinely discontent" (my emphasis)

OK Mr Bezos, I have to give you points for that one. It's a retailer's philosophy in two words. I could imagine that coming from Jack Cohen or Terence Conran.

To err is human, to expect better divine.

I thought this letter provided many a great lessons, up until "we write narratively structured six-page memos."

Wow, can any Amazonians speak to this? Sounds like a huge time sink to be writing a 6 page memo instead of preparing a few slides (or writing a 1 page technical design doc)

I don't work at Amazon, but narrative memos probably encourage better thinking and communication. PowerPoint often amounts to hand-waving assertions delivered so fast you don't have time to evaluate them.

https://www.linkedin.com/pulse/beauty-amazons-6-pager-brad-p... explains it well.

Yes, we do write six-page memos. It's not just leadership, even on the engineering side, design documents are six pagers. Contrary to your initial reaction, six-pagers are very effective for both parties—the writer and the reader. You're right in that writing narratives as opposed to making slides takes longer, but that's not inherently a bad thing. Writing a narrative helps me articulate my design well. If I cannot communicate my design via a narrative, it means there are gaps in my understanding/formulation of the same. For the readers, it gives them a coherent understanding of the problem, approach, pros and cons, along with an appendix of material. This helps them to attack the problem with almost as good an understanding of it as the writer—sometimes better, because they come in with some prior knowledge. The writing exercise takes a few months to get good at, but once you're comfortable writing documents, you retain that skill forever. I'm not kidding when I say this practice is something that'll make me stick with Amazon for a long time.

Can you talk about the format of this doc? I did some googling but came up with nothing / a bunch of fluff pieces.

There's no format as it is; whatever suits the occasion I guess. If you're presenting an engineering design, it is expected that your document conveys the what, the why, and the how—ideally many hows along with pros and cons for each. A good narrative would be data-driven, and not riddled with intangibles.

Ah OK. Basically, a written doc with sentences and paragraphs as opposed to a PPT or outline with bullet points then?

That's more or less it. There are training sessions you can take where people who are good writers, teach you how to be a better writer. And the meetings, since they revolve around the narratives, involve getting constructive feedback on your writing besides simply getting feedback on your design.

Yes

A common format is posed as a "Press Release" for the idea as if it existed, with quotes from customers — even if it's purely an internal service, the downstream teams are your customers

The documents are fixed length to 1, 2, or 6 pages for different purposes/stages, and reference data goes in appendices that don't count. It's also impossible to be hired as a senior contributor or any management role without submitting a writing sample, which gets reviewed in the hiring meeting along with everyone's interview feedback.

While I agree written communication skills are critical for senior roles, do you ever find this format stifling for more visual ideas and concepts?

Or are designs and comps just included in the appendix with a couple embedded highlight images?

They're either embedded in the text or as an appendix. Its not really a hard fast 6 pages exactly thing, more of a guideline. You want it to have depth but not be too long. Thats pretty much it. For more details: I'd say 1 pager is an idea. 2 pager is idea with the why and some details fleshed out. 6 pager is idea, why/how/people, design docs, stats/analysis and maybe even some tech choices.

I thought it was a good model look into and plan projects. IMO, programming should be / is only 20-30% of a softdev's job, the rest should be planning (and probably fixing bugs afterwards haha).

Source: Former Amazon dev.

I don't - there are plenty of ways to get visual ideas/concepts into the document. You can refer them to the appendix or embed it in. Appendixes are often quite large on complex topics.

The big benefit is that meetings start with people reading the documents and thinking about them - you'll see people making notes in the margins, etc. Only after everyone has had a chance to read the doc do you start discussing it, so you save a lot of time answering questions that are answered in the doc.

Now that you have acquired that skill, why don't you move elsewhere?

I meant the practice of writing docs is unique to Amazon. I would imagine it isn't the case at say Google or Facebook to spend the first 15-20 minutes of a meeting reading narratives.

By saying this will keep me at Amazon, I meant that I would like to continue doing the narrative style meetings for as long as possible. Also, there are plenty of other skills you gain by working on things at the scale AWS operates at. I don't believe I've learned all that I can, yet.

Amazon, as an organization, cares more about thoughtful, evidence-based argument than any company I've ever worked for and, I suspect, any company I will ever work for. Having left Amazon (twice!), this is the thing I miss the most.

We do "architecture docs" where I work. Although there really isn't as much of a narrative in them. It's mainly a summary, why we should do it, and a bunch of graphs/charts.

Honestly, I find it a bit arduous, although it's definitely necessary. I really just wish people just whiteboarded it all out and it was recorded.

"PowerPoint makes us stupid" -- James Mattis

(Source: https://www.nytimes.com/2010/04/27/world/27powerpoint.html)

P&G also does not do slides (most of the time). The typical format is a one-page memo for anything that requires a decision.

I used to work at Amazon as an engineer, and it is all about making meetings effective. Writing is an effective tool to thinking effectively. Six pages is actually about keeping things brief and to the point, and many people will cheat at make the font really small.

Current Amazonian, my opinions are my own.

I love the emphasis on writing these papers to drive a meeting, and 6 pages is an upper limit. I personally prefer trying to condense it to 1 page. I will sometimes need get Product, Tech, and Ops to agree on something, and I find writing the paper really helps me solidify my argument. If you're just trying to get your team and principal engineer to agree to a design, however, the 1 page technical doc you mentioned is sufficient. Full 6-pagers are more common when working across orgs, in my experience.

6 pages design doc is too short for a company with 1000+ engineers.

It is 1000% true that /High Standards/ are "teachable, and contagious".

But what a company, or indeed, society as a whole set those /Standards/ on matters so much much more than anything else.

Really well written and engaging as usual. If he ever writes a book I will buy it. His extreme ambition to build the world's most dominant company may get in the way of that, though.

And Bezos wept, for there were no more markets to disrupt.

Somewhat OT, but I can totally relate to the handstand anecdote. I've recently been learning to do handstands, and while I knew it was something I'd have to practice, I completely underestimated how hard they are. I mean, kids do them all the time, right? I'd say 6 months of daily practice is probably about right.

I've always been impressed by handbalancers, but now that I have some inkling of how hard it actually is, I'm totally blown away by it. Next time you see someone balancing their whole freaking body on one hand (something like 4-10x harder than a two-handed one according to estimates I've seen), stop and marvel for a while, for it is truly incredible.

100M prime customers making 5B purchases a year is $2 per purchase for 2 day shipping that costs amazon 5-8 dollars to ship. Considering amazon takes 25% cut from FBA products you can say any product sold for less than $15 is a cash negative sale for amazon.

Wait what? Amazon often packs multiple products into the same shipment. I assume each individual product is considered a purchase. And where do you get the $5-8 figure from? I can ship things for that much, so you’d have to assume Amazon gets a discount for massive volume.

Amazon charges 3P sellers for the shipping and my understanding is they're basically passing through their shipping costs.

For a typical $15 dollar product that's under a pound, Amazon currently charges $3.19 for shipping. Most of those can be shipped first class or even Amazon flex because it's close enough to the customer it can still get there in 2 days, and Amazon can easily pack and ship a light item for $3.

If it's over a pound that fee goes to $4.71 currently (in fact the exact cutoff is 12 oz for complicated reasons).

I very much doubt Amazon is losing money on any third party sales in general. The one thing they might be is when they provide Amazon funded discounts to customers, although the max they do is 8-9% and I've figured they're giving up referral fee profit rather than losing money.

Wish he addressed about inhuman working conditions in AMZ

Credit where it's due. Customers love Amazon and that is what really matters at the end of the day. The people they serve enjoy the service and products Amazon provides. Bezos has very successful worked his way up to become one of the top dogs of capitalism. 20 years is a considerable amount of time work towards a goal, but Bezos was patient and hardworking.

"Amazon was also just named the #1 business on LinkedIn’s 2018 Top Companies list, which ranks the most sought after places to work for professionals in the United States."

This statement/award strikes me as very far removed from reality.

I don't think so at all. Sure, Amazon has a certain reputation, especially in hacker-news circles. That said I think you're forgetting about the silent but huge majority of average joes who would kill for an Amazon job.

