The founders of Frontback originally made a much bigger, more complicated photo-sharing app. When it wasn’t getting off the ground, they were faced with the big problems of high investor expectations and low valuation of a seriously complicated product. The app was called CheckThis.
“We got stuck. We had raised a lot of money, but we had to change the valuation of the company by proving to investors that they should invest more,” says Frédéric della Faille, the CEO and designer of Frontback.
Eight weeks from running out of money, they did a quick pivot, validating on Facebook. “The way CheckThis was designed, it let you combine a bunch of different pictures. I had this idea of using the both cameras of the iPhone and only doing two pictures," della Faille explains. "I put one photo front and one photo back. I shared it on Facebook; I got a lot of comments from friends. I looked really really tired. They loved it.”
“My cofounder came from Brooklyn for a month. I said I think we should build this. We took four weeks. We built it. And we didn’t tell our investors," says della Faille. "Then we released the app. We had 100,000 downloads that first month. That completely changed my relationship with my investors," he says.
“We really did that as a hackathon fighting against time, removing features, shipping the fastest as we could.”
“We realized that only one thing mattered: the content. We wanted only one action: the act of photo creation and only one action for consumption--liking the content. At first we had no profile. We had no notifications. We had no list of likers. You had no idea who liked your content, or when. You had to go through the feed and see the post from two hours ago. But basically people don’t really care as long as the main action is there,” says della Faille.
“Then week after week when we saw the number of users growing, we thought ‘Okay let’s continue to work on these things.’”
But first: They had to tell their VC, Steve Schlafman, what they built.
“One Friday night I went out to dinner with Fred and Melvyn at Siggy’s Good Food in SoHo. I’ll never forget it,” says Schlafman. “I can even remember where we were sitting. It was at that dinner where they told me about it.”
“They were doing it on the downlow, but at some point they had to spill the beans,” says Schlafman, “It was certainly a surprise to me. But they really had no other options. At the time they only had eight or 10 weeks of cash left. I had two choices: to lambast them or say ‘I’m going to support you.’ Taking the latter approach was more pragmatic.”
“They showed me the alpha on my phone. I instantly got it,” says Schlafman.
It certainly didn’t hurt that the demo was well built, or a product with a uniquely viral appeal. But Schlafman said his reaction as an investor was based more on his belief in entrepreneurs--and his expectations around seed stage pivots.
“I think some investors would view being in the dark as being duped. I sort of come at it, particularly at the seed stage, that you get one or two pivots. I generally believe, and maybe I’m naive when I say this, that if an entrepreneur can clearly articulate why they were pivoting, it can work.”
“I also think every time an entrepreneur has their back against the wall they need key allies," he says. "It was one of those moments where they had to go all in. I like to think that my pep talk helped them get their heads around burning the boat.”
“From CheckThis, we had experience on building feeds and could code without thinking. We’d already learned about mobile,” says della Faille. “But when you build an app like ours, you don’t really realize that every feed is different because of the people. You have to build the feed so that when the user opens the app, the feed is there.”
“This is a JPG of our first prototype (above)," says della Faille. "I basically first did some drawing--took me two days. Then we built this interactive prototype using POP for iPhone--took me another three days. And Melvyn coded this into a beta we could touch--took us three weeks.
Cofounder and lead developer Melvyn Hills adds, “We learned a lot while working on CheckThis, we made many mistakes, also on a technical point of view. With this knowledge we were able to build something much simpler as an MVP and still very solid with the right technology choices that could evolve until now to handle more and more traffic without a hitch. We used the Heroku and Amazon platforms to be able to deploy and maintain the whole API very easily with only one backend developer.”
It also comes with surprises. For instance, how to build a custom feed. “When you build an app like ours, you don’t really realize that every feed is different,” says della Faille. “Building a feed can be complicated for multiple reasons: They contain lots of information from very different sources, must be infinitely scrollable, pageable, sortable, and last but not least, they must be really quick to load, as users expect them to be available instantly.”
“We had a bad prior experience on CheckThis with complex MongoDB normalization techniques, which was overkill for the size of the app and resulted in unexpected behaviors, slowness, and was hard to maintain. We went the simplest way using some basic caching mechanism added to PostgreSQL's basic but powerful relational database features.”
“More advanced techniques will probably become a necessity at some point for Frontback,” he says. but so far the K.I.S.S. (Keep It Simple, Stupid) approach has proven to be the best for us.”
“Because we started from scratch, we didn’t have any constraints," says della Faille. "I could see the trend of cards. Now in mobile, everything is about cards. Think about Tinder, Snapchat, every piece of content is one screen. With Frontback we didn’t have any constraint of a product we had to switch to mobile. So every piece of content was one swipe, created for one screen. That’s the problem we had with CheckThis. It didn’t really fit for mobile because the content was too long or too short."
Smart Design creative director Dan Saffer literally wrote the book on microinteractions. He agrees that cards are the way to go for people building new apps. Here’s what he has to say about why they work psychologically and which apps are doing them especially well:
“Cards provide an immersive alternative to lists or grid views or even menus. The reason we like cards so much is that they dissolve the Paradox of Choice, turning what could be an overwhelming list into a simple ‘Do I like this one? Yes or No?’ or ‘Which one of these two?’”
He adds: "The physics of designing cards is very important. The gesture itself can be very satisfying, to dismiss a card by flicking it away. It’s like crossing off an item from a to-do list. It can feel like progress. The active daters in our office swear by Tinder, because flicking through the cards feels like speed dating, but also like progress. There’s a dopamine rush in that you don’t exactly know who you’ll see next, what’s behind the next card. Traditionally, uncertainty is a bad thing in interaction design, but with a card-like structure, it can be used effectively, just as long as you don’t need to remember where a card is in a stack.”
“Focusing on microinteractions can be a very valid strategy, particularly in a crowded market where your features (like Frontback's) are basically commodities," he says. "Take the Mailbox app, for example. They focused on the microinteractions--even for signing up for the beta--and were acquired by Dropbox before they even officially launched.”
[Image: Flickr user Andy Rennie]