My favorite story of "unintended consequences" of belt-tightening like this was when I worked at a startup about a dozen years ago. We did well for a few years, but it was clear to people we were reaching a point of market saturation. Nevertheless, the CEO kept agreeing to doubling of sales revenue year after year. Well, no surprise, the last year before big layoffs, we were way off on sales goals.
The CEO announces we have to "put our heads down." Pretty much the main change enacted was no more ping pong. The problem is that were really only two guys constantly in the ping pong room. Those two guys just happened to be the two most senior sales guys, and the only two that were meeting quota.
They both left about a month after this and not-so-slyly told the CEO he was an idiot.
2008 GFC, startup of 25 (15 dev), everyone drops to 4 days a week, the newly hired 'CEO' drops 5k on a top of the line laptop for himself to basically run Outlook and Solitaire after telling everyone they were cancelling 'birthday cakes' (which the staff mostly pitched in for anyway). RIP entire company.
I've seen companies reduce top-performing sales people's commission because it was "costing too much money". So the good sales people left and the company didn't have that problem (of making lots of money) anymore.
> I'm still working on removing the virus, seems to be using drivers to mask its presence from the virus checkers I am using. I've got some ideas based on using a scanner on a live-cd.
I regularly practice just switching to a whole new system and restoring from backups. I think asking an infected system to scan itself is sketchy. Just blow it up, mkfs, and rebuild it. Binaries are disposable.
It would be interesting to have a view showing all comments from the above submissions on the same comment page.
I guess the ranking would have to be proportional rather than absolute, since I'm assuming more recent comments would have gotten more votes due to a growing audience.
Yes - I wrote software that makes it fast to do by hand.
You'd certainly be welcome to resume! I think including the date + number of comments is the important bit, plus not pointing to threads with zero comments - but IIRC you had both of those features.
Edit: one fun fact – it's surprisingly difficult to post links to past threads without coming off as somehow reproaching the latest submitter for posting a duplicate. In reality, HN allows reposts after about a year (and this is in https://news.ycombinator.com/newsfaq.html), but many people don't know that and interpret the list of past threads as an implicit criticism, which is a shame.
I've tried out a bunch of different wordings trying to minimize that misunderstanding. The simplest one seems to be the most effective - I just say "Related". Somehow the word "related" sounds less like a criticism. It also has the nice property of being inclusive, so for example it's fine to include other articles on the same topic.
One of these years, I still want to build software support for collecting related URLs and related past threads into HN's official UI. Then we can all build up sets of related things collaboratively. That will hopefully make HN more interesting.
My reading of it is that it isn't the soda, it's having the bean counter become an authority that can make decisions.
If you work day and night to get a product out, and then a CFO slides in above everyone on the team, and on top of that they're a cheapskate, that sends a signal.
Right, I think that's exactly the point of the article. In no world should free sodas make a substantial difference, but it's rather the signal it sends.
Much like a single cockroach at the entrance of a fancy hotel isn't by itself a problem if it's limited to exactly that cockroach...
Having been through this transition multiple times, I agree that it’s not about the sodas or the CFO. Why do you think they hire CFOs? To do things like this. To impose cost discipline. To be the person everyone is mad at instead of wondering why the company hired a CFO in the first place.
Same with “HR types ruining the culture” and “middle managers running amok.” Those people all get hired to transform the company, because leadership has decided some thing very fundamental: What got us here, won’t get us there.
PG once wrote that startups should frame their reason for existence as “trying to answer a question.”
Famous questions of the past: Will microcomputers sell? Will people pay a premium for bit-mapped graphics and a mouse? Can advertising sustain a search engine? Will the world use a social media web site built for college students?
But at some point, you know the answer is “yes,” you know that a sustainable business is possible, and you are no longer a startup. The difference between “answering a question” and “running a sustainable business”is vast.
Maybe killing free soda is not the best way to start the transition, but one way or another, all the people who want to explore the unknown, test their visions, and answer questions are going to leave, to be replaced by people who enjoy playing the world’s most cut-throat, high-stakes engine game, business.
If you truly enjoy what a startup is and how it works, you need to pack your parachute long before the free sodas go away. You need to pack your parachute when everyone realizes that you are no longer wondering if your company can become a real business, the only questions are how, when, and who.
On the other hand, maybe you want to build a real business once you know it can be done. Gates did. Jobs did. Long before them, Ken Olsen did at DEC. If so, accept that the sodas will eventually go and the controls will be imposed.
I don't disagree. I wonder if it's possible to work at a stable company (not a startup) that doesn't treat its employees as an annoyance to be constrained and mistreated.
Sure, but it's a "pick two" problem. Culture, Growth, Profit. Startups bleed money to create a culture (entitlements, informal benefits, etc) that attracts talent and generates very high growth percentages. When a company gets big enough it can either give up the culture in order to turn a profit while growing or give up the high growth rate to maintain its soul. The latter become those stable, durable companies that'll last a hundred years but never make pop investors jizz their pants. The former have the chance to become mega-corporations.
There is of course the third category like Google whose solution was to stratify into a caste system where your core employees maintain some of that original culture while most of your employees are contractors you treat like shit.
My 2c is that these transitions aren't about giving up culture, they're about changing to a different culture.
There are companies that make a lot of money and grow both organically and via acquisitions where people have a well-defined culture that attracts top talent... But it's a different kind of talent, it's the talent that enjoys building a lean, mean, money-making machine.
Historically, ITT was one of those companies during the Age of Conglomerates. Later on, IBM and Xerox come to mind. I recall that during the 80s and 90s having worked in management or sales for Xerox was a golden ticket to almost any job you wanted. Same with working in finance for GE.
What about working in management at Disney today? What about working in leadership with Apple or Amazon today? These are very well run companies with a strong management culture, growth, and profits. And investors love them.
My 2c is that these transitions aren't about giving up culture, they're about changing to a different culture.