You need to qualify that. Most people would not kill to work in a warehouse or customer support, which is a huge portion of Amazon jobs.

This absolutely. While we tech scene folks love to consider ourselves "average joes," Amazon employs around 100,000 warehouse employees, and for them Amazon is decidedly NOT the #1 place people want to work.

It's a fine place to be a software engineer, so long as you enjoy the high stress/low perks part of the tech scene, but that's a very different, air conditioned environment.

I know someone who works as a low level manager at Amazon and I've seen salaries of the average Software developer - for both the salary is like $140-$160K. That's nothing to live in Seattle with the cost of living. You can make that in 1 of 20 other metro areas that I know of with a much lower cost of living and much lower stress. Even when considering the signing bonus and stock, I was never enticed to actually fly out there and interview even after I got through some preliminary phone screens.

You know median seattle family income is like 80k right? Less if you consider more of the metro area. That's the family income, so it includes two earner households. NYC is like 50k.

What does the average household income have to do with someone who is choosing between a software engineering job in Seattle and another area with a lower cost of living?

That it's absurd on its face to say that 160k is "nothing to live in Seattle with the cost of living". The vast majority of people live with much less. The rest of your comment was fine re: programmers have other options.

It came with the assumption that I wanted the same standard of living. But the truth applies the other way. You won't live in a neighborhood with top rated schools (they will be safe), but you can live in a nice neighborhood with a 70k household income. When I bought my first home in 2002, it was a new build and 2800 square feet for around $170K. I was making around 70K. Looking now, that same house I use to live in just sold for around the same price about a year or two ago. The neighborhood is a safe, quiet, place.

Commuting to downtown was only about a 30-35 minute drive.

I'm not sure what planet you live when $140k dollars is nothing. As a software developer you might be able to do better, but you are incredibly lucky. I assure you the median wage is significantly lower than that, and of course many people make minimum wage.

In context - $140K for an experienced software engineer in a major metropolitan city isn't outrageous. A "software architect" often does hands on coding. For reference:

https://www.matrixres.com/resources/salary-survey/

If I can make $140K in metro Atlanta, where a newly built 3000 square foot house can be had for less than $350K in a nice neighborhood, why would I want to live in Seattle?

Politics. Nicer weather. Better environment for women/girls in schools. :)

The weather point is, of course, tough. Seattle summers are bloody amazing. The winter sorta blows.

The weather in Atlanta isn't much better than Seattle, if you compare season to season.

Seattle has one bad season, one good season. Atlanta is the same.

It is overcast 8-9 months of the year in Seattle.

The weather in Atlanta is just more extreme, honestly. Which can matter a lot if you like hot weather. Out here in Seattle it is almost always cooler. Hardly ever actually "cold" though, in that I think Atlanta actually got more snow and I had my pipes freeze more there than I have out here.

Now, the winter is much darker. The shift in sunset and sunrise is much more pronounced. This is somewhat offset by how much bloody sun we get in the summer. Which is truly insane. Sun is up by 4:30 most mornings, and you can see outside till near 10 most evenings. (Sunset, I think, is well earlier than that, but our twilight is bright. I am likely mistaken on this understanding, though. Haven't taken a journal or kept records, etc.)

>Politics

Says nobody that's been to Atlanta or has taken two minutes to use the internet to check preconceived notions about Georgia against the city.

https://www.myajc.com/atlanta-neighborhood-2016-presidential...

Politics isn't just local laws but also state laws.

Which state laws affect individuals more than they would n liberal states? Atlanta being such a major city makes GA very pro business. Businesses are doing a lot to keep pressure on the state to keep it from becoming more conservative.

Laws about abortions, bathrooms, access to the social net, etc.

Apologies for not responding. I actually forgot I posted this. :(

Lived in Atlanta for many years. It is much more liberal than surrounding areas, true. However, Seattle is in another league when it comes to this sort of thing.

Which is not to say that we don't have a ways to go, either. In particular, we have some really hard problems to consider out here.

So not sure what you define as experienced (2 years, 5 years, 10 years, 20 years?) but a 5 years of experience SDE at Amazon may make that in salary, but their total compensation will be quite a bit more than that.

