4 ms·
Alphabet could've reduced compensation per employee by as much as 50% at any point in its history and still would've had no problem hiring the same number of pe
by roncesvalles 4y ago
Alphabet could've reduced compensation per employee by as much as 50% at any point in its history and still would've had no problem hiring the same number of people. But it didn't, because success in software is extremely sensitive to quality of talent.
The best analogy I've come up with is that hiring software engineers is like hiring a bunch of book authors and asking them to write a NY Times bestseller. While your competitors are also trying to do the same.
If anything, I'd argue tech compensation isn't high enough. Software has unprecedented reach and (as a corollary) a winner-takes-all flavor. A team of a few hundred tech workers could build a business that brings the company in the order of a billion dollars in annual recurring revenue, justifying their own salary by several orders of magnitude. No other profession can do that. An accountant, plumber or lawyer can never create business value that is 1000x their own salary.
- Apocryphon 4y agoHow does this gel with: 1. The widespread meme that Google and other likeminded tech giants snatch up talent for the sake of denying competitors that talent, rather than for developing their own businesses. 2. Google’s seeming inability for the last decade or so to launch new products that stick around, as well as questionable decisions with existing products- with their chat messenger strategy being the intersection of both failings. Maybe the previous post only applies to SWE and not PMs.
- darth_avocado 4y agoYou know this. I know this. Pretty much everyone in tech knows this. But this isn’t about only Google paying lower, Google already pays lower than competitors and still hires talented people. This is more about market wide pressure to suppress wages. If every company does the same, suddenly the quality engineers start getting paid a lot less than they should be. Of course over a course of a longer period, this drives up competition and new startups capitalize on the opportunity, but in the short term it resets the wages until the cycle starts again.
- mgraczyk 4y agoThe part about Google still hiring talented people is not true. You should look around at offers. Google is hiring fewer good people because other companies are paying more. Google is losing their best people to startups that pay more. For ML roles, salaries are much higher than they ever have been.
- thinking4real 4y agoMany in tech are only their instead of starting direct competition to the major tech players because of the high compensation. Sure I could create my own startup and make more, but with lower pay and way lower headache i’ll take the salary. If I was being offered what European companies offer swe, i’d be out the door. I might even go take a job that’s enjoyable since the pay isn’t there. I actually hope this is a broad movement for big tech companies to pay less. I’d love to see an exodus of power from them.
- sacnoradhq 4y agoWhere I work, there are multiple senior ICs making 7 figures. There are many dimensions in the calculus of layoffs: - An opportunity to get rid of staff they don't like - Trim down for the particular economic environment (Meta went insane the past 5-10 years, zooming up in head count like a hockey stick. Google has some bloated departments.) - Virtue signal to investors that they're "using their money well" - And, tamp down on wages If high impact engineers had any sense, they would: 0. unionize 1. form worker-owned co-ops to capture the market value for their labor rather than whatever crumbs a corporation decides to grant them.
- roncesvalles 4y ago>form worker-owned co-ops to capture the market value for their labor There's a word for that - "startup"
- bluedino 4y agoGive bonuses to that 1% of SWE's then (pretty sure Google does this)
- rchaud 4y ago> A team of a few hundred tech workers could build a business that brings the company in the order of a billion dollars in annual recurring revenue, Like who? "Software" in and of itself doesn't accomplish anything. Nobody today is selling CDs in boxes that run entirely offline with no dependencies. What is Airbnb without a giant network of IRL brokers that actually make inventory available? What is Uber without thousands of real people running ID checks, lobbying politicians, and negotiating down AWS and Google Maps API bills ? Figma, everybody's favourite unicorn, only made $190m in revenue. It's only in bizarro world Silicon Valley that that results in a $20bn acquisition.
- roncesvalles 4y agoSoftware has near-zero marginal cost to manufacture, very few barriers to distribution, and can displace giants in a very short amount of time solely by virtue of being Good. In other words, software is viral. Valuations are an extortion fee. Here I have a doomsday device that will kill your multi-billion-dollar business faster than you can say Mississippi, and I'm selling the key.
- rchaud 4y agoI don't know if any of that holds true anymore. Marginal Cost to Manufacture - Not in terms of cardboard boxes and CDs, but you can't scale up your user base without paying hefty fees to cloud providers or payment processors these days. In any case, software companies aren't valued on the basis of gross margin but recurring revenue and operating cash flow. Barriers to Distribution - OK, this is still mostly true. I will say that "enterprise" type software tends to require RFPs, SSO integration and regulatory compliance that can let monopolies fester, like Microsoft (and Adobe). I wouldn't say software is displaced at all though. Companies are simply buying the competition as in the case of Figma, (or Instagram, or Activision) or pouring money into developing vertically integrated solutions to strengthen their moats further.
- novok 4y agoValuations are also about the future based on current adoption trends. It's like asking why did facebook pay so much for whatsapp, it's because it was very close to pulling a tiktok on FB themselves. They had half a billion people and was the defacto network effect locked in messenger for most of the smartphone world outside of China and the USA & Canada, on top of being extremely more staff efficient. If they bothered to add social networking, it would've been the beginning of the end for FB in those markets, and I wouldn't be surprised if it was a WIP in WhatsApp when acquisition talks were happening. Figma was pretty much acting like the whatsapp to core parts of adobe business based on adoption. If they let it go on further, a similar thing would've happened.