Conventional engineering wisdom is to ship a product barebones, get some feedback, then iterate. But Mick Hagen wanted his app, Undrip, to work differently than other reader apps. Maybe that's because Hagen isn’t just the company's founder, he's the team’s UX/UI designer.
“There's this idea that you should get your MVP (minimum viable product) out there and then you're off to the races getting user feedback,” says Hagen, whose 80-person education startup, Zinch, was acquired last year. “Whatever we deem as ‘minimum,’ I still want that to be absolutely gorgeous.”
But how does a startup app shop like Undrip balance agile development with perfectionist design and still manage to throw a few easter eggs into the basket? In any startup there's a ticking clock: Good design takes time, and getting an app to market fast is crucial for a small team’s momentum. Undrip is an especially intricate experience, one that leaves you with the impression it took years to create. Yet Hagen says his team works more rapidly than most, pushing a new build at least as often as Apple approves the previous one.
"We're like a duck swimming upstream," he says. "Above water our feathers carefully placed, calm and composed, but underwater our feet are moving like crazy." Without speed, Undrip would be chum for bigger competitors like Flipboard. "If we weren't so agile," he says, "there's simply no way we could compete with them."
The use case for Undrip is similar to any catch-up-on-your-day social network reader, but with less extraneous nonsense. Where news readers try to auto-curate articles based on general interests, Undrip merely pulls content out of your social feeds and displays it in a beautiful, digestible and easily shareable format. Content is sorted into articles, videos, photos and so on; the most popular stuff bubbles to the top.
While Undrip has made subtle visual evolutions since its beta launched in June, the backend, which Hagen cobbled together after teaching himself to code this past summer, has since been gutted and rebuilt by his team. "Funny thing is that all our engineers were originally PHP developers," he says, "but when I started working on it I started in Python, so everyone else had to learn and adapt." The Python framework Django helped the team save time as they migrated to a PostgreSQL database; a Redis caching layer stores key values to aid its performance.
The team's constant time crunch means mistakes are tilled under without drama. "We have a quick 20-minute standup meeting every morning," Hagen says. "What will you accomplish today? Boom, done. No big processes, no big meetings, no layers," he says. "We talk about the objectives, goals and desired outcomes for each task, and the engineer takes it from there. They are the CEO of that task. He or she owns it. The vision is clear, the destination is set--now they create the path. I'm not concerned how we get there. I just need to know that we will."
Yet what spurs Undrip's team to work so quickly is as much about the code base as the fact that the engineers are empowered to bury little fascinators into the app on the fly, often without even telling each other. “When you pull to refresh, you see Mr. Rogers,” Hagen says. That’s not the weirdest part. “If you tap him, he says ‘Hey neighbor.’ That little Easter egg I didn't do.” The Easter eggs are the crux of the app's allure, says Hagen. “If people appreciate your product at many levels, if they put together the clues, they'll fall even more in love with it.” (There are more where Mr. Rogers came from, we won't spoil the surprise.)
Which begs the question: What if some engineer has too many beers on a Friday afternoon and decides to build in his version of the Aristocrats joke? “They know I'm the designer, and I'm particular. That’s why it's so important to have a CEO that values design in a big way,” says Hagen.
The rest of the team thinks before they place a pixel, but mistakes still happen. "I do my best to be tactful and nice," he says, "but when something is bad, I'm not afraid to say it. We certainly will fail sometimes--probably most times--but there's nothing wrong with it. When bugs get into the product, yes, I hate it. But I know that with the pace we're trying to move, it's gonna happen. We're not planning quarter to quarter, we're living day to day and week to week."
The secret to keeping control requires generosity of spirit and caprice. "There's no major consequence for getting things wrong," he says. "There's no guillotine hanging over our heads. I will never be upset for someone going out on a limb and trying something new," says Hagen. "What will upset me is when we stop trying."
Undrip currently works on iPhone and iPad; the team plans to release an Android version soon.
[Image: Flickr user Tracy Lee Carroll]