One thing I’ve always disliked is visiting someone’s website or reading their bio and only seeing The Hits. When things are going well for a person, it’s easy to assume they’re magical. But odds are they have a string of missteps in the past which have since been swept under the rug.

I don’t want that to be the case, so I’ve long kept a list of my “failures” on my website. None of these are projects I regret, and many of them did alright in the grand scheme of things. But each of them underperformed what I had originally hoped.

Let me know if you make your own page like this:

📬 Arrows Outreach (2019)

We launched Arrows Outreach back in August 2019, and it landed with a thud. There was a lot of interest at first, via Twitter and Product Hunt. But only a handful of folks ever really used it. There was an interesting strategy involved, where we wanted to build a CRM but couldn’t invest multiple years into developing something worthwhile. So we figured we could launch a handful of small tools which, over time, could approximate to a fully-featured choose-your-own CRM type of product. It still might be able to work, but it’s probably not the best strategy for our skillset.

🧾 Arrows early explorations (2018-2019)

Before we made Arrows Outreach, we made a bunch of never released tools for Arrows. A few of them we charged for, but most of them failed for a variety of reasons. The tool was wrong, people didn’t understand it, people didn’t want it, we couldn’t replicate the experience for multiple customers, and more. Many of these iterations and explorations will make their way back into the future version of Arrows, in some form or another. So it’s not totally lost! We just spun our wheels for over a year before we found some traction in a direction we like.

🛋 Housecraft (2017-2019)

Asher Vollmer made a prototype app for placing furniture in AR when the first ARkit SDK launched. Then he posted a video on Twitter which got ~7,000 likes. So we decided to turn it into a real app to launch alongside iOS 11 in September 2017. We felt strongly that we could make a little money off the app, even before launching, because it was really good. Asher had a relationship with Apple, so we assumed we’d get some nice App Store attention. All said, Housecraft in March 2020 has over 625,000 downloads and is still listed on the AR Apps page in the App Store. It was also featured in an Apple commercial for augmented reality. But at the end of the day, we had a few really interesting opportunities to get partners or sell it, and they all fell through. I did ultimately find a buyer, but the money was tiny and they don’t seem to be doing anything with it (we still own a small percentage of future earnings).

💨 Shortwave (2015-2018)

Back in spring 2015, Joel Corelitz came to the office I was working out of one day and showed off a cool way for a group of people to control music together using their phones. It was an interesting prototype, and I immediately saw the possibility for it to be more. I pestered him until he let me help out. We improved the code and the UI, starting showing it off in an art installation-like format during events and parties. It was a hit with the people who played with it, but never really found its footing. We explored a lot of possible routes with it, but none really made sense for us.

📦 Driftless (2016-2018)

I had a week-long client project scheduled, which would’ve paid me a few thousand dollars that I really needed at the time. But then at the last minute the company told me they didn’t have the money anymore and cancelled. So I pulled from an old list of project ideas I had and tried to make some money to cover my upcoming rent payment. The idea was to ask the question of people “What’s the last useful thing you bought on Amazon?” The distinction being useful, meant to capture those odd things you buy which end up being awesome but aren’t necessarily something you’d ever recommend publicly. I also thought the question and objects were uniquely interesting and might lead to a semi-viral effect. It made money by using Amazon’s affiliate system, which shared a percentage of all sales on Amazon for the 24-hours after clicking on a link. And since a lot of folks would click on Driftless links to see more about the product, then just buy other random things on Amazon, I assumed it might do okay. And, honestly, it did. The biggest issue was getting sustainable traffic to the site (do you see a trend in all these projects when I was younger?!). If I had a good idea for traffic, it probably could’ve become a decent source of income for me.

🖼 Mocky v2 (2015-2016)

Once Mocky v1 didn’t take off like I’d hoped, one of the reasons I thought was that it required users to build a new habit. So I decided there was a really interesting product in building tools and UI on top of file changes in Dropbox. Unfortunately I spent months dealing with technical challenges, and never really doing full-on sales and customer development. So by the time the product launched, it was interesting but not actually that useful to most people. Ultimately I was out of money and out of energy, so I let it die a few months later. RIP.

💰 Hermes (2014)

I’d mostly been building products first, then trying to launch and sell them. So i decided to change tactics. I noticed lots of friends came to me for help finding jobs and also for hiring people, and I also helped a lot of people with salary and stock negotiations. It seemed there wasn’t enough liquidity or clarity in the jobs market for designers and developers, and I wanted to rethink it. But I wanted to learn the system first and get paid to develop a product. So I essentially became a matchmaker-style recruiter for tech companies and designers+developers in Chicago. It worked decently well and I got paid a little bit, but the business model was flawed and I got burnt out being a recruiter. It wasn’t fulfilling and I didn’t see a path forward for a software product.

💪 The Shakedown (2014-2015)

There was an app which launched in early 2014 which I thought was pretty clever and funny at first. It was called Yo, and all it did was allow you to “Yo” your friends, which sent them a push notification that said “Yo”. That’s it. It blew up because of the absurdity, and I even was convinced by a few people that it was oddly useful in certain situations. But… then they went off and raised $1-2 million in venture capital, and I became super annoyed. These people took a clever, simple, fun idea and then ruined it by trying to force it into the VC ecosystem. So one day I started thinking to myself: what is the absolute most ridiculous interaction you could have with your phone, which might then trigger a notification to a friend? Shaking your was the best idea I could come up with. It was in-part a spoof of Yo, but then became its own fun idea in the vein of the original version of Yo. So what was it? All users started with 10,000 points and they’d see a leaderboard of their friends when they opened the app. When you tapped on someone’s profile, you’d see their score and if you shook your phone you would steal points from them. The harder and longer you shook, the more points you stole. And once you stopped shaking for a few seconds, it would send a push notification to your friend that you stole some number of points from them. This had the added benefit of being a socially emergent system, but the shaking also caused people in real life to ask what you were doing. It was really fun, had about 15,000 downloads at some point, and ultimately died off after a few weeks. Benedict Fritz, my current cofounder on Arrows, was my partner on The Shakedown. Some day… we’ll bring it back and do it again.

🖼 Mocky v1 (2014-2015)

After leaving Svbtle, my friend Hiten Shah said he’d give me some money to build a small tool if I had any ideas. He had a really cool 5-letter domain that he said I could use. So I built Mocky, which was a tool for collecting feedback on design work. Ultimately it failed because I never did sales or marketing and I wanted the first version I launched to take off right away. It probably could have grown into a nice little business if I’d stuck with it and continued improving the product.

💰 Svbtle (2013)

For a few months in 2013 I was a technically an employee of and a “cofounder” of a blog network called Svbtle. That was, um, interesting.

🎨 Colorplane (2013)

While building a private forum for my friends to hang out in, I decided I wanted a better color picker. I hoped it would catch on and the idea would be more widely used, but I don’t know of anything that stole the idea besides Svbtle. Luckily, I was able to incorporate the concept into OneShot a couple of years later, and it was appreciated there but still not done anywhere else.

And plenty more…

It’s even easier to forget all the ideas and prototypes and experiments which never even had a name or launched publicly at all. There were tons more.

Start small, and keep going.