Why this edtech startup decided to rebuild its product from scratch

January 9, 2017

In the tech world, where customer needs and expectations are constantly evolving, your product has to evolve, too. But sometimes, limitations in the way a core product was designed mean that the only way to keep up is to start over from scratch.

Two years ago, Albert.io, a Chicago-based edtech startup, found itself in precisely that position.

Originally named Learnerator, Albert was founded in 2011 by Luke Liu, an undergraduate at Northwestern University at the time. By 2014, the company had built up a sizable customer base for its platform, which helped high schoolers study for advanced placement exams through interactive testing.

But its core technology product — the development of which had been cobbled together on a shoestring budget — was quickly becoming more of a liability than an asset.

“We knew that the code base on Learnerator was really poorly done,” said Liu. “It was completely unmaintainable, it was buggy, it was slow, there was no way we could build a mobile app and there was no way we could do cool things like weighted averages and interactivity.”

After much deliberation, Liu came to the conclusion that the code base would have to be rebuilt and got the process underway in January 2015.

While the engineering challenge of rebuilding a tech product from scratch is daunting enough, Liu was concerned about how customers would respond to the new platform.

To complete the rebuild in a timely manner, Albert’s engineering team had to make hard decisions about which features to prioritize. In the short term, that meant taking away functionalities that paying customers had gotten used to — a downgrade, albeit a temporary one, for the sake of creating something that would be better in the long term.

“It was exhilarating, but also terrifying,” said Liu. “Here you have all these users using Learnerator and getting value out of it... At the same time, they were asking for all these features that we knew we wanted to do, but we just couldn’t give it to them.”

The rewrite ended up taking more than a year. Albert onboarded the first customers onto an alpha version of its new platform in December 2015, but it didn’t officially shut down Learnerator until the summer of 2016. And although a lot of functionality has been added since then, Liu said his engineering team still has an enormous backlog of functionalities it's looking to add.

Liu said he launched his company in the high school AP test prep space primarily for pragmatic reasons. In the time leading up to launch, tests like the ACTs and SATs were undergoing a lot of changes and he was concerned about spending limited resources on content that would be outdated in two years. While the ACT and SAT test prep markets were fiercely competitive, the smaller AP market was largely untouched by the tech sector.

Now that its redesigned platform is in place, however, the Albert team is aggressively moving into college and graduate school entrance exams as well. Liu said that transition has been fairly straightforward, both because of the versatility of its new platform and because the content for SAT and ACT test prep is easier to create than high school AP content.

The rebuild gamble paid off, Liu said, even though the process was painful at times. Since 2014, the startup has scaled from three employees to 36, and it’s still growing. But despite the rapid rise, Liu said he’s been lucky to find great people for every open position.

“There’s not a single person on our team that I wouldn’t re-hire,” said Liu. “Really, a huge part of my motivation now to make this company successful is to be able to keep working with them, and to be able to hire more people like them.”

Images via Albert.

What's your company's story? Let us know with a tip or a tweet @BuiltInChicago