As a regular reader of Rachelbythebay - I agree with the sentiment of a lot of other posters here, it's not the money that makes people quit.
A classic story here is the "Sodas are no longer free" one: https://steveblank.com/2009/12/21/the-elves-leave-middle-ear...
The total cost of free sodas for engineers is probably small compared to their salaries, and for the individual engineer, the new $0.50 per soda cost is probably negligible compared to their salary. But moving from free (as in free soda) to non-free signals a culture change, and so the best engineers started updating their CVs.
A person close to me works at a law firm. She was feeling a bit stagnant, so she connected with a recruiter. She got a very solid offer with a significant pay bump. She gave her two weeks' notice to her firm, including appointments with the partners. One of the partners asked her for a half hour. They came back with a massive pay raise, and a promotion to partner if she would stay. She was in a state of shock, but then informed the other firm that she was staying at her current firm.
By way of contrast, an engineering firm I am familiar with had an employee who had been there six years, and knew the company's very complex product inside and out, every nook and cranny. He was one of the only people who had such deep understanding of the system that he could fix any issues that might come up, hardware, firmware, software, everything. He gave his two weeks' notice, and then went to a different job. He's a very talented guy, who would command a very attractive offer, but his talent to the current company is vastly greater than his generic value on the market, because of his detailed knowledge of the product. Although he diligently documented his knowledge, the company was still left in a jam after his departure. It would have been great if the company had fought for him the way the law firm fought for the other individual described above.
Amazon used to have an internal Slack channel that allowed employees to share their total compensations anonymously. Thousands of people shared and the numbers were eye opening. Basically, if one was not a star and got consistently promoted every two or three years, she would not get compensated at market price. As a result, an L6 who stayed in the company for years often got lower total compensation than an L5 new hire who was actually an L4 before jumping to Amazon.
Note: No L7 or above shared, or so as I knew. That said, like any large company, resources tend to concentrate to the top. L8 and plus were still compensated really well, but L7 were so so because L8 became the new L7 after waves of rapid promotions inside Amazon.
This is from 2021. Since then, they hired their Head of People a month later, and ~6 months later published a follow up:
https://www.goethena.com/post/a-public-and-transparent-formu...
Management Mindset: As long as the employee is not complaining, why throw more money? Besides, they can't be "that valuable and hard to replace".
I have even seen management who don't even bat an eye or goes extra mile to hire some 5-10x more expensive(even compared to market value) limited time contractor but zero interest or energy in putting up a fight to give a fair pay bump to one or more severely underpaid employee(s).
The tainting is a mind game: an employee suddenly hears from their friend or random recruiter about some lucrative offer(usually at least 25-100% raise in pay), tries out their luck and lands an offer, but now manager(assuming a sensible person) needs to put a fight[1] with upper management to bring that same bump(or comparable).
However, the relation is now tainted: employee now feels that injustice had been done and there is no guarantee that in future further raises/promotions will not come easy[2] and the manager might be looking into replacing them soon. The manager is now also sad that, they got under pressure by a leveraged employee+upper management will push to de-leverage such assets(yes a term managers use), and seek to replace such expensive assets soon. Now, both sides are running on bad trust and it gets immensely uncomfortable, hence the employee leaves anyways or get replaced by some shiny new hire who will command more than the original employee being replaced attained at their level.
[1] This is often difficult, unless your manager is politically influential, because once you have neglected an employee too long, adjustment can be bit high, it is now difficult to justify, how suddenly a particular employee became so much valuable out of the blue. Human mind is wired in such a way that, slowly ascending a hill feels less tiring compared to ascending steep jumps, as the latter takes more energy(or in case of employee big pay bump).
[2] Even though the adjustment being a correct valuation, any raise/bump of significance also brings expectations of more responsibilities and higher performance which is the same ol' undervaluing again in action.
It's always puzzled me why many tech companies seem so reluctant to give counter-offers to departing employees. Trying to pay as little as possible and not easily giving raises, even to their most productive employees, is at least somewhat understandable. But if a valuable employee is about to leave, possibly taking some critical knowledge with them, you'd think that a 50% raise should be on the table out of pure self-interest. So why doesn't this usually happen? Is it just part of keeping up the charade (a big raise is a tacit admission that they have been underpaying massively)? Is it to increase the barrier to negotiations (you have to actually leave rather than just get an offer)? Is it because HR people get more points for bringing in "new talent" rather than keeping the current workforce? Or is it just that neither HR nor management have any clue who their critical employees are, so they are unable to prioritize?