If you hang on through your last 2 years of vesting (which is notoriously painful).

Don't fret, there's a good chance HQ2 comes to Atlanta and hikes those home prices. In fact, if I lived there, I'd probably make sure to buy ahead of time

There are plenty areas of metro Atlanta where the house prices are much lower than $335K for a nice size house. I purposefully chose the most affluent county in Atlanta (and one of the top 25 in the US) for comparison.

"Metro Atlanta" consists of either 5 counties or 10 depending on which definition you go by. There is so much sprawl in Atlanta and so much undeveloped land that builders will jump at the chance to build. This is still the South. The government doesn't make builders jump through as many hoops to build.

Yeah, exactly. Amazon pays less than Google and FB, but it is vastly more than your average SF startup, for example.

(wife works at amazon (but in palo alto) (and is 1 year out of grad school))

a. That salary is not low. If you chose to run a family of 4 on one salary of 160k, it's your choice. It's still quite good.

b. If you discount 30% of the comp because you feel like it, I wonder how many of the 20 other metro areas are going to match up.

Comparable houses in Seattle compared to metro Atlanta is a lot more than 30%. For reference, the area of Atlanta with both the highest average house prices and the highest household income is Forsyth County. The average house price is around $350K and the average income is around 90K.

Thats like a military Major's payrate when converted to civilian pay (bc allowances)

I don't know if I'm just being taken advantage of, or if your data is just wrong, or you're lucky, but I've never been offered more anywhere that wasn't Sillicon Valley big tech, or Seattle big tech.

- I posted a link from a recruiting agency (that I have used) but because you haven't been offered that amount, my data must be wrong? The link is "2017 salary survey".

- looking at another recruiting agency I've used, I found this job posting.

https://prestigestaffing.jobs.net/job/lead-net/J3W7C8754PTSH...

- from anecdotal experience, I have a group of four friends from a former job. We all still get together about once a month and we are in a private Slack group. All of us are in our early 40s and none of us have any desire to go into management. We all consider ourselves "full stack developers". Our salaries range from $125 to $145K.

- I just turned down a job for $145K as a lead developer (still hands on coding) because of the commute. I accepted a job that paid slightly less that was better commute and used newer technology.

They aren't slaves...

I met a couple of a guys at a bar a few months ago. They both had on Amazon gear and, when prompted, would excitedly talk about their job at a fulfillment center and what it's like to "work for Amazon". They had worked their jobs for 1 and 2 years respectively.

Remind yourself that the warehouse employees are in areas where their choice of work would normally be Walmart or a coal mine. Working for a big name company like Amazon has as much prestige for them as it does for us "tech scene folks".

Correct, but I believe the LinkedIn report was in reference to LinkedIn users. Many of which are not looking for warehouse positions.

A lot of it are "permanently temporary" jobs with no prospects of raises, promotions, etc.

I don't believe LinkedIn is concerned with warehouse workers as warehouse workers aren't using LinkedIn to find jobs.

I hear a lot of disaster stories from corporate at Amazon too, some from coworkers who have worked there. It has a well-known bad reputation in general in tech, even for non-software fields. My girlfriend has heard bad stories about Amazon from the hardware side as well. This doesn't even include some of the absolutely terrible things they do on the warehouse side either - I have friends who have first hand accounts from having worked there, and none of them are pretty.

It's one of the few companies on my list of companies to avoid, unless they improved their employee treatment (and got rid of their bad employee RSU vesting schedule).

> I don't think so at all. Sure, Amazon has a certain reputation, especially in hacker-news circles. That said I think you're forgetting about the silent but huge majority of average joes who would kill for an Amazon job.

They are the most aggressive at recruiting. I've never gone 6 months without at least one e-mail from an Amazon recruiter, despite asking to be removed multiple times from their lists. This leads me to believe they have turn over problems that would make your assertion incorrect.

For anyone arguing it is growth, consider they are the _only_ company that does this but they are not the only company that is growing at that pace.

