Two Dimensions of Tech Scalability
Every investor, and we are not an exception, is interested in the scalability of business. And because many companies are built on technology, it involves scalability of technology.
In conferences, you hear about systems scalability – how to build a website to handle millions of visitors; how to store petabytes of data and search them; how to pass a hundred thousand messages per second thru your Kafka cluster.
These topics are sexy for geeks, me included. And they are necessary from a certain point. But that’s just one dimension of scalability and I dare say the easier one to handle. That’s because it’s already fairly well established field and there is a ton of material and tools to help you. Cloud platforms removed the necessity to build datacenters. Thanks to tools like Kubernetes it’s easier than ever to run auto-scaling apps. Traditional relational databases are able to handle unbelievable workloads these days.
Don’t get me wrong. I’m not saying these things are easy, but it’s easier to tackle than the second dimension of scalability and that is scalability of your tech team.
It’s great that your product was developed by your genius of a co-founder and it’s the most efficient piece of code on the planet. But how about when you actually do get that investment and you’ll need to hire ten more developers to work on it? Suddenly, things other than your search algorithm being 10 milliseconds faster than your competition begin to matter more.
Is your code understandable and readable? How long does it take for a newcomer to deploy their first task to production? Is your system divided into logical modules or microservices, so more people can work on it? Do you have a fully automated build and deploy process? Does the build include a test run? Do you monitor and measure all important business and technical metrics so you can immediately react when deployment goes wrong?
These are the questions we consider when evaluating scalability of technology. It’s not just about those thousands of requests.
More articles on