We've tried and found that remote engineers are less effective, they miss out a lot from group whiteboard discussions, overhearing office chatter "Hey bill, service XX crashed again because of YY (and Mike overhears and says "Oh yeah, I saw YY in service ZZ too and fixed it"), and just general team cohesiveness. In theory we could hire more people for less salary if we had a remote workforce, but we're still a small company (less than 100 people) and it's a lot harder to manage a 100 person geographically spread out engineering team than a 50 person local team). Paying the San Francisco premium for salaries (and rent and everything else) is less than the cost to build and manage a distributed team. As we grow, that will likely change as we'll be able to have specialized teams and open satellite offices, but for now hiring in the Bay Area makes the most sense.
We could likely build an accomplished team in any major metropolitan area, but since our investors and board members are here (as are the other companies they invest in and manage), it makes sense to have the company here. (not entirely our choice, when someone invests millions of dollars into a company, they want to be close to it)
I have found that this says more about the ability of the management and less about the ideas underpinning remote work. Large companies figured out some time ago that they could have a large remote workforce. So it isn't just Basecamp or whoever that is doing remote work. Remote work has it challenges but (and here is the important point) they can be managed if the team knows how to manage.
FOSS proved over two decades ago that you can build complex competitive systems with people you have never met. The fact that you couldn't make it work doesn't mean that it doesn't work. It means your company could not figure out how to make it work for you.
Large companies manage to not go out of business while still having a large remote workforce.
Large companies often have a ton of inertia, not a lot of competition, and they often take a very long time to fail. Just because they've started doing it, and are getting away with it for now, does not mean it's a good idea.
I'm not saying it's NOT a good idea, I'm just saying you can't tell based merely on the fact that they're doing it. They also have lots of employees who put in the bare minimum and punch a clock to draw a paycheck. Does that fly in a smaller, less well-funded company?
Google, a very large company with nearly unlimited technical resources, continues to bus workers 40 miles down the peninsula every day because they want workers to stay on-site. Google, of all companies, could afford as much office space in SF as they need, yet they still choose to only have a relatively small satellite office there.
The experience of large companies that have large remote workforces and FOSS projects that have workers so motivated that they'll work for no pay probably doesn't translate well to a small startup that doesn't have the resources of a large company, nor the employees so motivated that they'll work for free.
Remote work, as you say, has it's challenges, and sometimes it's better to pay the money avoid the challenges than to pay the money to confront the challenges.
Hmm... it could be argued that Google's repeated failures in the social networking space are compelling evidence that it's one of the least likely companies to succeed with remote workers.
When teams have problems with remote v local employees, I've found this to be because they actively treat them differently, sometimes without realizing it.
It takes work to establish your culture to work remotely well -- if you just hire some people that you only ever hear on Skype during standup and give them work, they don't become enmeshed in the fabric of the company. Cliques form everywhere but they can be especially brutal in excluding remote workers from the 'core' teams that are seen as successful within a company.
I'm sorry to hear it hasn't worked for you. If you attempt it again, make sure you evaluate whether you've built a culture based on 'being there' before hiring people who can't be. Lots of people make this mistake and just see cheaper workers.
When teams have problems with remote v local employees, I've found this to be because they actively treat them differently, sometimes without realizing it.
Yeah, some teams like to take advantage of the far richer communication that's available in person rather than handicapping everyone to the lowest common denominator.
If you attempt it again, make sure you evaluate whether you've built a culture based on 'being there' before hiring people who can't be.
IOW, discourage informal social interactions between employees. Don't permit them to go to lunch together and talk at leisure about things.
Companies do not like spending money they don't have to. So since the Internet hasn't killed business travel, there must be a very good reason its still worth it.
This is definitely a thing that happens. Some teams also do a lot of work in hallway conversations, and this sort of accidental interaction is critical. I witness this all the time.
I would much prefer that my company let me work remotely. The Bay Area is full, seriously, and I'd be happy to work from many other places.
But denying reality doesn't help.
I think there are potential technical solutions to this problem (VR office at home), but I suspect it'll take legislation (aimed at getting vehicles off the road) to make remote work a real thing that a lot of people do every day.
That kind of smalltalk definitely happens, but where it happens varies a lot. At some places probably 80% of the idle work-related chatter happens on the IRC/Slack, while at others it happens mostly in the coffee area or hallways. Seems to be a company-culture difference in large part. Both are ok for me, but the places that use IRC/Slack a lot are easier to be remote at (or even just on a trip) without being excluded from all the impromptu discussions.
It's actually not that noisy normally, just the clatter of keyboards.... until there are problems with something, then it can become pretty active. Most people stick with Hipchat or email for routine communications like getting stuck on debugging a piece of code or where to eat lunch -- unless they are talking in one of the break rooms, or they duck into one of the conference rooms. People tend to be pretty cognizant of the open office so if you hear people talking, it's generally something worth listening to.
We could likely build an accomplished team in any major metropolitan area, but since our investors and board members are here (as are the other companies they invest in and manage), it makes sense to have the company here. (not entirely our choice, when someone invests millions of dollars into a company, they want to be close to it)