Doing what? The point of Amazon is to eliminate whole swathes of human-powered jobs from the supply chain. You're either getting paid six figures or more for coding or you're busting your ass for minimum wage on the wrong end of the supply chain for just as long as it takes them to get the robots that will replace you working right.

What jobs do you think your "average joes" would kill for?

I would bend over backwards to put a few years worth of Amazon on my CV. I assumed that was common especially in this crowd?

The 5/15/40/40 vesting schedule is _really_ rough. If given the choice between Amazon and "no-name company," yes, Amazon will do wonders for your resume. But, if comparing an Amazon and Google offer, I'd want a big, big signing bonus from Amazon.

Amazon does a 2 year sign on reward, the first year in bulk the second year in monthly increments. No other company does that, but all we ever hear about is the vesting schedule.

When all is said and done, Amazon pays a little less than Google/Facebook (the caveat here: most Google/Facebook jobs are in California, not Seattle), but with the stock performance we're seeing, they end up making more on average.

The question, is when Amazon's stock stops appreciating so quickly, is other comp going to make up for it?

Oh interesting. What precisely is a 2 year sign on reward?

Amazon has a target compensation for you for your first 4 years, with a cash signing bonus making up for the lack of stock in the first two years, so it evens out.

Of course the 5/15/40/40 model only applies to your signing; all further yearly stock grants are over the next 2 years from grant time.

Is it common to get additional grants during that first 4 years?

It's common to get a second vesting in your third year (or else your comp would go down). It's almost common to get a promotion around that time -- managers are responsible for promoting their employees, and it's taken seriously.

It also means your destiny is in your manager's hands, which can be unfair if you end up with a bad manager, or several new ones. And unfortunately at Amazon, this is very common (current employee here).

That's literally every job ever. You aren't going to have a successful career at Google if your manager doesn't want you to. The org golden handcuffs are gone, so if you have a bad manager, change teams.

For the record "bad managers" are no more common or less common at Amazon as any other company.

Stock aside, they still pay healthy tech salaries, well over those of many professions in America. The stock is a bonus over that.

Never take upward stock performance into account in a comp offer.

If you think Amazon stock is going to go up and you have an offer from another company for more money, just take the offer for more money and buy some Amazon stock. Upward stock potential is never a valid reason to join a publicly traded company.

This is a misunderstanding because it doesn't take into account the asymmetry of the situation: you have more to gain by RSUs going up than you have to lose by them going down.

Imagine my yearly pay will be 150K at Amazon or 200K somewhere else.

Amazon stock could go up, and my pay in my second year could be 250K, in which case I am making more by being at Amazon.

Or it could stay the same or go down, in which case my pay in my second year would be <= 200K, but in that case I can just ditch Amazon for another company to bring my comp back to market levels.

Also, if the stock doesn't perform well, Amazon will grant you more to keep you in the expected band; but it won't take any away if you are making "more than you should".

They will only grant extra if the poor stock performance isn't a result of real problems in the company or economy.

They won’t take away terms of your existing grant, but they will nerf future annual grants when they see that you’re “already making plenty” with the shares that are yet to vest.

You can buy call options covering the entire vesting amount of RSUs with less than the 50k that would give you the potential upside that you would have as an employee.

Remember, predicting that a stock goes up is just as lucrative for an outsider as it is an insider. Never make a stock prediction part of your evaluation of a job offer because there are plenty of financial instruments to allow you the same upside as an outsider.

Can you explain exactly how to do what you describe? with Amazon's vesting schedule of let's say $200k RSU over 4 years? how much would it actually cost to "invest" like that?

You can always tell the people blowing smoke about stock options because they are speaking in the hypothetical instead of the past tense :)

As I used to tell people about why I was selling my options at [some other name brand company] as fast as they vested, choosing to work for the company is an investment. A huge one. And since I was far more likely to be laid off if the stock tanks, owning shares and working there was not diversifying my assets.

It’s the only time I made money on stock options and most of what I earned I earned by... profit taking on shares of a different company that enjoyed two stock splits before our stock started to crater.

Here's past tense: My grants have tripled in value since I started, and as a result my total comp over that period of time has been considerably higher than the offer I received at Google, which was in Mountain View. The higher COL of the Mountain View offer PLUS the outperforming stock put my comp way above what Google would have paid.

