Scaling a business

We Ruby on Rails developers are used to the barbs. “Ruby doesn’t scale” they say. Ignoring the fact that Twitter famously deals with eleven thousand requests per second.

For those of you for whom the above was gibberish, let me explain. Scaling a web application is making it deal with large numbers of simultaneous users. That in-house application that happily deals with all thirty people in your office using it at the same time may fall apart if thirty thousand people start using it concurrently. And, in this age of Digg, Techcrunch, even Slashdot, getting hit by a sudden spike in users is quite likely.

But there is a problem with scaling. Namely the processes and procedures you need to follow to make an application scale can sometimes be in direct opposition to those processes you need to follow to make the application fast for a small number of users.

There is a similarity here with making a business scale. I have spent all of my career working for small companies (although many of my clients have been large). The reason? Just like a small-scale application can be fast and responsive, a small-scale company can make quick decisions and run fast and loose with its procedures. This means that every staff member needs to do a bit of everything, every staff member has a say in decisions, every staff member can make a real difference.

The problem comes when the business starts to grow. How do you keep your corporate culture when suddenly you find that “newcomers” out-number the “old-timers”? How do you keep that “folk knowledge” (no, we don’t do it like that - because two years ago X happened and we nearly lost two clients) when your staff no longer fit in a single room?

Now this is looking far-ahead for me. Currently, 3hv is a one-man band (the odd contractor or freelancer aside). My procedures live in my head; no-one can do the wrong thing because I get to decide what is right and what is wrong.

But I fully believe in starting as you mean to go on (reserving the right to change my mind at any time of course :0)). And, more importantly, as things stand now, there is a limit on my income. I have to eat and sleep, and most importantly, I have a family. That means I cannot spend more than ten to twelve hours per day earning (and ordinarily eight to ten hours is all I can manage). So, if I am paid by the hour, the only way to increase my income is to increase the number of chargeable hours. And the only way to do that (apart from a time machine) is by bringing in more staff. Yet once I do that, how can I guarantee that the quality of the work (the thing upon which 3hv’s reputation depends) is maintained? I don’t want to be hiring and then throwing away the extra hours by monitoring everything my staff do. At the very least, that is going to be totally demoralising to the staff.

So instead I must start to scale my business from the off. Document my procedures and processes. Emphasise (and write down) the values that make 3hv what it is. Share them with my customers, my (future) staff, my acquaintances. Like scaling an application this may make me slightly less responsive now. But, when the time comes, it means that I can grow as quickly as needed with no loss of quality.

And, of course, this being the age of transparent internet business, I intend to share these processes with you. Why? I get feedback (from you, dear reader), you get to watch 3hv grow and (hopefully) come back here to see what I will come up with next. And it should promote trust between me and my potential customers - as they can see exactly how 3hv works.

You know you want to subscribe to find out more. Or maybe you just want to hire a web developer who will talk to you in plain English and deliver you some code so beautiful you will cry with joy.

Photo of Oresund Bridge is by sundstrom.

Posted on December 14th, 2007 | By: Rahoul Baruah | Filed under General

Leave a Reply

You must be logged in to post a comment.