How I lean startup’d my way to $240k+ on the saturated App Store
I cofounded a social app startup at age 22.
While I learned a lot about programming ambitious apps and working with high caliber people, we failed to find product market fit.
Our strategy was basically “Let’s brainstorm ideas and ship massive features and hope people want them”. And once I started learning Lean Startup approaches, the team gave me lots of heat for trying to implement them.
Two years, zero sales, no friends outside of work, and countless failed iterations and pivots later, we ran out of funding.
Wanting to prove that I could use Lean Startup methods, I opted to go solo.
Looking to avoid building an idea that no one wanted, a friend told me she was looking for a password keeper. Thinking I was being “Lean” I went with it.
I saw that there were definitely other password managers like OnePassword, saw some of the flaws they had, and took that as “validation” that people wanted a new one.
I did all the app & website dev and design and made them all as nice as possible (and you can still download it, its a fun app). It took five months, all my savings to ship.
It flopped. We lived a poverty lifestyle so my over-engineered app and website could have negligible downloads. My girlfriend had 90% lost faith in me.
We scrambled to find money to afford a trip to the animal ER to save my dying cat Footfoot. We borrowed money to save his life.
My cat, finances and (ex) girlfriend were all victims of my shitty approach to entrepreneurship.
The pain taught me to execute with more tact. The next few months led to the start of an App Store portfolio that I ended up selling two years later, netting over 1 million downloads, nearly $240k sales and a modest exit.
How I found a legit idea
Here is the process on how I was approaching ideas:
- Get a tasty idea in your head
- If Competition exists: I could make this so much better
- If No competition exists: The market is all mine muaha!
- Romanticize about all the awesome features you will build
- Start building the product immediately.
- Make it perfect. Then make it perfect some more.
- Build it in complete isolation. No on can see it until its beautiful.
- Start marketing it after you launch it
Turns out this is actually the guide on how to almost kill your cat.
I’d learned that if I were going to commit to building anything, an idea would have to prove itself over and over that it was worth building, and 99% of ideas fail to do that.
Which means I needed lots of ideas. I needed a habit of ideas.
I wanted to generate ideas that I got from other people because making up ideas on the spot is lazy and almost never works. So I built the habit of finding ideas that other people gave me clues for.
This means asking people about their problems. I learned to tell people I was an app developer and ask people for their app ideas. Even had my waiter tell me one.
It also mean immsersing yourself in people’s problems online. Seeing what they’re saying. I’d message them and talk about their problems.
I wrote down every app idea I discovered. I’d force myself to spit out a few every day and ideas quickly became my life.
After turning myself into the Sherlock Holmes of iPhone app ideas, I finally found one that worked.
GifShare came from someone telling me about needing a better solution for posting his Gifs to Instagram since Instagram doesn’t let you do it in their app. They told me because I wrote a post on Reddit about being an app developer and they reached out to me and eventually I asked them about their problems.
How I validated before building
“Post GIFs to Instagram” passed the research test (which all my other ideas failed):
- People were actively searching for a solution.
- Some people were even making some clunky workaround to solve it already.
- Traffic, as indicated by Google Trends, Keyword Planner, were good enough for an app
- There was none or very little competition
Old Me would’ve immediately jumped in and built the best product ever.
But New Me knew too much pain which I highly recommend not experiencing.
How could I validate without building an app? Aha! A landing page.
I used QuickMVP, threw together a promotion in like 30 minutes that said “GifShare: Post GIFs to Instagram” with a call to action that said “Download for $1.99”. I made an awful app Icon that took me like an hour.
I wanted it to be shitty. If users took action on an ugly landing page, imagine what they would do when I optimized it later.
I posted a link to my landing page on Youtube videos, Yahoo Answers and forum questions I’d found that were ranking high for relevant search terms like “Instagram Gifs”.
Within these pages, others were talking about the problem and seeking solutions (do you see a trend? Get people involved before you build!).
Curious users were flooding to my site, but they wouldn’t click the download button. Was the price too much? I changed it from $2.99 to free and people started clicking like crazy.
Since I didn’t actually have an app yet, I asked them for an email after clicking through. Emails were coming in. More than I thought would.
I was ready to build.
From a crappy app to $900 in first month revenue
I cobbled the whole first version together in a week.
As minimal and “just good enough” as an app can be. Same app icon as before. Stock interface design. Free to use icons. A single App Store screenshot with a sentence or two for the app description.
I even outsourced part of the app’s solution on StackOverflow using a bounty :)
To make money, an In app purchase to remove the watermark from the app.
Compare that to the 5+ months I did on the previous app. I spent hours grinding out every single detail to perfection. And for what? (Spoiler: overwhelming anxiety & shame)
I wanted to see some beta testers to prove an extra level of validation, because beta testing an app at the time was a chore. I had to get them to click through an email, sign up for an extra service, give me their iPhone UDID which they’d have to look up how to do, then recieve a download link through email then HOPE the app works.
I thought if I even got 10 people to endure that process, that’d be enough. I got 20. I even set up a fake In app purchase to see if people would remove the GIF’s watermark to see if they would pay.
Many of them paid. I finally had the confidence to launch.
The first month of launch did $900ish. I wasn’t about to quit freelancing, but if I could figure out how to “growth hack” this bad boy, maybe I could soon.
Growth hacking my way to $200k in sales
- Getting new customers
App business = big # of downloads. At its peak, I believe my business did ~1.2k downloads/day
After MUCH experimenting (failing), I found a few things mattered here:
- Improving the App Icon
- Improving the App Store screenshots
- Making the watermark that showed on the gif to be prettier and more noticeable. This worked because people would see “Made with GifShare” on Instagram and download my app from there.
- Encouraging good, abundant reviews
Keywords barely appeared to matter for me though it mattered for other businesses. Neither did the App Store description. Test test test test.
2. Making more per transaction
This was simply adding more In app purchases (paying for filters, fonts, etc) and… charging more for the IAP that was working well (removing the watermark).
The original version, I charged 1.99 USD. By the time I sold, the same IAP was 3.99USD. All extra IAP were .99 or 1.99.
I tried ads, and they added decent revenue. ($600ish a month). However, I eventually took them down because I felt a bit scummy — so scummy that I actually lost motivation for working on the app until I removed them.
3. Creating more loyal customers
I struggled the most improving loyalty. Users often would use the app once or twice then never again.
I found a few things worked pretty well though:
- Get people to follow the app’s Instagram account
- Build an email list and email them occasionally for feature updates
- Make the app a super usable, fast, and rewarding to use
- Encourage sharing multiple GIFs (after they finished one, I’d try to get them to create another one)
The Process
In order to discover these changes, I wanted to move as fast I could which is a challenge on the App Store when you’ve got release cycles that sometimes last a full week or longer.
To get around this, I used a backend service like Firebase and I would turn on/off various changes (like what color a button should be, should I show certain things or not). This allowed me to test if something worked or not without having to wait for a new update to push.
It’s important, no matter how slow/fast you move, to only make one change at a time.
If you enjoyed this article, like below and sign up to my newsletter and I’ll teach you how to make apps (and other software) that sell like crazy :)
More where this came from
This story is published in Noteworthy, where thousands come every day to learn about the people & ideas shaping the products we love.
Follow our publication to see more product & design stories featured by the Journal team.