When that behavior ends, if Amazon doesn't adjust my comp, I can go to some other tech company, while still having one of the best names you can have on a resume (although, Google would have been arguably better in that case). Furthermore, if we consider my stock gains as part of my annual comp, it gives me a vastly better bargaining position.

Still blowing smoke?

>Still blowing smoke?

Yes, you're not correctly understanding that you depended on a gamble to outperform the Google offer. That same gamble you could have explicitly made in the stock market while taking the Google offer and capturing the upside of the Amazon stock.

Not at Amazon.

They take stock performance into account and it effects your future raises.

So yeah stock can go up, but if it does, you are never getting a raise.

That's not quite accurate. They take it into account for future stock grants. If the stock does well, you won't get more for the given year, but you will still get a raise.

That's a great point, thanks.

> Amazon does a 2 year sign on reward, the first year in bulk the second year in monthly increments. No other company does that.

That is a really bold way to frame "We don't even give you all of your signing bonus up front" as a positive. Kudos for coming up with it.

Their sign on bonuses are higher? And who cares, everyone's golden handcuffs are 2 years, so you aren't actually getting it up front anyway. It's pro-rated loan until then.

Maybe inside the SV bubble Amazon is viewed as a career stepping stone or prestigious or whatever? I can't relate but I'm in this weird career niche where open source commits trump or entirely replace resume/cv so I don't claim to be thoroughly knowledgeable about what motivates others. Personally I'd rather deliver pizza than work for a company that exists on the back of a pool of exploited low-wage workers. YMMV.

>Personally I'd rather deliver pizza than work for a company that exists on the back of a pool of exploited low-wage workers. YMMV.

Easy to say when you're making six figures. I doubt most pizza delivery drivers would agree with you.

I drove delivery for five years so I know at least one pizza delivery driver who agrees with me.

So did I. I was also a waiter. So I know at least one who disagrees. I guess it's a tie.

......dont pizza companies exist on the back of a pool of exploited low-wage workers

Yeah, but they do get free pizza.

Amazon, meanwhile, won't let anybody get free snacks from the bins except for mice.

I mean, if you want to play that game, capitalism itself exists on the back of a pool of exploited low-wage workers.

I've worked in a pizza place and known other people who have worked for pizza places, and none of us described it like a prison or felt the need to piss in a bottle.

No, instead they have a dangerous business model where they force their drivers to get to a location in 30 minutes or less causing them to weave in and out of traffic and crash in hopes of getting a tip. Much better than bottle pissing.

The what? That 30 minutes or less thing was outlawed in the 1980s, specifically for the reasons you cite.

Having a big-name company on your resume helps you get your foot in the door pretty much anywhere else. That's just the way it is.

You can use that to your advantage or not, but you won't be helping any low-wage workers by choosing to work at a small tech company over a famous one -- well, maybe you will if you pass on a job offer from Uber, but for the most part, tech companies pay low-level workers better than non-tech companies.

Amazon employees are literally pissing in bottles because they can't take bathroom breaks without getting fired. Several have been hospitalized or died from heat-related injuries. This isn't merely about pay.

If you constrain "reality" to the LinkedIn list[1] that Jeff referenced, Amazon is #1. If you drill into the urls and read the LinkedIn ranking methodology[2], you'll see it's based on views of the companys' job listings and career page.

If it was a ranking based on The New York Times stories and reader comments, Amazon would definitely not be #1.

Yes, when Jeff translates LinkedIn's limited criteria into "the most sought after places to work for professionals in the United States.", he's editorializing. That's not too surprising since it's a letter to his fellow owners. Cheerleading is to be expected.

[1] https://www.linkedin.com/pulse/linkedin-top-companies-2018-w...

[2] https://www.linkedin.com/pulse/how-we-created-2018-linkedin-...

"Already today, a portion of our European delivery fleet is comprised of low-pollution electric and natural gas vans and cars, and we have over 40 electric scooters and e-cargo bikes that complete local urban deliveries."

Sad to see even the mighty Bezos misuse the word "comprise." The whole comprises the parts!

