How Are You Staffing Your Startup?
protocolostomy.com
protocolostomy.com
2) Awful central point (about the sysadmins). Early start-ups are staffed primarily with developers because developers are what start-ups need early on, before the product is launched. Put it this way: if you have a lion running after you to eat you, carrying some extra water and food might be very wise for the later part of your trip, but is completely irrelevant at this point in time, and you're better off throwing it to the ground to run faster. First, you have to escape the lion, or get the start-up off the ground and get users to love it. Until then, every penny spent on "problems that will hurt you later" is wasted.
Start-ups must be super-optimised in that manner, unless they're floating in an ocean of cash - but that's not the case of most start-ups, and that has other problems.
3) More awful points: "They’ve also given very little thought to how to enable their workforce to communicate". If your stealth-mode technology start-up needs to give thought to how to enable its workforce to communicate, you're fucked. Pack your bags and do something better with your time, stop wasting your money.
4) "they’ve either assumed that systems folks’ jobs are so easy that it can be handled by the developers" >> If the 2-3 developers in your pre-launch start-up aren't capable of sorting out their own development environment, you're fucked. See point above. If you've got a lot more than 2-3 at this point in time, you're probably fucked too.
5) Hiring someone that specialises in X isn't the only way to do X. Huge fallacy here that because "start-ups don't hire specialised sysadmins", therefore "they don't care about systems administration". That's utter bullshit. Good start-up developers are generalists and can do anything that needs to be done. Those who can't, aren't good start-up developers.
In short, this article is a fantastic example of premature optimisation applied to start-up hiring strategy.
I do think scaling should be a consideration from the beginning, but it is sufficient to have someone on the team who has actually worked on a high volume web site. If your team doesnt have a developer with the experience, ask around. Most developers are willing to contribute a few hours of advice.
If anyone has questions about architecture or scaling I have built some pretty large (>500 million queries/month) sites and I like helping out. You can find my email in my profile.
We're targeting a well-established business vertical and support is an important component of our offering. We had been planning on hiring a more entry-level person and keeping more of the resources focused on development, but I can keep coding until we hit beta and we can push that hire back a bit.
Still, if support really is so central to your business, that might make sense.
You shouldn't have to worry about things like systems administration, email, and scaling from the start. If you are, you're probably not focused on the most important things and may well run out of runway before you ever really need them.
OTOH, if you focus on your app and your customers and are so successful that your infrastructure can no longer support you: what a great problem to have.
Good to see we agree on something for once.
When I read your comment here, I thought how much better you said the same thing.
I don't remember disagreeing with you much, now I have an excuse to explore searchyc later today :-)
Happy hacking!
If you pretend that production is a mere afterthought then you're likely to end up being Twitter. Successful, yes, but scrambling to stay in business (I'm not talking about the revenue model issue here).
But I don't spend unnecessary money before I have to and if I spend more than 5 or 10 minutes thinking about scaling or optimization, I quickly snap back to reality and focus on my app.
The trick (as in most things) is knowing where to draw the line.