| |
Why did he leave? The tl'dr is the L6 glass ceiling and the remote pay cut. I can speak to both of those:First, for those that don't know, new grads get hired at L3, PhDs at L4 (which IMHO is actually a mistake but that's another issue) and there is an expectation of getting promoted to L5 (Senior Software Engineer). There's no set time frame for this. I've known people who stayed L4 for many years and were quite happy. L6+ start to require increasingly extraordinary influence and technical contribution. > It's hard to get promoted beyond L5 for open source work I'll clue you in on a dirty little secret: it's hard to get promoted to L6 for anything. This is by design. It's how the company controls costs and maintain retention (by dangling that L6 promotion, possibly for years). One of the things that Ruth Porat did when she became CFO is lowered the promotion target percentage. This meant less people got promoted at each cycle. How promotions work is a committee of higher-level engineers (eg L7s consider promotions to L6) and will have 10-15 packets to assess. These will get stack ranked by the committee. If you're above the line you'll get promoted. If you're not, you won't. Ruth's change just made that line a little higher. As for the pay cut, this requires some context. By the time you reach L5+ the majority of your total compensation comes from equity, both your initial grant (for the first 4 years) and annual refreshes and discretionary grants you get along the way. So less than 50% of your income comes from salary plus bonus. For those employees who go permanent remote, there is a location-based pay adjustment. SF and NYC are at 100% rate. Other areas at 85-95% of that. That only applies to the salary plus bonus (since bonus is derived from salary) part of your total compensation. So a 10% cut for L5+ is in reality <5%. You don't save on taxes by moving to San Diego but it is cheaper than the Bay Area. If you move to a no-state income tax state however your net pay can in fact be higher. So, the argument for no pay cut is you're saving the company money by not having to pay for office space, a desk (etc), free food and so on, all while doing the same job. This seems to be the author's position and I'm certainly sympathetic. The argument for a pay cut is that your compensation isn't really associated with your output at all and is entirely market-based. For many people not going into the office and being able to live somewhere that isn't the Bay Area is a huge plus so people are willing to give up something for that. Plus you could argue the company doesn't want people going remote who don't really want to by offering the same pay. By lowering your pay slightly it creates a psychological barrier such that only people who really want to do it do so. I'm sympathetic to that argument too. |  | |
Long-time former Googler here, author of go/l6talk (internal presentation on getting promoted to L6). While there I coached a lot of people on the L5-to-L6 promotion process. It seems a lot of Googlers get disillusioned about getting promoted above L5. Broadly speaking Google's culture is heavily focused on levels and promotions, which is too bad, since many L5's are doing just great and getting paid extremely well.If it weren't for the fact that Google makes level information public within the company, I doubt there would be quite so much emphasis on getting a promotion. Most people want to get L6 for the recognition and the fact that it elevates you above other engineers in terms of status and influence. When I worked at Apple, levels were private, so you had to treat everyone the same -- a person's influence was not dictated by what level they had on their company profile (there is no company profile at Apple). In my opinion this approach was much better, and significantly lessened the internal competition for promotions. | | | |
> I'll clue you in on a dirty little secret: it's hard to get promoted to L6 for anything.It's hard if you're on the wrong project (read: low impact and/or low visibility). There are different routes to L6 at Google, but by far the most successful was to become TLM or just M of a medium impact project. It's hard to make the case as an IC for L6, but colleagues did it multiple times. Large enough, long-running projects also reach a critical mass of L6+ people and then getting high-level support becomes much easier. Having been through the process, seeing the process from an IC, TL, TLM, and promotion committee member, if you think that L6 is unachievable, there is either a problem with your project or your expectations. | | | |
"Having been through the process, seeing the process from an IC, TL, TLM, and promotion committee member, if you think that L6 is unachievable, there is either a problem with your project or your expectations."Lots of managers are also not good at talent management (and not just at Google). Which is sad. So they tell people it's a promo committee's fault, or ..Realistically, it's also the manager's fault. If they don't have scope/work that can help someone grow to L6, the right move is to help the person find a place that does, rather than string them along. Stringing them along ends up with them leaving or unhappy or ...Helping them grow elsewhere ends up with people who will likely want to come back to your org at some point when you have the need/scope/role for them. | | | |
I know that IC is Individual Contributor, but what are TL, M, and TLM? | | | |
Techlead, manager, and techlead-manager. | | | | |
For those of us without direct knowledge what are IC, TL and TLM? | | | |
Individual Contributor, Tech Lead, and Combo Tech Lead / Manager. | | | |
"The argument for a pay cut" ...Another argument is that if tech paid everyone the same, everywhere, it would greatly exacerbate income equality issues in a lot of places. Tech is certainly not altruistic, as you point out, and generally wants to pay market rate rather than output rate or whatever. But it doesn't make less true that it would worsen a whole bunch of social problems by paying people the same everywhere. | | | |
That is a ludicrous argument. The whole income/wealth inequality is not about working people earning more, it is about capital owners earning vastly more due to various reasons.Some businesses do earn more due to ease of access to additional capital, and hence can pay employees more, but the root of the income/wealth divide is certainly not employers paying employees more. It is certain employers (capital owners) getting disproportionately more in the first place. Tech employees are simply in sufficiently high demand relative to supply that they can demand a piece of the pie. But there is no point in trying to address a problem by addressing its symptoms rather than causes. Specifically, in this case, paying Alphabet employees less in the name of helping the income/wealth divide only helps other owners of Alphabet stock be even richer. | | | |