The article ignores that "the market" works.
Switching jobs, with new physical locations, new informal structures, new colleagues imposes a change cost on the software engineer too. That's why a lot of people stay. Additionally, paying every engineer the market rate wastes money on those that would have stayed anyway.
As a similar problem, countries that pay for babies find that 98% of their money is wasted on couples that have babies anyway. The Economist says: "schemes in Poland and France cost $1m-2m per extra birth" https://www.economist.com/leaders/2024/05/23/why-paying-wome...
Now that's a scale problem. Is one engineer staying really worth $10m? $1m? You'll pay way more than you think.
It is simple.
Because they have no money.
For startups, they want engineers to take a pay cut to work 996-like hours.
Especially in the UK, where there is a running joke where companies complain about a “skills shortage” when in reality companies cannot find engineers at top universities that want to take a pay cut and work for less pay.
Fairness.
I'll work hard for less money if it feels like we're in it together and the potential compensation is fairly distributed.
But if someone pays me less just because they think they can get away with it, I'm out.
My 2 cents - I think this is also a function of the domain the organisation operates in. IMHO, there are predominantly two types of companies - one where the technology creates the business vs. one where the business uses the technology as an enabler.
For the former, think of companies like Boeing, nVidia or the Windows/Office divisions at Microsoft. Here, the product of technology is what brings in the money. For the latter, think of almost any "IT company" in the world which pieces solutions for clients in various businesses by leveraging existing technology/tooling and applying some amount of customisation.
The former tend to value engineers more (exceptions are always there) as there are usually less number of competitors, but the risk of the employee switching and enriching the competitor's knowledge (and hence, business) is significant. The latter tend to look at engineers as replaceable (with little to medium effort) resources, and value their payroll expenses more than retaining top talent.
Hiring people is like gambling for companies, as in most tech companies hire some person on the off chance they're a 10x developer, though expect them to be a 1.5x one on average. Even if you're a 2x developer, you haven't lived up to the hypothetical 10x ideal to your employer, so they aren't willing to put in the extra money and effort to convince you to stay, there's more incentive to put that money into hiring the next potential 10x-er
An analogy is, let's say you get a treasure box that can be opened for 100 dollars and may contain 1000 dollars. You open it and find 150 dollars. There is a smaller box inside which you can open for another 50 dollars which could contain at most 75 dollars but will most likely only contain 50 dollars. Do you spend those 50 dollars on opening the smaller box, or do you use the money to buy another larger chest that could contain 1000 dollars?
By the time the Engineer has gone through the effort to find some place to go to, it's well past time for the company to increase pay.
They'd have to pay more than the difference, so they don't. And they'd have to be able to predict which engineers are about to leave (and churn models are terrible.)
Middle and upper management never want to be put in a position to explain why they designed a team where it is not easy to just swap someone else in instead of paying more to keep someone. Usually if this happens, they will pay more to keep you until they find a way to solve this problem.
I worked in the UK at a FAANG, so I was on a different bonus system, but my US colleagues had a very significant lock-in. Both your starting bonus and annual bonuses were implemented in terms of share grants that vested at 25% each year. I heard from several of these colleagues that the annual bonuses were often about the same as the base salary, so for some of these guys it was around $200k. So, at any given time they had between $150k and $200k locked up in un-vested shares that would be lost if they left the company. It's framed as a benefit rather than paying someone to stay, but ultimately it's structured this way so it serves the same purpose of providing a strong disincentive to leaving.
I wonder if what's going on is that most of the work is done by people who joined within the last two years.
Or maybe somebody leaving and having to be replaced by a new hire kind of rolls the dice on headcount, which low and mid-level managers might find interesting.
The answer I like the most is that, as the article says, when you hire a new person you have to pay, on average, market rate. When you're evaluating whether to change the pay of Joe Lifer, you sometimes have to pay market rate but often you don't have to. Maybe you'd save money by giving raises to Joe Lifer, but who knows? You're hiring people anyway, so if Joe Lifer wants to leave, that's a shame, pay the new guy instead of him.