That's a distinction without a difference.
it's the talent that enjoys building a lean, mean, money-making machine.
Precisely. You start a company as a problem-solving customer-pleasing machine and it becomes a money-making customer-exploiting machine. The ideal world for the MBA is one in which your customers hate you just not enough to stop paying you.
Curious, are there good examples of companies that were able to delay the "pick 2" problem for a long time?
Im assuming such company would need to be a dividend-paying machine bc otherwise the profits would go into retained earnings (vs staff growth or R&D expense, etc) and that would make investors really unhappy.
Steve Blank (the author of the article) put this very succinctly.
"A startup is an organization in search of a scaleable business model." When the company moves from exploration to execution, it is no longer a startup.
Oh wow, I hadn't thought about it that way, that the end of free sodas isn't necessarily a bad thing, so long as done at a natural inflection point where it's transitioning out of startup mode and into a steady business model.
My informal model of startups is that they're business that are bad at any one specific thing they do (or at least, not-necessarily-good), but are good at quickly iterating between different business models. Edge cases (all IMHO):
Uber is still a startup because they have huge unresolved inefficiencies in their core ride-service model and are looking for ways to best make use of the network they've established.
Amazon is not a startup because they have a steady business model that enables long-term production at scale. They certainly enter new markets and offer new services all the time, but do it much slower than a startup would (and often do it just by buying a startup). Same logic for Apple (not that anyone was saying otherwise).
Quibi was not a startup because, from the beginning, they locked into a specific model and were slow to iterate.
Although Luke did not lose his arm in wampa cave on Hoth (he lost it at Cloud City) he could've gotten a new one at the Redemption which was stationed at Hoth at that time.
Ontopic: I wouldn't give a shit about soda or free soda, but coffee and tea is always free here for employees in my country (quality may differ!). I simply would not work for a company which wouldn't feat this. So the social norm defines it and keeps its own behavior status quo. That's also a reason why its difficult to get rid of smoking/smokers tobacco in society
I'm not entirely sure what you're trying to say here (who is the bear, the employees or the company?) but anything related to text adventures / interactive fiction has my immediate upvote!
It’s sort of the natural evolution of companies as they grow in size. When they are small, things are just done, but as they grow and as they bring in more people, the cost of those resources also grow. The free sodas are a fun example (and here outlined as the wake-up call) but where it really starts is when someone higher up notices the cost of a department that has grown from 5 to 20 people, or something similarly eye raising, and then bring in the E&Y type consultants to help transition into a scalable business. Which simply involves reporting, cost-trimming and looking into efficiency. The sodas are where the rebellion break out, but the seed are sown long before that as the “bean counters” and “process people” slowly begin making business intelligence part of everything because it lets everyone report on everything.
You can’t even say that it doesn’t work, because it’s how every major company operates. On the flip-side, it’s not like Coca Cola has really invented anything of worth for like a 100 years. So while bean counters are financially good for investors, they are probably pretty terrible at running our society, because it’s the engineers that actually build things and the founders who come up with the saleable ideas.
Anyway, if my career has taught me anything it’s to do your thing. Being part of the transition from startup to enterprise can be a lot of fun as well, so long as you know that you can’t really fight the MBA types and win.
> I don't think the soda cost per person would be different for 20 people though, so it seems a weird thing to raise the eyes about.
Think of it from the management perspective. It’s not about the soda, it’s about the “we’re giving this middle manager $1million a year to pay for his/her employees now, and we sort of don’t know what we get in return”. In startups nobody has time to think about that sort of thing, you simply do what’s necessary to get stuff done, and, the amounts are smaller. As size grows, so does the perceived need for bean counters.
I say perceived because I’ve never personally experienced a company who didn’t go that road, so I can’t say what would happen if you continued to trust all your employees and middle managers rather than start monitoring and structuring them, but I do know that a lot of the business intelligence and documentation systems that I have seen showed little actual value once they got disrupted by covid and everyone working from home.
Of course it's not any different. But the focus has changed. When you're an early stage startup, a 0.7% overhead for junk food (made up number, but that's probably not too far off) is a fundamentally negligible thing. It's like 3 days of runway, if that.
Basically: who cares, the company will live or die based on products and funding rounds, and you're just trying to make the company not die at this stage.
But to an established company trying to make money, 0.7% of revenue is a huge (seriously, huge) line item in a CFO's accomplishment list. Consistent profitability is built on a long, grueling ladder of tiny incremental bean-countings. And to an established company trying to make money, that's where the focus turns. And those are the employees who get rewarded.
It's not that either side is "right", it's that they have opposing priorities and that growing companies need to manage that tension and not just give in to the temptation to lock the soda machine.
>But to an established company trying to make money, 0.7% of revenue is a huge (seriously, huge) line item in a CFO's accomplishment list.
Ok, that seems pretty weird right there, 0.7% overhead of an early stage startup is now 0.7% of revenue even though we agree it is just increasing as the same rate of hires?
In other words while each new programmer hired will drink the same amount of soda as the others it should be evident that the percentage of revenue would be decreasing, otherwise they have some serious problems that getting rid of free soda will not fix.
Indeed, but also it’s a sign that senior management collectively - inc the CEO - now see trivial cost cutting as a worthwhile use of their focus and a good way to generate value for shareholders.
I agree with this. When management think in money only, that's a good sign for me to leave. If they think that happy employees are just magically happy, and don't care about making employees happy, that is just ignorance. Don't happy employees perform better? Does employee performance not matter? Profits have long been the end goal for American companies, and it's too bad. Chasing money, instead of treating employees well, treating the community well, and treating the environment well, is a bad way to run a business, and we are all suffering for it now.
Unfortunately, this anecdote re-enforces a stereotype. The CFO in this story is bad and so is senior leadership and the board for allowing the new CFO to run amok. Not all new CFOs are bad and not all CFOs with leadership will destroy a company.
