Small company or big company?

A choice I will likely have is whether to work at a larger company… or a startup… I was wondering if you had any particular feelings on this question.

In the long run I’m hoping to work at a non-startup due to hours/general quality of life as well as salary and stability for the considerations of family life and the fact that the projects I’m interested in don’t necessarily have good prospects as short-term revenue generally. … I worry that startups are often competing with other startups in a crowded marketplace [and that working] at a startup means selecting one that will eventually “win” either through getting a hold in the industry or getting bought up. Choosing one that “loses” means my work may go entirely to waste? … Larger companies of course don’t have a great track record at significantly shaking things up.

  • I don’t think startup vs big company is the high order bit for quality of life. For instance, most people at Wave work reasonable hours and take lots of vacation, whereas literally everyone I know who’s worked at Microsoft says the work environment is dystopian.1

  • The determining factor in a startup’s success isn’t competition, it’s whether they can actually make something good enough that they don’t die. Many spaces have multiple successful companies, or successful products that used to be different companies (random examples–ridesharing: Uber/Lyft/Grab/Didi/etc.; messaging apps: Whatsapp/Messenger/Line/Viber/WeChat/Telegram/Signal; money transfer: WorldRemit/Azimo/Remitly). It’s true that the top company is often 5x more valuable than the second biggest in these spaces, but that doesn’t mean the second biggest company is useless. For one thing, preventing a monopoly from emerging produces a ton of social value :P In that sense, you might actually have higher leverage by working at the second most successful company and helping them compete.

  • That said, it’s true that if you work at a startup that is making a bad product your work will be useless. Don’t do that (to the extent that you can tell a priori). But your point of uncertainty shouldn’t be “will this startup be killed by competitors,” it should be “is this startup going to make something 10x better than the status quo,” which is easier to evaluate.

  • You can tell whether a startup has [already] made something people want by whether it’s growing really quickly (and its retention is good). At the early stages Paul Graham says 5-7% [a week] is considered good and 10%+ is considered exceptional; I don’t know as much about benchmarks for retention as it depends on the product. Fast growth is really good not only because it means you’re doing something valuable, but also because you will be forced to do a lot of new stuff and solve lots of growth-induced problems. Fast user growth usually translates directly into fast personal growth (or that’s been my experience at Wave).

  • Lack of short term revenue opportunities [for your areas of interest] is a more compelling reason not to work at startups, but there are also startups that plan to raise a lot of money and then do R&D for a long time (e.g. Magic Leap).

  • Salary: any good startup should pay you more than enough to live on [as a recent college grad with no dependents], even in NYC, if you don’t ratchet up your standard of living a bunch. But a big company would probably let you pay off loans or buy a house a lot faster.

  • In general I think the startup vs. established company distinction doesn’t add any information if you already know what the company’s product, plans and work environment are like. That is, it might be useful in your initial sourcing phase (it might be a better use of time to look at a list of “big company augmented reality initiatives” and evaluate each opportunity, than to do the same for “augmented reality startups”)–but once you’re at the level of having a few accepted job offers you should be looking mainly at the actual details of the jobs.

  • To reiterate a more specific but important point, if you care about quality of life you should DEFINITELY not work at Microsoft.

Thanks to my anonymous friend for prompting this post!