The people running companies are mostly "professional managers" who see engineers as a fungible resource. If anything, tenure in the same role is mostly a negative signal. Why haven't you gamed the promotion system yet? Don't you want to grow into management or 'technical leadership' and spend your days in meetings? What is wrong with you?
I've heard the argument that an engineer who wants to go should be let go. They won't be motivated anymore even if you pay them more to stay. Not sure if that's valid.
> Sep 26, 2021 > 5 min read
Although it was only 3 years ago, from a macroeconomic standpoint, this feels like centuries ago. Since then, the tech industry has weathered several erosion events (SVB implosion, interest rate hikes) and it seems that aside from generative AI (which with the recent NVIDIA drop, may be showing first signs of slowdown), at least the VC funded startup space is at a local minimum in terms of market heat. Anecdotally, I've seen a lot of colleagues I've worked with at previous colleagues in hotter markets struggle with finding their next role for much longer periods of time with much more frustrating processes. The market for candidates seems tougher today than it's been in over a decade; in fact, I'm not sure what the last time was that it was this rough -- perhaps right after the 08 crash?
"The impact" of an engineer? How to measure that? Hmm. Let me think ....
Think of a new company the customers will "love". Start the company and, being a good engineer, be the CTO and CEO. "Impact"? The value of the new company. Pay? They OWN the new company.
It's honestly wild. Just had someone leave because of an incredibly dumb money-saving corporate policy reason. The overall impact will DEFINITELY be that projects get delayed and we waste a ton of time training the new guy, it would have been much cheaper to fix by just paying him more (or fixing the actual benefits issue). It's mind boggling how short sighted some of these fucking decisions are.
> "With a market this hot"
Maybe the title should get a [2021] because it sounds like the market has cooled substantially in the last year or so.
The article doesn't seem to take into consideration market cycles, assuming market rate always goes up at a rate that outpaces cost-of-living comp adjustments. While this may be the case more often than not, how are companies supposed to absorb market downturns? A salary reduction, if legal (?), seems almost as bad as redundancy except you risk being saddled with disgruntled workers who might decide to jump ship at the next moment that is convenient for them.
Some companies do. At my previous company (I have since retired), I would get yearly outsized stock grants to motivate me to stay. I sincerely doubt I could have found matching compensation at another company.
I never had any expectation of loyalty from the company and it didn't from me either, otherwise the stock grants would be unnecessary. I had some loyalty to my manager and had promised him to give a heads up before leaving the company. He knew I would retire once I hit my target.
In my experience, the job hopping for an increase only works up to senior / lead level (or engineering manager for manager counterpart). Anything after that requires time in role having a consistent, positive impact, and building up a portfolio that you can then use to sell yourself at the company for one of the coveted principal / staff / otherwise distinguished roles.
I see one obstacle by this transparency approach: How to create a fair measure for "impact"? I would say that it is very hard to measure the actual impact of a senior developer. He gives advice here and there, he influences decisions in a good way, he asks the right questions, he recognizes risks early and so on. All that can hardly be measured.
it's the same everywhere, not just with jobs and careers. contracts for (mobile) internet are always much cheaper for new customers than for old customers. or old customers don't automatically get the better conditions of new customers. and that although old customers are even more valuable, because they have already earned their acquisition costs.
The issue no one here seems to be talking about is that paying people according to tenure is anticompetitive for the company on both sides.
If you pay people more because they've been at your company longer (the way the follow-up post by Ethena describes[0]), you're explicitly choosing to pay them more than they can get elsewhere. You don't want to pay more than you have to for talent, so this is a hard sell to whomever manages the budget.
On the flip side, if you're attempting to pay people in proportion to their worth to you as a company, you're going to be paying less than the competition, because the competition front-loads this money. A new engineer takes a few months or more to ramp up, so if you're making an attempt to pay engineers based on their impact, you will be outcompeted by companies willing to give sign-on bonuses and extra comp to convince people to switch. That's built into the plan of the four-year cliff - you pay them a lot to start with and hope you get the savings on the other side when they don't spend the effort to switch jobs later.