It’s easy to attribute revenue to products, but difficult to fully account for their costs. It’s hard-to-impossible to attribute revenue to free sodas (and the staff who drink them), but easy to account for their costs.
Many business decisions are heavily influenced by this dichotomy in accounting.
I'd say its more sinister than that. It really is signaling a shift from creating to wealth extraction.
The no free sods is just a rules for thee but not for me situation. Do you think the exec floor with the said CFO no longer has free soda? Of course not. The execs have free soda, coffee, donuts, meals, open bar, beer, private jet reservations, wine and all kinds of other extravagances.
Free soda (well, usually coffee) is just the indicator...
The company is either broke, and can't afford soda (so it probably won't afford any raises and bonuses), or it's trying to save money on meaningless, cheap stuff (and will try to save money elsewhere... like on staff and wages). To me, this is a signal to slowly leave.
> No one on the board or the executive staff was trying to be stupid. But to save $10,000 or so, they unintentionally launched an exodus of their best engineers.
Maybe the insight here is "keep your engineers happy, it's important". But I believe there is more to it on this story. The engineers heard "we could give you soda when we were small, now we're big, we cannot give you soda" and, to the rational mind, that makes no sense.
Maybe they left because they were sitting on a precarious equilibrium and the free sodas tilted the balances. But I think it was also that the engineers saw the sign on the wall, and thought "this guys have lost their collective corporate mind, let's bail before they have any more bright ideas".
I think there’s something a bit deeper going on, and it’s implied in the post. Small companies (on the whole) are more fun than big ones for a certain kind of person. At a small co. you get to wear a lot of hats and have a much more holistic feel for the business. As the business grows, you get more specialized, and restricted. Some people, and I am one, get bored and frustrated in that sort of environment. Things getting more “corporate” is a signal that transition is happening, so the people who don’t like that kind of environment leave. Some companies handle the transition better than others.
> At a small co. you get to wear a lot of hats and have a much more holistic feel for the business
I like the hats.
Once, a (junior) coworker asked what I expect from the job.
I figured it was 6 things:
- Money, of course
- Interesting work
- Learning, or at least putting good stuff on my resume
- Ego, wanting to say "_I_ built that"
- Control, wanting to say "I built that _with my tools_" or getting to work on "_my_" projects
- Socialization
I ranked them in order of how concretely the company offered them to employees.
Money is the most concrete, it's in every contract. Socialization is the most abstract, the company can't make any promises about that.
The middle ones were fuzzy. The company's product is somewhat interesting, and won't change often, but learning, ego, and control were not in any contract. These were things I quietly extracted under the table because I just really wanted them.
I would like to reify them. I'd like to say to my manager "I'm going to work on some pointless internal tool nobody needs, just so I can play with this new language. Then when I realize we could never deploy it, I'll feel better about using an older language on our real product. I can't control my inner child, and letting me play with blocks and Play-Doh is cheaper for us than hiring a company therapist or firing me."
A couple times my boss' boss teased us about "10% time, like Google does", and it never manifested. I've just been taking it anyway. Probably less than 10%. My last performance review was great and every month they give me money, so I haven't been caught. Maybe it works because, unlike Google, I'm not trying to write a new chat app every time I need a promotion. I'm just occasionally doing things nobody told me to do, based on my intuition, to satisfy some sub-conscious wannabe hacker drive.
One more thought - Sometimes I wonder if manager-types read these threads and think, "Okay, so when it's time to stop being a startup, we drop the free soda and replace the whiny unherdable hacker cats with work-a-day copy-pasters. All according to plan."
So I wonder, maybe if I ever stop liking my company, it doesn't mean the company's bad, it means I just need to find another startup and ride it for 4 or 5 years until it gets too big again.
What part of "free soda doesn't scale" is irrational? I get that it signals a change in management priority -- specifically a change away from prioritizing staff and toward prioritizing money -- and I'd probably leave too. But I don't see how you have to lose your mind to achieve that perspective.
“free soda” doesn’t have to do with scaling, or being small, or anything other than what a company is willing to pay for their employees really.
A company cutting employee perks without further explaining what employees will get in exchange is irrational and being stupid. I mean even drastic lines like “you get to keep your job that otherwise we would cut” are usually explicitly stated to try to convince the employees there’s something in for them.
We decided to be hostile to you, soda-drinker, personally, we are some combination of tremendously greedy and out of money, and the trend is towards more bright ideas of this sort.
> What part of "free soda doesn't scale" is irrational?
All of it.
Free soda for 10 people is hard. They have different tastes. There are too many things to keep enough refrigerated at the same time to make sure you hit soda rush times of the day. And you're probably running to a local grocery store, at least sometimes, to make sure you don't run out.
With 100 people, you've probably settled into something more standardized, no more special Puerto Rican pineapple soda for that one person with the interesting tastes. Everything can be delivered by the same one or two vendors. You know roughly how much you need and when - much more predictable. And you have more fridges now, constantly stocked with more soda. Moreover, the free soda now costs less per person.
The only problem with free soda for a big corp is that people are more likely to take the mick and take more than their fair share, as they’re less likely to feel obligated to the company to be reasonable.
There's a fairly low natural limit to how much soda one human can drink. Yes if they're dragging pallets out to resell on the street that might be an issue - an issue of the company not paying enough such that hocking loose cans is worth it for their employees.
You're working very hard. Now and then, you grab a free soda. You work so hard that the corporation has a chance to grow. What is your reward for all that hard work? The chance to pay 50c for the previously free soda. From this point of view, there is nothing rational about it.
Similar story -- back when I worked at a physical engineering company, one of their cost-cutting methods when things got tight was to cut down the amount of time they ran the A/C during the day. This applied to all floors of the building except the one where the partners' offices were.
This led to a number of people routinely using their breaks to take the elevator up to the "partners' floor" and cooling off / airing their sweaty BO for the bosses' benefit.
I think that these cost-cutting schemes never really work out, they either cause the best people to leave or trade money for work-done.
