How to Scale
22 June 2015

How to Scale? 3 directions from examples

Every entrepreneur has dreamt of the day when their own company will have large headquarters, employing thousands of workers, and carrying a globally renowned brand. They know it is not easy, that it is only for a few, but they all long to become the next Google, Facebook or Twitter.

The development of tech giants shows that making this dream a reality is not impossible. One of the key ways to do this is scalability, a startup’s growth model: it contains steps startups have to take to start generating income while keeping costs down. All the big players of Silicon Valley have faced a similar same challenge. How have they managed to overcome it?

1) Scale technologically

After verifying that the business model works, the first thing to take into account in order to maintain reasonable costs is technology. Todd Underwood, engineer from Google, explained at the Usenix LISA (Large Installation Systems Administration) conference in Washington that automating all possible processes for each of the company’s products helped reduce costs while increasing productivity.

Twitter have experienced something similar. August 3, 2013, was a turning point for the little blue bird company. On that day, over 143,000 Tweets per second were published in Japan during the airing of the animation movie ‘Castle in the Sky’. This was 25 times greater than the global average of Tweets per second, which at the time was at 5,700.

Despite the spike, users did not encounter any incidence on the social platform. “That’s one of our goals: to make sure Twitter is always available no matter what is happening around the world”, explains Raffi Krikoria, VP, Twitter Engineering in a blog post. By then, the infamous ‘Fail Whale’ which appeared too frequently during the platform’s origins, was already a recent memory.

The leadership team at Twitter had learnt from their mistake during the 2010 South Africa World Cup. The spike of Tweets caused the apocalyptical whale to appear too much, and that was something that a $1 billion-worth company could not afford. Twitter’s engineering team set goals such as reducing the time a Tweet takes to be published, as well as the number of machines the company needed for the ‘microblogging’ platform to work, and spotting small errors before they caused a system breakdown.

“Not everything we tried worked and not everything we tried, in the end, met the above goals” explained Krikoria, “but we were able to settle on a set of principles, tools, and an infrastructure that has gotten us to a much more desirable and reliable state today”.

2) Scale on infrastructure

Many companies have turned to Amazon Web Services (AWS) to relieve their server load when their aim was to get millions of users. Pinterest, Netflix, Reddit or Spotify are some of the startups that recognised Jeff Bezos’ platform as the technology that would allow them to grow without affecting their accounts. Why would they keep their own servers if they could use someone else’s?

That was the same question that Snapchat founders asked themselves. However, they followed a different path. Instead of AWS, they chose Google App Engine to build their App. It is a free alternative during a startup’s first steps, which allows you to benefit from the infrastructure of the almighty search engine without spending a penny until you reach five million users per month.

It is a key issue, but sometimes a few technological changes are not enough to guarantee a company’s scalability. This is what Airbnb learnt when it saw its users multiplying after changing the photos of rooms for rent and replacing them with high-resolution pictures. This change was far away from the code that had doubled their weekly revenue to $400 per week.

3) Scale on hires

As soon as a startup starts seeing itself as needing to scale, changes produced are not only technological, but also in employment. The business grows, together with the number of employees. Obviously, managing a team of 100, 200 or 500 workers is far more complex than coordinating a team of five to ten people.

In order to handle the situation, almost all the big companies of Silicon Valley decided to organise their engineers in small teams, granting them the ability to making decisions in order to solve problems quickly or improve a product without the need of having the leadership’s approval in each and every step.

However, each company has its own philosophy, and has to organise its team accordingly. Aditya Agarwal, VP, Dropbox Engineering, explained this in an interview. He used to manage a team of 2,000 workers in Facebook when the social network already had 700 million users (in 2011). Currently, he coordinates over 200 engineers working for the storage service.

“At Facebook, we were famously moving fast and breaking things; in Dropbox’s case, the focus is on maintaining quality”. It is about protecting personal data of hundreds of millions of users, a delicate task that requires greater supervision.

Increasing efficiency from a technological viewpoint, making yourself known through marketing, and giving greater responsibility to staff, the number of which is growing simultaneously; these are some of the keys for an optimal growth of a startup. This recipe should be coupled with effort, and the wish of achieving the dream of becoming a giant, even if this is something that is already present in the entrepreneur’s DNA.