Lastly, turnover isn't as much of a negative for the company as everyone seems to ascribe. Being forced to keep up with industry best practices and technologies to be able to recruit talent, to onboard new devs when someone leaves, and to retire unmaintainable legacy cruft when the creator leaves are not strictly negatives - they are risk reduction. That's not mentioning the benefit of fresh eyes and fresh ideas.
(Honestly, I think all of this discussion on both sides ascribes too much rational decision-making to what is essentially cargo-culted hiring processes. The biggest companies copy decisions from each other to the point that it's literally collusion[1], and everyone else follows suit because they are smaller and don't have the economies of scale to make researching alternatives positive expected value. People at the top setting policies don't have lines of communication to front-line managers to be able to determine whether their particular company needs extra focus on retaining developers for business continuity reasons, so to the extent that it's a conscious decision at all, it's based on industry-wide studies or company-wide turnover statistics.)
[0]: https://www.goethena.com/post/a-public-and-transparent-formu... [1]: https://www.theguardian.com/technology/2014/apr/24/apple-goo...
It's challenging being in big tech right now. I spent a year applying for equivalent level jobs at peer institutions and startups. Finally got an offer from a startup that was heavily stock weighted. Took the offer to my boss and got a $100k raise within a week.
TBH suspect I was being seriously underpaid.
1. Roles have budgets and companies only have so much cash to pay people.
2. Raises done arbitrarily to match market rates snowball across the organization. One 20% raise is going to turn into 2, or 3. See #1.
3. "Losing time" because of a junior hire is, oddly, less of a disruption than issue 2 becoming issue 1.
How can a company effectively measure an employee's impact? In my experience, when performance metrics are tied to salary, employees often manipulate the numbers to reflect a more favorable outcome, which may not accurately represent their true impact.
While the conclusion that engineers leave because they aren't paid enough does hold some truth, I'm much more compelled to cite the old "People quit managers, not jobs" saying.
In my experience money is absolutely part of the reason why people quit. Not everyone lives in places or is in circumstances where finances don't matter that much.
I am making over 100k more this year than I expected at my current job due to stock increase, that’s pretty damn good retention.
I wrote a related Article "why aren't software developers paid more" (https://jonpauluritis.com/articles/why-arent-developers-paid...)
the reality is that we get comp way wrong... and its related to hiring, and its related to Venture Capital and related to a million other things
So the new head of people will help come up with a magic remuneration formula… right…
I think this reads a bit differently today than 2021… should be in the post title
All this effort, metrics, measuring impact, performance reviews, blah blah blah. You want to make sure your engineers paid top dollar? Tell them to go interview in other places. If you truely value them - beat whatever they were offered by X%.
From 2021.
With apologies, they are idiots. That they can't frelling figure out even remotely who is worthwhile to keep around or not is 300% on them, something so few orgs actually calibrate on, or do so based off awful anti-signals on, and it's embarrassing that this pathetic foolery has just gone on and on and on and on, with zero signs of improving.
My boss loves me, swears I'm so full fo awesome knowledge & smarts. But every review, there's some VP pissed off I asked a question about why we were switching to OKRs or how to was going to work or how it would help or some security team pissed off because I have complained about NetSkope or their shitty undocumented USB drive blocking policies. There's always some fuck doing a shitty job offended that some engineer dare speak up, and my boss is telling me again and again it's costing me many dozens of $k in raises.
It's so so so sad. Even if there aren't the dogshit losers, there's still so little discernment & taste. And there's so many people who can look busy and suck up, but actual deep knowledge is so so rarely respected. There's so many orgs that straight up deserve to fail, based on how bad they are at supporting their truest asset, their human resources.
greed
Supply and demand
Salary?
[dead]
Because it's easier to enslave them with non-competes.
if the salary of a role is 100k and you can pay seniors up to 400k (or some number). After some point, it won't make sense financially to increase salary as you can hire people for this role with much less. So the limit of the pay is set by market. If engineers want bigger and bigger pay, they would need to change to other roles that pay more.
This entire piece starts from the conclusion that engineers leave because they aren't paid enough then works backwards from there. This is wrong from my experience. Anecdotally, I have never left a job because of money. Sure, I left jobs and got ones that ended up paying more money but the reason I started looking in the first place wasn't money. It was for other reasons that usually included frustrations with management.