I worked for a bank where they had terrible coffee on our floor. We used to joke about how bad it was until one day these fancy automatic machines showed up that ground the coffee for each cup, they even had milk! Times were good, we'd walk over to somebody's desk--"I'm running a build, do you want to grab a coffee?"--and we'd chat in the break room about how great these new machines were.
Then we had a couple of quarters in a row with bad numbers, cutbacks were being made, if we do this then we don't have to have layoffs, etc. We lost our fancy machines.
The impact to morale was huge, we only had had them for a couple of months, taking them away once we got used to them was a huge deal. But more importantly, most people on our floor were now used to drinking partly-decent coffee. A couple times a day we'd go to the nearest coffee shop and buy our own, which involved walking to the elevator, waiting, going 23 floors to the lobby, badging out, walking across the street, getting coffee, walking back, badging in, waiting for the elevator, going up 23 floors, then finally back to your desk. This had to waste at least 20 minutes, more if the elevators were busy.
There are places that do cost-cutting the other way around. I recall from a place a long ago where it was handled by 1) freezing annual raises for everyone, 2) cutting managers' salaries by 5%, 3) cutting executives salaries by 10%.
Sometimes that's called, "Leading from the front."
Targeted layoffs of people everyone complains about because they are bad at their jobs and also terrible as team mates sends a way better signal than keeping them and removing perks from employees who do work.
When someone impacts morale, forcing a team to work with that person is highly damaging.
The managers all knew who these people were, so they could have given them severance without calling it a layoff. I'm mostly referring to a situation where everyone else would have been relieved instead of spooked.
Severance may come with it's own difficulties: state unemployment payments, contractual obligations, bureaucratic overhead, manager pushback.
I think leadership-oriented types tend to esteem headcount the way nerds esteem CPU core count and frequency - up is fine, down is not acceptable unless you're absolutely forced. Neither is a really rational position, but they overall tend not to be too detrimental so there's no widespread change.
I've been railroaded into manager hiring, at a huge company everyone knows, and so I've gone over the manager hiring documentation. Assuming you don't fuck up the interview, the level you get hired at as a manager ... is directly related to the number of reports you claimed to have had at your previous position (there is some verification, but nothing "real").
It's not the only factor, but it's by far the most important.
Managers fight over headcount because it directly affects their personal ... everything. Career, money, power, ... all are dependent on headcount ("number of reports").
If you have 100 useless employees and you offer them each $50k to leave of their own volition, you will reduce the number of bad employees and save money. You won't have any legal obligations because they're leaving on their own.
This is how it works at various big, powerful tech companies (although the severance offer often comes with a "performance improvement plan" as the alternative, a bit of a carrot/stick situation which usually gets people out the door pretty quick).
Where your comparison falls apart is that every nerd knows that you can't run all cores at 100% all the time, that's just wasteful.. thus underclocking and shutting off cores when not in use is part of every modern processor.
In my experience, unless you lay off a key person, layoffs tend to generate a lot of drama for a day or two and then things quiet down. It's a short, temporary pain that hopefully can be justified with real numbers.
Cutting perks signals a permanent change in how they approach things like that, and is really hard to justify since in every single case I've seen it, perk cutoff is highly targeted at the rank and file and execs still get catered lunches during meetings, etc.
We had something akin to this happen at my company. For context, this is a well established mid-sized company with around 5000 employees. Some beancounter decided that they could no longer afford to subsidize the plastic spoons in the break rooms. So they announced that when the current stock ran out, they would not be replacing it.
The backlash was pretty strong. Not because engineers can't afford plastic spoons of their own, or some other alternative, but because it was such a petty thing. These spoons come in cases of 1000 for under 10 bucks, so nobody can argue with a straight face that it was about the cost.
I don't know that anyone actually quit over it, but it leaves a sour taste in your mouth.
This reminds me of a startup CFO I worked with, who could see the big picture. In the early days of the company, they worked in an office that shared a bathroom with a half dozen other companies. And the toilet paper in said bathroom was terrible. Like, not just standard-office-building terrible. Really, really bad. So CFO takes it upon himself to rectify the situation. He orders a couple dozen rolls of standard-issue "premium" office toilet paper (still not that great relative to home-use TP, but good by office standards) and leaves it in the bathroom. Two days later, it's gone: someone (not clear whether it was someone at the startup or one of the neighboring companies) decided to take it home with them. Everyone in the office starts grumbling about the state of the TP situation again.
Not to be outdone, CFO places another order; several cases this time. And it takes a few more days, but sure enough, it all disappears.
At this point, CFO is fed up. He orders two full pallets of TP. After lunch one day, several volunteers unload the toilet paper. Multiple stacks, floor to ceiling, in the bathroom. More stacks in the supply closet. Stacks in the hallway. Stacks in the lobby. Everywhere you look, there's toilet paper. And if people were stealing TP after that, it wasn't at a rate that anyone could notice.
Including the ongoing costs for the subsequent 18 months until they moved out of that building, the "unlimited TP" policy might have cost the company $10k. But it bought a lot more than that in terms of employee goodwill. Knowing that the bean counters are focusing on "our employees are producing a product worth $x; let's figure out how to have them do more of that" makes a big difference.
> All these engineers were still heads-down, working their tails off, just as they had been doing since the first few months of the company. Too busy working, most were oblivious to the changes that success and growth had brought to the company.
This gets at the vague steps that the author mentions at the end about supporting transitions of _people_, but I was waiting for them to get concrete about how the engineers should have _positively_ been made aware of the success and growth. Pay raises to market rate? A bonus? I’m guessing that in actuality, it was little more than verbal pats on the back.
I guess it’s different for each company culture, but this makes it sound like the engineers worked hard, the company got solid, things were taken away, and then the superstars started to leave. I’ve seen similar patterns a few times. What are some positive ways that people have seen these patterns fought?