"That is a ludicrous argument. The whole income/wealth inequality is not about working people earning more, it is about capital owners earning vastly more due to various reasons."Uh, what? No, it's about where the money is concentrated and how that concentration changes (or doesn't) over time. It's true that other forms of earning have a bad effect on this, but i'n not sure why that's particularly relevant. To whit - tech folks, paid in SF dollars, but living in various cheap ares of the US, are in the 0.1-1%. Easily. The fact that they got there by being paid rather than some other form of wealth accumulation, is irrelevant. They will stay there just the same as anyone else. Why would it be any less unequal because someone is starting in the 1% by getting paid a ridiculous salary on a regular basis rather than earning money through the stock market on a regular basis? Why would one not transform into the other?It's rhetorical of course, since that is what will happen.In fact, it's what the tech folks moving want to happen. They want more money to buy a large house and invest and retire and be richer than they can be in SF. " But there is no point in trying to address a problem by addressing its symptoms rather than causes." You haven't, at all, explained why putting more people in a seriously unequal position makes anything better, or why it's "pointless" to avoid doing that. It's not pointless. It avoids more inequality. Does it avoid all of it? Of course not. But again, the other way demonstrably makes things worse. Do you have a concrete argument as to why it makes sense to keep doing that? Because your entire comment feels precisely like the rest of the discussions around this - "it's not my fault, i shouldn't earn less money, it's the other people who are at fault. Particularly the ones who have 10x/100x what i do". | | | |
> paid in SF dollarsThere are US dollars, there is no purpose in framing something as SF dollars. > You haven't, at all, explained why putting more people in a seriously unequal position makes anything better, or why it's "pointless" to avoid doing that. They are not being put in a more unequal position. An economy with money flowing in is helping lift people out of an unequal position. Maybe not all at the same time, but if anything, it is helping in the long run, not hurting. > It's not pointless. It avoids more inequality. Does it avoid all of it? Of course not. But again, the other way demonstrably makes things worse. The other way is money staying with the company? I do not understand how paying employees less and letting the employer and hence shareholders keep more could be better than spreading wealth to more employees in more places of the goal is to reduce income/wealth inequality. The other way is some places continuing to stay poor and some places continuing to stay rich. > Do you have a concrete argument as to why it makes sense to keep doing that? > Because your entire comment feels precisely like the rest of the discussions around this - "it's not my fault, i shouldn't earn less money, it's the other people who are at fault. Particularly the ones who have 10x/100x what i do". The argument is that you do not want to disincentivize people working to earn money. That is a good thing for society. Go ahead and increase marginal income tax rate/implement wealth taxes, but there is no reason to regulate prices between market participants. If tech ends up paying more, then more people will be attracted to tech, and supply of labor will increase, thereby reducing the inequality. | | | |
Another another argument would be that income inequality is not the fault of employees but rather the fault of management, whether politically, culturally, socially or all of the above. | | | |
There's a few things here that are subtly wrong.> How promotions work is a committee of higher-level engineers (eg L7s consider promotions to L6) and will have 10-15 packets to assess. These will get stack ranked by the committee. If you're above the line you'll get promoted. If you're not, you won't. Ruth's change just made that line a little higher. FWIW this is no longer true for a lot of orgs, and will likely soon be untrue for all orgs. Also fairly sure that stack ranking within a single promo committee was never how it worked. > That only applies to the salary plus bonus (since bonus is derived from salary) Currently, however management is on record stating that this will change in the future, since I don't know exactly how, I'm not going to speculate, but the intent is for equity to reflect the compensation regions in the US eventually. | | | |
10% of the company is at L6 or higher so it can't be that hard. | | | |
Most of those got promoted by going into management and lucking out that the product they worked on grew and therefore needed to hire a lot, and you easily get promoted as you start to manage more people. Getting promoted on technical merit on the other hand is really hard. | | | |
This seems like a super-biased view, and having sat on tons of promo committees for IC, ICTLM, and EngMgr for many years, i'd suggest to you it could not be more wrong.Managers don't get promoted for growing teams. I've watched lots and lots of people who write "grew team from x to y" get feedback on their promo packet that this isn't, by itself, a valuable thing. | | | |
I guess their real promo is when they interview elsewhere and tell how they've been managing a big hoard of ICs. A few years later they come back and tell the story how they've been a director/VP/etc. and get a matching level which is above the level of that guy who wrote "not a valuable thing" feedback. | | | |
Google doesn't matches levels based on title when hiring.The main cases of "title inflation" of this kind are related to M&A rather than hiring. That's obviously much rarer. In hiring, the only thing the title/size of org buys you is an initial guess at level. (IE i get asked to interview the person for a potential director position). | | | |
I didn't say you get promoted for growing a team, I said you easily get promoted when you manage a lot of people. | | | |
This is also, in my experience, wrong.But you also started with "lucking out that the product grew" as your view for how that happens too. | | | |
TLM L6 here. I completely disagree. "Be a manager" is sufficient only to L5. Simply managing a team, even a large team, is not going to build a case to L6 at all. The primary challenge that L5->L6 TLMs and EngManagers face is finding time for doing the other stuff beyond just the baseline of management to build a case for L6. | | | 
