Businesses are expected to grow, and the websites and applications that support them must be able to grow too. Scalability is a core concern for many founders. But building an application that can smoothly scale from a handful of users to tens of millions is challenging and complex. It takes expertise, time, and an expensive infrastructure platform.
Early-stage startups have limited resources. Is building to scale the right way to invest those resources, or is it better to focus on building and marketing a great product. Should you choose infrastructure that can scale to support millions of users, or start small and worry about massive scaling when it becomes an issue?
Scalability has been defined as the ability to increase resources and have performance increase proportionally, where performance refers to the amount of work the system can do. Most websites and applications are scalable to a point. When your WordPress site gets too big for a shared hosting account, you can migrate it to a dedicated server.
The same is true of bespoke web applications, but I’ve often spoken to founders who don’t think that a simple server solution is enough. Most are building businesses based on a CRUD app. They plan to build it from the ground-up to scale out across multiple servers, and so choose a cloud platform that allows them to spin up servers on a whim. As developer Jacques Mattheij wrote several years ago:
Web developers the world over are always terminally worried about whether or not their application will scale orders of magnitude without keeling over. It’s a funny thing, because plenty of those applications would have stood a fighting chance if their builders were not that worried. Worries like these lead to premature optimization, and premature optimization is a huge timesink.
Premature optimization consumes time and money. The more effort a founder plows into building their app to scale massively, the less time they have to add new features, to iterate on the features they have already built, and to create a product that people want to use.
There are some exceptions: if your business model requires millions of users a month, you have to think about scaling early on. Twitter might have suffered fewer growing pains if more attention had been paid to scaling in its early days. Most businesses don’t follow the volume model: they sell a product or provide a service ” and they’re being paid for it.
These businesses don’t need to attract a million visitors a month. They need to attract a substantial but much lower number of visitors who convert to paying customers. Resources far better spent optimizing the site’s features and the businesses sales funnel than on premature optimizations for scalability.
Limited resources should be spent on solid and reliable infrastructure, rather than on a massively scalable cloud platform ” throughout its life, a physical server is less expensive than the equivalent in cloud infrastructure, and it’s likely to perform better too. A dedicated server hosting an application built with standard open source tools like PHP and MySQL can support hundreds of thousands of visitors a day.
In the early days of your business, choose software and infrastructure that will help you to get up-and-running quickly. The ability to scale your site by orders of magnitude will not help you to grow your business, because your customers don’t care about scalability ” they care about your product and what it can do for them.