Based on how I've seen it mismanaged, the only way I see it working is hiring "provisional VPs". In the same way that you might have hired slow and fired fast in the beginning for competent ICs, you need to do the same with the now executive management skillset that's required to break off and scale whole divisions of the working organization. The other side of that is you need to equip them with the actual subset of the talent they need, and not hold back some employees who were there in the beginning for special considerations.
The ICs have been there for years, knife fighting with the market and the competition, and then someone comes in from some adjacent market or some other vertical entirely, and the expectation is that all the work done to go from zero to one needs to be fit to their agenda, and not the other way around. Everyone who has lived it knows its backwards, and yet I've seen the trigger get pulled on this so slow that it had a measurable effect on the exit.
A small, mildly entertaining, anecdote about soda:
When Oracle acquired the company I was working for, they made the soda free. That is, Oracle made the soda dispensing machines simply dispense soda when you pressed a button.
The same company that had a self-service "I Quit" page.
Free soda is an indicator about how a company treats its employees, but perhaps not the best.
I had a client that stocked kitchen fridges with free basic foods - drinks, meats, breads, etc. - and quality, healthy stuff too. That always impressed me, sending a simple, strong message that they cared about their employees and people like me, and it had a positive impact on me when I ate my share (and didn't have to worry about taking time to get lunch).
At one point they brought in a relatively large team of contractors to work day and night on a critical, time-sensitive project. A week after the team started I was there, talking casually to the manager. They told me that there were complaints that they were running low on food in the fridge because the contractors were eating it too. What would they do? Ban the contractors, with only arms-length loyalty to the business, working day and night, and on whom the business's fortunes depended, from the kitchens.
> At one point they brought in a relatively large team of contractors to work day and night on a critical, time-sensitive project.
To me that is a huge red sign that something is seriously wrong with the company. Bringing in a bunch of new people and having them work long hours does not seem conducive to good code.
Not everything is permanent code that has to be maintained forever. Sometimes you have a migration, or a bunch of data to suck in one time from a vendor or acquisition, or whatever. You can run appropriate quality checks to bound the error rate.
Accounting/PeopleOps meets the Real World! Great anecdote.
The free meals are a perk paid for by the people creating the revenue. I suspect that perk probably wasn't budgeted for addition to the Contractor work. An oversight, perhaps? Hopefully folks working got fed.
The loss of a single food item in your own house has a much larger impact than the loss of a single food item in a kitchen stocked to feed dozens of employees - in the former case, your lunch disappeared but in the latter case there's just one less pack of crisps in a box of 100 that nobody will even notice and the positive impact of that (in terms of morale, especially for regular contractors) outweighs the 50c the pack of crisps costs.
I worked for a post-startup company early in my career.
The Christmas parties were awesome, my first few years there included parties at the Shedd Aquarium and the Field Museum. There were subsidized vending machines as well. I knew engineers who would spend a dollar on the vending machine and get a few bags of chips for lunch.
Subsidized vending machines were the first to go. Then the Christmas parties. Then a bunch of other changes which showed derision for customers and employees both.
Is Oracle all that bad of a place to work? Everyone _hates_ Oracle for product reasons but I've never met any engineers that said it was a terrible place to work or particularly worse than any other big tech b2b companies (e.g. ibm, microsoft, etc).
Immediately after the Oracle/Sun acquisition, there was a huge exodus of senior engineering talent (e.g., most of the core Solaris kernel development team). This writeup of why James Gosling left has some of the obvious reasons (e.g., he was asked to take a significant pay cut), but also this anecdote, showing Oracle managements' general attitude toward the talent:
Making his point about the “creepiness,” not only with Ellison but with Oracle’s power structure, Gosling said he sparked a notion to try to improve morale amongst the Sun faithful who endured the Oracle acquisition. He said the company decided to rent out the Great America amusement park in Santa Clara, Calif., and allow the Sun folks to have a day of fun. Scott McNealy and Sun CEO Jonathan Schwartz signed off on the project that came in well under budget and all systems were go, Gosling said. Except a few days before the event was to occur, Oracle Co-President Safra Catz got wind of it and put the kibosh on the thing.
“Safra found out and had a fit,” Gosling said. “The word came down that Oracle does not do employee appreciation events. So she forced the thing to be cancelled. But they didn’t save any money because the money had been spent – so we ended up giving the tickets to charities. We were forced to give it up because it wasn’t the -Oracle Way.’ On the other hand, Oracle sponsors this sailboat for about $200 million.”
I've always found it somewhat amusing that Ellison's facial hair is virtually identical to the episode of Star Trek where they cross into another dimension where they are all evil. Like somewhere there's a clean-shaven Larry who is the life of the party and getting recognized by the City Council for his humanitarian work.
1. "Oracle does not do employee appreciation events" is not the same as treating staff badly. Events like that have all kinds of frankly important accessibility and inclusion concerns that are tricky to handle. Not saying it's the choice I would have made, but I think it would be fair to, rather than navigating that poorly and leaving some people unhappy, simply not have a budget for that type of event and (theoretically) fold that into salary. After all, those perks are technically part of your comp.
2. The fact that you segment technical staff from other staff is exactly the kind of tiered treatment a blanket policy would seek to avoid.
Every ex-Oracle engineer I've met says it was a terrible place to work compared to the big tech company where I work now. I also don't know anyone who went to work for Oracle after leaving my current company, fwiw.
Oracle was one of my favorite places to work ever, I barely did anything at all. Extremely low expectations. Got a raise every year and eventually a promotion to Staff. Got to go on free trips to different datacenters. Free soda, catered meals on holidays, lots of morale parties.
Their products are total shit, but it was sweet where I was!
Super anecdotal, but I once interviewed an engineer who was currently working for Oracle. During the interview he told me that every six months he would go apply to jobs at startups, get an offer or two, and bring them back to his manager and ask for a raise. He had been at Oracle for 15 years doing this apparently the entire time. So it seems some people very much like working there. Needless to say I recommended we pass on him...
it strongly depends on which org you're in, with an average of "boring, with a big steaming pile of bureaucracy" based on my observations spending five years at the seattle-based cloud org. if you had a good manager and skip-level stuff could be pretty alright, if you didn't god help you.