Isn't that what he said?

The European delivery fleet is comprised of low pollution cars. Low pollution cars comprise the European delivery fleet.

I think he's saying that the correct form is: The European delivery fleet comprises low pollution cars, etc. Not "is comprised of".

The way I always keep it straight is, if the sentence still works with the word "compose," then that's the right word to use, and "comprise" isn't.

Whether that's correct depends on what dictionary you choose to cite. Some have given up on this particular battle.

Garner's Modern English Usage is the final word on grammar, as far as I'm concerned. Other dictionaries are pathetic postmodern excuses for language guides.

Hmm, I don't have that one. What does it say about Bezos's usage of 'comprise'?

"Comprise" works both ways.

No, it doesn't. Not if you care about usage.

http://www.lawprose.org/lawprose-lesson-145-is-comprised-of/

Congratulations and thank you to the now over 560,000 Amazonians who come to work every day

Wow, that's incredible. How much pay could each one make and the company still be profitable?

That’s a hundred thousand more people than the population of Florence!

Using 2013 census data, if Amazon employed only the population of Wyoming and its entire population was in the labor force, Wyoming would have an unemployment rate < %4.

Why Florence in particular?

I live in Italy, sorry I should have picked a US city.

Meh. Not all of us here are American :) "One sixth the population of Melbourne" would have worked as well.

It’s almost .01% of the population of Earth!

Could be as much as 20% of the population of Earth back in 10,000BC. Modern society is amazing.

this is a global community! Yay Florence!

'More than 10x the population of Florence, South Carolina!'

It's difficult to believe that this is FTEs. Very easy to believe it's FTEs, contractors, temps, and vendors. Including warehouse workers etc.

EDIT: I was curious, so I took a peek at their 10-K [1]. "We employed approximately 566,000 full-time and part-time employees as of December 31, 2017." I couldn't find any info on their FTE count with a quick keyword search.

http://phx.corporate-ir.net/phoenix.zhtml?c=97664&p=irol-sec...

It's probably noteworthy that their reference date is in December. I suspect their total would be lower in April.

Their 2017 profit was 3B, so they could each receive a $6,000 raise with that.

I'm so use to Amazon not reporting a profit consistently, I had to check to make sure....

According to Wikipedia their profit was 3B last year. If they have 566,000 full-time employees at 2,000 hours each, that is 3 billion USD they have to distribute to 1,132,000,000 hours, so they could raise the wage of each employee evenly by around $3 an hour, and would then make $0 instead of $3B.

When a company has this many employees (compared to some other large tech companies like Facebook with only 25,000), the math isn't as nice as you might imagine. Amazon has very low margins.

I've never been a big fan of Amazon OR Jeff Bezos' libertarian nightmare - but, seeing it framed like this, in terms of a private entity providing employment for so many people and for little profit makes me question my own beliefs. It's amazing what a little profit at scale adds up to.

For the vast majority of their logistical staff, $3 an hour more would likely be a substantial raise. Or at least equal to an extra yearly raise.

Bear in mind that hundreds of thousands of those employees are in the fulfillment centers etc.

According to my guesstimations, engineering should be 80-60k, which is still a healthy number.

I don't think high standards are the same as expectations, and it seems obvious to me how that misunderstanding can lead to poor management practice.

Some people, given a certain task, max out lower or higher than others. Which is, people have different aptitudes for each task. If the bar to be 'ok' is high enough to where not every person may meet it, and they're truly doing their best, then there's a management problem.

You can push people psychologically, and gently, and be within the confines of what I think is considered good management practice. But pushing people physically, and pushing them in a way that's mechanical, like a global quota, that's harmful to a person.

Just simply falling short of something and hearing "well, you fucked that up, don't let that happen again." instead of "hey, you're fucking up, do you need help? Is there anything wrong?" is enough to almost ensure that the person will fall short again.

It's cliche to say that honey catches more flies than gall, but damn if it isn't true. However, starving the flies until gall is all that's left is diabolical.

Sorry for rant, but I've lived like those DC workers are living, and its not pleasant.

Reply on news.ycombinator.com