Do startups work well when managed remotely?

14 points by jmtame ↗ HN
I've had limited startup experience. I do primarily design and LAMP development. Currently doing JSP (new to MVC architecture but making the change fairly smoothly, usually I script and do front-end design).

I joined a startup just this past summer that received some funding from a popular VC firm and worked there as an intern over the summer. I'm accustomed to being in the incubator, and being in the presence of the co-founders, while I work.

However, in my current position, everyone works remotely from their own locations. There is office space, but it's out of the VC firm, there's not really dedicated incubator space for us. The product is already developed for the most part, we're just focused on growth.

The entire team (about 6 and counting) operates this way. Can anyone chime in and let me know what your experience and opinion on remote work on a startup is like? Do you feel it's effective?

9 comments

[ 2.8 ms ] story [ 32.2 ms ] thread
37signals champions working remotely here is one example of their praise of the topic, but you can find many more on their site.

http://www.37signals.com/svn/archives2/getting_real_the_alon...

For me, in my limited experience in the area I would have to say that it is completely dependent on the quality of the people and the level of autonomy their job responsibilities can afford.

http://pikluk.com was built out of 5 different locations in 2 different countries without even a central office. The only "central" thing we had was a Subversion repository.
This is the way my startup is currently operating. There are certainly challenges that come with that. You need to be internally motivated on a day to day basis, and it's important for the team to have a high-level plan that you can work towards so that you know whether you're "late" or doing well, but overall it's working fairly well.

Another key element is to ensure there is a touching point each day, where all the team members are online at the same time.

I second the high-level plan and the brief daily calls.

I'll add that if you're working from home, put your shoes on when you start working and don't forget to spend time outside.

Sure startups can work well remotely but they work even better in person.
Let's consider a more extreme version of the question. Do startups work well when all development is outsourced?

Assume you have a very capable outsourced team. You will naturally have more difficulty communicating with the external team, both in terms of expressing requirements to them and in getting feedback from them. On the other hand, you might get better people and be able to work with them on a more flexible basis. If you can understand and communicate your requirements in written form, and they don't change very rapidly, this may be a worthwhile trade-off. If you underestimate your product's rate of change, or overestimate your ability to communicate requirements, your life will suck. This is extensively documented in the lore of outsourcing.

Distributed teams are incrementally better than outsourced. You get better people, there is less turnover, and the team members are focused on your project. You might get decent communication if everyone uses chat and a wiki. But you'll still have to work hard to communicate, and you'll probably misjudge your difficulty/capability ratio. You're essentially copying the model of open source software development, and should take note of where that model has been successful.

Remote work is great because developers can focus for hours at a time without interruption. If you are at a stage where most people are working on big chunks of implementation or fixing bugs, this is a huge productivity boost. The other major benefit is that you can hire great people who are blocked from most tech jobs by their location and/or family obligations.

Remote work sucks because communication is more difficult. A group brainstorming session conducted by phone is, optimistically, one fifth as productive as an in-person session. If you are trying to generate ideas or generate a specification, you should expect this to be very difficult to do by phone.

As far as I can tell, having worked for about 2.5 years in a distributed tech company, the determining criterion of success in distributed work environments is specialization. If you have separate people in charge of design, marketing, and coding, it works well. But this often isn't the case in startups; sometimes your best design ideas come from some hacker with an offhand interest in usability, or you decide you can't find a good marketing person and you'll just have everyone pitch in a bit. Those sorts of group-work decisions imply a communication cost, and if you don't realize that your team will feel strangely hobbled and directionless.

Remote collaboration can work. It depends alot on how the goal and objectives is defined and relayed to the team. There is a strategy project consulting company in SF Bay Area that trains project teams to collaborate remotely.
One more thing. it is all about the process preceding the technology not the other way around.
You need to have an excelent software development cycle. I am in the US, my partner and clients are in Zurich, and my development team is in Cali, Colombia, South America! and it works perfectly!