If you have a good manager, and your project doesn't span across orgs, you have a ton of freedom. Crossing orgs creates red tape you wouldn't believe, since such requests "must go through Larry's office".
They also have an absolutely stellar health plan - I know of one (US) employee whose spouse got cancer, and they didn't pay one penny out of pocket for the entire treatment.
Somewhat ironically, it was the only way to do it. If you tried to talk to your manager, they would direct you to the self-help page. There were just too many scripts tied to it - from disabling your access to notifying your manager to setting an end date in their payroll system.
There were legends of folks who failed to press the button, and were being paid for years thereafter. While I imagine a kernel of truth is here, I can only shake my head at the thought of Oracle's lawyers getting ahold of the ex-employee in question.
I assume the form has some dissuading text along the lines of, “Upon submitting, a goon squad will be sent for your immediate removal from the premises.” Or maybe they just send spiders through the ethernet. It’s a big company.
With 200,000 employees (at my time of employment), they couldn't care less about turnover. Acqui-hires and people wanting a working retirement keep Oracle afloat in the tech department.
Realistically, if you weren't on the Oracle DB development team and didn't draw too much attention to yourself (for example, via daily reimbursed meals at elite restaurants accompanied by the best wines for years on end), you didn't really matter.
To present an alternate point of view, I never grokked this idea of free soda, free food, catered lunches, sleeping pods, dry-cleaning, massages, ping-pong table as a way to attract talents. If I took a risk to work for a startup, I want a lean machine that pays me in ridiculous cash or ridiculous options, and obscene milestone bonuses. The only thing I think I'd really care about is "bring your dog to work" and "onsite/nearby child-care", there's no substitute or purchasing-equiv to that. A startup is an intense high-stress environment where focused work and time is invaluable; none of those "perks" directly drive the goals, if anything they seem hedonistic distractions from the goal.
I think the point of supplying food and drink on-premises is exactly that, to foster "focused work and time "
The objection isn't so much the price then put on the food and drink. It's the conversion of 'valued employee' to 'hourly crank-turner'. No longer a valued critical member of a team, now the corporation regards you as a 'resource'.
E.g. did the executive perks disappear too? No? Now you know what pigeonhole you've been placed in (demoted to).
I tend to agree with this viewpoint. I actually found that my happiness decreased when a company I was working for started providing free lunch every day. The lunch break - along with a change of scenery, a short walk, and the freedom to choose whatever I wanted for lunch - was important for my well being. Also, employees would often complain about the free lunch options on certain days, which is kind of sad.
I suppose it would have been different if it were a big campus at a FAANG, though, where they have lots of different restaurants to choose from spread out through multiple buildings. There you still get most of the benefits of the lunch break that I felt I was missing out on. And I assume there are certain efficiency benefits to be had at that scale vs. having thousands of employees all having to fend for themselves.
I thought I remembered reading somewhere that Google (or some company in a similar position) had worked out the math and figured that the free meals more than paid for themselves in increased productivity (or at least seat time) because employees weren't getting up to go off-site to find lunch. As often as not they'd be quickly back at their desk coding away.
Never mind startups, I once had a client who was in the printing business, a numbers-driven blue-collar culture at the time. They worked out the loss of productivity from workers going out to the coffee truck and decided it was a win to spend a little money on coffee if it kept workers from going outside when they heard the truck’s horn.
As an employee I also find it more convenient. I don't really want to pick people to go to lunch with, wait to be seated at a restaurant, wait for someone to take my order, receive a standardized quantity/proportion of food, wait for someone to come with the bill, decide how much to tip, etc.
Instead I serve myself the exact proportions that I want, go find friends to sit with, eat, chat, then drop off my dishes and leave. The rituals of eating out and having food prepared to order are expensive and don't provide a lot of value to me over self-serve, generally speaking.
What’s funny about Google is that people will travel across campus (sometimes even to other nearby campuses) for better food! It’s still much faster than eating out, though.
I once worked on a company that made me understand the value of this stuff.
My project was late, I was coding desperately, sometimes skipping lunch to have enough time, I had to leave using a bus on a specific time so I couldn't just stay late.
I realized then, that whenever I wanted to drink a coffee or a tea, there was one, hot, on my table, and I didn't fetch it!
Snacks too sometimes!
And whenever I left to go to the bathroom or some random reason, when I got back, my place was clean too!
This meant I could focus on my work, just coding, I didn't had to think about fetching coffee or tea, all I had to do was work, and whenever I needed coffee or tea or food, it would be there, ready, on a clean desk.
It made me realize the service staff worked as a sort of "force multiplier" for the company, you hire a few of them, and a lot of programmers can squeeze more work hours, and what you spend on the service staff wages, you save on programmers wages that you would be wasting if they were wandering around looking for food or caffeine.
Nothing saves time like not having to leave the office for snacks and drinks.
Maybe you're organized enough to have the stuff you want stocked for yourself and ready.
Many people aren't.
And as the article points out - this stuff doesn't cost much.
You're not going to get a significant raise if they don't have a pingpong table and Le Croix.
It's also a little bit of an accounting trick IIUC. Those things are expenses for the company and tax deductible. If you buy them yourself - you're probably paying >40% income tax on it.
I like the explanation from this article [1] titled "Why perks matter": because perks like that avoid context switches and remove "distractions and other things which could disrupt [my] flow".
As a human being who regularly needs to eat, not having to worry about having the right amount of snacks and/or change at hand can make or break my day. Trying to code while hungry is a waste of everybody's time and money, which is how free food pays for itself. And don't get me started on trying to code when all you really want is a nap...
I don't see it as unintended consequences of Soda Not Being Free. It's an attribution mistake, not entirely different from how SEM tends to get attribution for conversions it didn't really earn.
The Elves leaving is unintended consequences of the company getting larger, and not finding a way to manage it in a way that retained top talent. The soda exposed it, but it was going to get exposed one way or another.
As a proxy variable, it's a proxy for things that were already true, they just hadn't realized it yet.
It's like how there's a lot of turnover after bonus season. The bonuses aren't the reason they're coming/going (though they're often shaped significantly by the real underlying reasons they're coming/going), but they do effect the timing of their coming and going... because that's the point where it becomes impossible to ignore the underlying reasons for the turnover.
It's a general problem of being, as the Brits put it, "penny wise, pound foolish".
Some bod in accounts can show an obvious change in the bottom line by cutting a cheap perk. The problems in retention, hiring, delivery are all someone else's problem and much harder to quantify.
See, I think that's not getting it. Cutting the drinks isn't what cost them the people. It's just what caused the loss of people to be evident. Particularly since the organization had changed, it would have been unwise NOT to bring the perk policy in line with the rest of it.
But that's exactly what the article states. It doesn't say they left because of the soda; it says it was a wake up call that made them realize it was no longer the company they liked and which they helped build.
Yes, I'm in no position to contradict the narrative of the story, and so I'm not.
I'm quibbling with the characterization of it being an "unintended consequence" that "launched an exodus". The changes had already happened. The exodus had already been launched. It's just no one had actually left yet. If anything, I'd argue the sodas were the unintended consequence, but the exodus, or at least the impetus for it, was very much intended.
It's worth watching the little stuff. When you see a retail establishment with a partly burned out sign, and it doesn't get fixed quickly, you know they're in trouble.
> When you see a retail establishment with a partly burned out sign, and it doesn't get fixed quickly, you know they're in trouble.
As others have already mentioned, this only holds true for a specific kind of store and market. Many businesses don’t care to fix their sign. The Safeway near me had a broken sign for almost two years.
It’s classic elastic vs inelastic demand. Grocery stores don’t need to worry so much about their brand image. Not zero, but less than a store selling luxuries.
It's the deviation from the norm that's a potential warning sign. You'd know the cool bodega might be in trouble, with a lame new owner or something, if the sign got fixed or a bunch of painting was done...
Traditionally the good Asian family restaurants were the ones with bad service that didn't update their sign. Though I don't think this is true anymore because of generational change.
I have wondered similar. Tradies need clean and tidy vehicles to present an image of competence, organisational skill and attention to detail. But if they were to start turning up in a Porsche Cayenne it might signal they cost too much. Fine line in image management.
Certain businesses and situations allow for the owner to squeeze as much money out as possible. Image doesn't really play into it or is unnecessary. Captive markets, knowing an eventual sale of the land will be a windfall, etc.
Coffee is extremely popular, and too easy and cheap to meter when produced on site. Soda costs more and more hassle to buy (in cans/bottles), is extra work to make cheaply (install and maintain a tap), and mostly is grossly unhealthy and most people (outside of coder stereotype) don't want it at work.
SodaStream-style device fits the coffee mold, but it's less popular and not traditional.
I don't have numbers, but my expectation is that cost is of the same order of magnitude. I don't think coffee is all that cheap. I'm thinking of all the problems I've seen with the plumbing leaking, the machine being left on and burning requiring extensive cleaning, broken pots, and so forth, all on top of the material expense. Soda doesn't have to deal with the heat, but it does have pressure, so that probably kinda balances out.
Recently joined a company with legit soda machines(the fancy ones with exotic stuff like cream sodas), slim jims, ice cream, cookies, cakes and which buys lunch for the engineers every weekend. It's amazing. I came from a place where you had to pay 2 $ for a soda!
It's definitely one of the things that you don't realize when you don't have it, but when you do it's like heaven. It's really helped me get more work done since I can snack away and it encourages me to get up for walks around the office.
Eh, there's a downside to freebies, which is curation.
No matter how luxurious and all-encompassing, your free corporate fridges are not the supermarket, your free corporate cafeterias are not an east coast diner. They don't have everything you might want, just a decent selection.
We're deep in first world problems here, but there's a certain frustration to always drinking the drink that doesn't quite hit the spot, to searching the cafeteria menu for the least bad menu option. At least when you are paying the money yourself in the wider world, you can actually pick what you want.
Years ago (like, 1993 or so), our college CS class visited Jones Farm. One of the stops on the tour was a break room. All of the drinks were free at the time, whether it was coffee or coca cola. One of the engineers there remarked that at Microsoft only the caffeinated drinks were free. LOL. I just about believe it.
Contractors don't get perks like that anywhere, thanks to Vizcaino v. Microsoft. Withholding resources afforded to permanent employees is weird and counterintuitive, but it's how employers demonstrate compliance.
people really don't understand what a blessing it is to not have sodas be free. that kind of culture with an expectation of it (and 'snacks') being free will create a lot of fatness in the industry
> Maybe I should start my day with a line instead?
Bad idea.
Once, when I was less than a year out of school, I was in a design meeting with two of the most senior engineers on the project. They were WIRED, difficult to reason with, and sent me on an obviously wrong direction. I thought they drank 4-5 cups of coffee.
A few hours later I realized that the nonsense email that I got 10 minutes before the meeting, from the department head, titled "Lines in the conference room" meant that they probably snorted cocaine right before the meeting.
About a week later I had to redo all of my work using the exact design I pushed for. 2-3 weeks later I was relieved when I was laid off.
---
If you want to legally and safely know what cocaine feels like, take 600mg of caffeine pills (3 No-Doze) as soon as you wake up in the morning, on an empty stomach. Don't do it before you need to do something important, or if you're doing something where you could loose your temper. Also, don't do it if you have heart problems.
Oh, I never would, and in that case I have no idea why people would do cocaine at all, because when I have too much coffee I get full of unsourced anxiety and the need to poop.
