Why We Canned a $2M App Business to Make Indie Games

My name is Ben Lochmann, I’m the CEO at Benjamin Lochmann New Media GmbH, a German game developer, now better known under the brand “Pixel Maniacs”. I thought I’d take the time to write down our story –  a successful app developer giving up its business because of falling in love with the desktop gaming industry.

It Began With an App

The whole story began during a train ride with no internet. I was bored and decided to play around with my new Macbook Air. The only cool thing I found was an Xcode documentation. I managed to “code” (it was more copy & paste) an app together called “Mäxchen mit Fake Button”.

“Mäxchen” is a popular German drinking game, somewhat similar to Liar’s Dice.  The app had a “fake” button hidden in the top left corner, which would make players aware of it win every round they’d like to.

This first app went live on Feb 14, 2011. It sold quite well for the effort I had invested into it.

(I took tons of pictures of a dice cup to build a perfect dice cup animation. Very fancy)

At that time, my main job was doing arbitrage performance marketing campaigns, which is a complicated way of saying that I searched for well performing affiliate offers that were paid per lead or sale, and booked ads on Facebook and other portals on a pay per click basis, to promote them.

Going Legit With an Actual Company

Two weeks after my first app went live, I invested that money into employing our first dedicated dev to build apps, Johann (https://twitter.com/zumJohann).

We then ramped up the headcount quite quickly in the first year:

Chart

We had no office at that time and started building apps at my place. This, of course, made little sense after more than two employees, so we moved to into our first office.

Back in 2011, you had great visibility on the App store without a lot of effort, since there were “new” sections and only very few apps.

Over the years, getting free visibility was getting harder and harder, so we started using our performance marketing knowledge from investing in ad campaigns, for our apps.

The only way we promoted our apps was through burst campaigns, which meant investing our app’s entire marketing budget on day 1 of the launch – and nothing afterwards. We mainly invested the money into cost-per-install-based ads, most of which went to facebook:

The campaigns were always focused on one type of device (iPhone or iPad) and one specific geo. The reason for this was that the main goal of a burst campaign is to send significant numbers of downloads to an app, to have it rocket up either the Google Play or Apple’s AppStore charts.

As you can’t afford to get into the top charts of every store (if you’re not Supercell or something), you need to concentrate your budget on a specific store. The iPhone and iPad rankings were different, which is why it would not have made sense to mix them.

In addition to those paid campaigns, we added a cross-promotion ad into all the apps we had launched so far – which is still quite helpful as, we can exchange marketing power with other app developers, , on a day-to-day basis.

The AppStore and Google Play – They’re Not the Same

The AppStore’s algorithm was quite quick to give feedback to this method: When investing ~$1,000 per hour, you could sit in front of the charts and wait for the next 3 hour-update, to see how it reacts. After a couple of campaigns, we got a good sense of whether a campaign  would flop and were able to stop it asap,  to not lose too much money. Sometimes, you could even reduce the marketing budget as you could tell an app was going to rock the charts, without any further investment.

Google Play was much more tricky. Contrary to the Apple AppStore, they have a section called “Top New”, were you can only get ranked in the first 30 days after your launch. If you reached a decent Top New position within the first week, you usually kept rising in the Top New section, until the end of the 30 day period, which gets you to Top Overall. The problem here was that you should invest all of your marketing budget on the first day of the launch. You can’t wait 1-2 weeks until you get the first retention KPIs, as you’d be losing valuable “Top new” visibility during this time. We tried to reduce the risk of investing heavily (“Heavily” in our dimension means $10,000 - $20,000 in the first 24h our the launch.) on an app which had a bad retention rate, by launching on iOS first, since there’s no Top New there. If the retention on iOS was high, it usually was high on Android as well, and we knew we were able to market the app properly.

Another difficulty on Google Play was, that the chart algo only updated the rankings once per day. This meant we couldn’t adjust our campaign spendings during the 24-hour campaign, in accordance with the rankings – as one would do on Apple’s AppStore.