Two thirds of Americans (European numbers are similar, higher if anything) drink coffee every day, which is, not to put too fine a point on it, an addiction. I'm one of them for sure.
It's easier to just give people their fix, and there's a long history of this, paid coffee breaks were an early concession unions demanded during industrialization.
I used to be in a management role a number of years back and faced this same issue.
Thankfully I was able to convince executive management that for our several hundred employee company, running a Coke soda machine was basically the cost of 1 Tech Support headcount.
That's it.
One of the most popular in-office perks, the Coke freestyle machine was 1 tech support headcount.
That was years ago. And we still have that machine.
The cynic in me wonders if management and the soda industry colluded to make free soda a symbol of something deeper, namely, management pretending to care about their workers. Come for the free soda, and stay for the open plan offices and Agile Scrum meetings.
I once read an article, I think in the Wall Street Journal, about Levi-Strauss sending out promotional materials to HR departments about how to set up "Casual Friday" at the workplace. It was a promotion for their Dockers brand.
The 2008 recession was especially fun because of bean counters. Some of the big Indian outsourcers used this to remove all sorts of things from the office stating "we can save money with these things and avoid layoffs." Among things that were removed at the outsources I worked at? Paper towels in the bathroom were removed in favour of the hand drying machines.
If you study this story carefully, you'll notice that it wasn't a case of "new CFO was an idiot". The ENTIRE BOARD, including some experienced VCs, agreed that they were "too big for that now".
My analysis: What does "too big for that now" mean? Does it mean that they can't afford it? I doubt it. They were still going to sell the soda "cheap" (presumably 25-50% of the normal soda cost). So what was it? My theory is "big" ISN'T a synonym for "large-scale" here, but instead is a synonym for "professional". Why is free soda a negative indicator of professionalism? And to whom? I think that once you reach a certain level (non-startup or series-Z-startup), investors expect you to be extremely bean-countery. So the ultimate source of this problem is mega-scale investors which want to see a tight ship. The board was likely feeling pressure from them before the CFO was ever hired.
I've been around long enough to see this cycle through twice before, but it still rankles how stupidly myopic it is every time.
Beginning this year we're no longer allowed to have the company pay for an in-office team lunch without pre-approval of each individual time from a 3-skip manager, who's in a department that might as well be a different company. At the same time they want to mandate 60% return-to-in-office time. The actual leads had all ability to incentivize people to come back cut off, and know reactions are going to range from work-to-rule through quit.
This reminds me of when Microsoft stopped offering the free towel service. In the buildings with large bike racks, there were locker rooms with showers. Normally they would have a free towel service so you didn't have to bring your own, like a gym. But then one year they said that they were spending too much money on it compared to the number of people using it. There was a huge uproar.
Then again, if it had been the free sodas, it would have been a mutiny. I remember when they got rid of the spicy v8 and a mutiny was considered, but declined.
I wonder about the "unintended consequences" takeaway here. Correlation is not causation. It seems to me that the talented startup engineers leaving and the decision to discontinue free soda don't likely have a causal relationship between them. Just both are things that happened along the growth path from startup to established company. More of a common cause explanation. If they had kept giving free soda would it have really kept the engineers from leaving? I have doubts.
My thoughts as well, but it _is_ plausible that if the sodas had remained free, it would have taken a few more months for the "elves" to notice that this was no longer the kind of place they wanted to work. The "uproar" basically sent the message of "this place is run by accountants now, not engineers", and that was what sent the engineers to the exits.
I could see myself being really upset by this change. It’s not just about the cost of the soda - it’s also about convenience. Now if I want to have a soda at work I need to remember to keep change or small value bills in my wallet. And if one day I forget then I need to leave the building in search of a treat, or else just go without - in either case I’m annoyed.
And more than that it’s about feeling valued and respected. It can be really upsetting when something you’ve gotten used to is taken from you. Especially when you know the cost to the company is just a tiny fraction of the overall operating budget, and unlikely to make any significant difference to the bottom line.
Interestingly at my first job that I worked at, sodas weren’t free, and it never really bothered me. But if they had been free, and one day stopped being free, that would annoy me. I think there is an asymmetry between the satisfaction people get when they use a perk, and they dissatisfaction they get if you take it away.
The interesting thing is that in MANY of these situations, you can guarantee that executives are getting free drinks, lunches, etc. as part of the perks.
I've seen it happen. No soda, no food. But the executive assistant will do a lunch run every day for executives. And will pick up drinks too. Or there's usually a stocked fridge for executives near their wings of the building.
It’s rare that a company can cut this kind of cost in a way that leaves the employees feeling trusted, valued and respected. That feeling is what hurts retention and dries up hiring referrals.
Additionally, the service has a value in time and money saved. There are bound to be employees for which it tips the decision to leave sooner. When many different conveniences are cut, more employees cross that threshold.
Finally, there’s a meta aspect to this. If you know a decision is harmful to the company and you see your managers making the harmful decision, it can be harder to keep working for them.
It's a combination of things - if company takes away benefits and provides no new benefits - you as an engineer lose money. Meanwhile in current market just changing the company means a rise. Why settle for an effective salary cut when you can trivially get a rise instead? The difference between no rise and cut is huge mentally.
Additionally if the company saves money on such a small scale it sends the signal things are BAD.
Hmmm ... anything is possible. You don't offer any evidence for your hypothesis and by that standard, anything is possible. The OP was there, at the company, saw what happened, and has observed several other similar circumstances.
It never actually bothered me that we had a CFO in any of the organisations I worked at. I like having a clear situation regarding who's financing what - especially when it comes to business trips.
A HR manager where there was none previously is a whole other story though.
The CEO announces we have to "put our heads down." Pretty much the main change enacted was no more ping pong. The problem is that were really only two guys constantly in the ping pong room. Those two guys just happened to be the two most senior sales guys, and the only two that were meeting quota.
They both left about a month after this and not-so-slyly told the CEO he was an idiot.