With these technique, we reached several #1 Top Overall positions in many countries, and got more than 25M downloads on the AppStore and 21M Downloads on Google Play.

Risk-Analyzing an App – Without an App

We tried to reduce the risk of an app’s failure, well before we created it, by booking ads for the non-existing app, and sending the traffic to a landing page, which indicated if users clicking on the ad, would actually download the app (if it had existed).

Apps with 0-100k downloads usually were flops, 57% of our 152 apps did not exceed that mark, which is a fail-ratio we could live with.

Here’s a chart showing the number of downloads, our various apps reached:

Chart

A Watershed Moment: Our First Game Jam

In April 2015, my collegue Andi had the idea to participate in a game jam called “Ludum Dare” – where teams around the world create a game around a specific theme within 72h. I honestly had never heard of game jams before (shame on me, I know), and was quite sceptical. We participated, just for fun. The whole team met at the office on that weekend, many installed Unity the first time, and tried to build a game that matched the theme: “An unconventional weapon”.

Our unconventional weapon was color, and so came into being our first game, ChromaGun:

We had so much fun that weekend, building an actual game, unique design and all, that we made the obvious decision to finish the project and launch it on mobile. The first Ludum Dare version was built as a WebGL prototype, and received great feedback from the other participants, so we were quite confident about the launch.

It completely ran into the ground on mobile.

When I say completely, I really mean completely, in the most literal sense of the word. We got around 100 downloads – nothing that would justify the months of work we invested into our biggest project so far.

Being distraught that this – in our opinion – really cool game had no chance of becoming anywhere near profitable on mobile, even though it had received such great feedback from players, and various media outlets, somebody from the team brought up the idea “why not launch it on Steam?”.

It sounded like a fantastic plan. Obviously, the game’s issue was the platform, and the controls, so making it a PC game would solve all our problems. A few clicks here, some minor adjustments there, and we would get tons of sales from Steam’s gigantic user base.

On Feb 16, 2016 we launched ChromaGun as our very first game on Steam. It was awesome having your first own “real” game out, and strangely,  it was a much bigger achievement for us, than all our #1 overall positions on mobile before, with more marketing-driven projects.

Of course, it had no chance to bring in enough revenue to finance the 10 months of development we had already invested into the project. We did absolutely no pre-launch marketing – simply because we didn’t know we should –  so I started promoting the game as I did it before with the affiliate programs and the apps, on a performance basis.

This was quite difficult, as there’s no Facebook SDK to measures installs of your game, as you’d do on mobile. As a performance driven company, we never invested a penny into marketing without knowing (or calculating) what we would get back from this cent.

So we built a small workaround, sending clicks through our own ad server, and adding a small HTTP call into the game, which made it possible for us to match (within a margin of error) if a client had installed the game after seeing one of our ads.

I treated promoting a desktop game the same as promoting any other product online.  If you know the bare minimum about video game marketing, it won’t be a surprise, when I say those campaigns were among the worst performing campaigns I ever witnessed in my entire performance-marketing-career. It was time to find out why.

How Promoting Apps and Games are Supremely and Entirely Not the Same

Our app revenue came from ads. The large majority of our apps were free, and stayed free. Contrary to a lot of other mobile companies, we had no significant revenue from in-app-purchases, or direct sales, as the following chart shows:

Chart

Getting a potential customer to download your free app, with a well-optimized app title and app icon was no big deal –  Getting them to consider investing $15 of their well-earned money however, was.

The customer journey we experienced during our time as an app developer was something like this:

A customer journey for desktop games often seems to look more like this:

No chance for my fancy conversion tracking script to measure anything useful here after a user clicks on an ad.It took us (especially me) some time to realize that we needed to radically change the way we promote our games. For years, we didn’t even have a brand. The apps were simply published under my name “Benjamin Lochmann”, as nobody really cares about this on mobile, if you are offering rather small, casual games that aren’t very deep in content.

Pixel Maniacs: Genesis

Finding our brand took several weeks. We collected ideas, rated them internally, and checked the legal side of the best ones.

We were looking for  a name that: