The Great Icon Experiment – A Great Failure

“Yesterday's failures are today's seeds.” – Martha McCormick

I Con Do It!

For the past 6 months and for the next 200 months, I plan on uncovering as much information as I possibly can about the abyss of the iTunes store and app marketplace. There is seemingly no clear path to success the way there is on the web (yet) and I am going to continue to experiment until I find out what the hell is going on. This latest project was designed to see what the delta on an icon and screenshots were for downloads and conversion rates.

Quick Background

I built the Oscar app using the exact same gaming engine as Alpha Combat but with a completely new look and feel. Cartooney, retro, and simple – but used virtually the same meta description (not keywords). It's actually a lot of fun to play and I thought for sure it would get me some good downloads and revenue. I did none of the external marketing I did for Alpha Combat, including a dedicated website, YouTube video, paid reviews on app review sites, and a handful of other things.

The app launched as a free app and I was giddy to see how it performed. Take a look:

Um – yeah, not that great. Like – horrible actually. So, instead of getting down, I made a list of what could possibly be causing such an epic failure which what is seemingly a fun spin on a winning business model. Things that could cause this problem, controlling for the Alpha Combat elements:

1. Icon
2. Screen shots (essentially the graphics)
3. Keywords
4. Name
5. External marketing

The keywords and name change would require a deletion of the app and the external marketing is much larger test than what I was willing to put in, so the obvious answer was to update the icon and screen shots.

Make Me Beautiful So People Love Me

I hired my go-to designer out of Eastern Europe to whip up some hot new graphics for me. I needed a new icon and screen shots. I sent him a bunch of examples, explicit instructions, and let him run with it for a week. Here's what we came back with:

A definite improvement, at least for $150. I was stoked. Next step was to release an update with this new found aweseomeness and start booking tickets to Hong Kong for a long weekend celebration. Umm….yeah not quite. Here's what happened when I released the update:

On that first day I took a big sigh of relief – finally! I was going to back in Alpha Combat world where the money poured like candy and the schnozberries tasted like schnozberries. The problem, however, is that this new improved conversion rate didn't trickle down beyond those that already had the app. The new icon and screen shots were appealing to people who saw them, but no one was seeing them. Ugh.

Remember the Titans – Keywords Crush It In App Land

After a few weeks of depression and retro induced panic, I snapped out of it all and gave the app a serious audit. What is going on? Why does Alpha Combat bury this app on raw downloads and eyeballs? I was looking through my iTunes Connect dashboard and reviewing all the data I had put in for the Oscar app and had a jaw dropping moment – there was only one keyword there: “arcade”

Well, great. At least now I have a better idea of why this is not doing anything for me. So here's what I did to get going on that.

I deleted the app from the app store (which is why you probably can't find it if you're searching right now) and erased all it's data. I took the Xcode project and updated all the necessary data, graphics, GameCenter, and in-app purchase data to reflect a new binary upload. I changed the name to Alien Combat – Galactic Space Invasion for two reasons – keywords and the fact that it starts with the letter “A.” Then I added the following keywords (you're allowed 100 characters):

alien invaders,space wars,space invaders,super laser,alien,atari,space ace,cows in space,mr space

Notice that all those keywords are HIGH VOLUME GAMES and not generic keywords. I have a theory that one of the biggest reasons why Alpha Combat is killing it is because I put in MetalStorm as the primary keyword – type that in to the app store search and you'll see why I get so many downloads – the icon looks just like MetalStorm and I just ride all their marketing horsepower. Who knows if this is the truth, but you'll see that in Alpha Combat, the second highest “Customers also bought” game is MetalStorm. Go figure.

Waiting Is The Hardest Part

Now I wait a week to see what this sort of update will do for the overall volume of the game. In summary, here are the biggest takeaways from this experiment:

1. Icons and screenshots INCREASE CONVERSION – a greater percentage of people that see the app will download the app
2. Icons and screenshots do NOTHING FOR INCREASING EYEBALLS – they do very little for getting more visits to your page and getting in front of people

So, this experiment is slowly turning into a question of how to get lots of people looking at the app, then converting them with awesome graphics. I'll definitely report on this as the info comes in.

Don't forget to sign up for the email newsletter to get the first look at the outcome of this experiment!

See ya!



  • Daniel VanderMeer February 21, 2012

    Thanks for the article, Carter, good stuff! Do you think increased competition and app saturation is one reason for not getting as much attention in the app store? When did you release Alpha Combat?

    I’ll check back in to see how the experiment came out.

  • Carter Thomas Carter February 21, 2012

    Hey Daniel,

    That’s a really good point. The market has definitely become more saturated in the last few years and months and marketing within that has become an art. Here’s a bit of background info that I believe will help paint a picture for you:

    Alpha Combat was released Oct 19, 2011. It has about 200,000 downloads to date. It’s made probably $7,500 in that four month period, putting it on track to make about $23K for the year if I continue the same rate I am now. In case you haven’t read the other articles, Alpha Combat was a gaming engine that I licensed from a developer who has had great success with the games he designed on the exact same platform – $70,000 a year per game. He released his in 2009 and there is a marked difference in the performance as the years have continued because of your very point – competition.

    He did a very similar experiment and found that a more poorly designed version of his master game made about 25% of the original – everything else was the same, except that the new game actually had Openfeint – it was a better app! But the design was much less exciting. I’m going to do something like this with Alpha Combat in the next few months, except I’m hoping to go the other way, and make Alpha Combat the lesser of the two.

    What’s most interesting is that he predicted my sales almost perfectly just based on “how it looked.” There is no data for this – it’s a combination of the design, name, description, everything. It’s more of just a gut feel type of thing where you know what something is going to do. My take on that is that you can apply science and data to give yourself the best shot, but you have to really put your heart into something if you want it to sky rocket. I put my heart into Alpha Combat – I definitely didn’t into the Oscar Retro app.

    I think competition is a very important thing to recognize, but I also think it’s critical to realize that the profile of apps out there is staying the same – a lot of crappy ones, some OK ones, and a few good ones. The overall volume of that profile is increasing, but the bell curve isn’t. Similarly, the audience is increasing at an alarming rate.

    What this means, to me, is that great apps are going to win irregardless of competition or the space being crowded. Until everyone is making awesome apps (never going to happen) all one needs to do to win in the app game is to make a great app. Look at websites – there are literally millions of them out there, yet you can still get to the top if you build a great one…just because there are so many that aren’t.

    With this experiment, it’s going to be very interesting to see what sort of traffic the keyword and name changes bring. If they don’t do anything significant, I may have to just accept the fact that this app doesn’t have that greatness I’m talking about and that it’s better for me to spend my time on another Alpha Combat – esque effort and project than 5 more lesser projects.

    Thanks for your comment,


  • david February 25, 2012

    hi,carter, really awesome article. question, why on date 1/19/12 the download is the highest? what happened there?

  • Carter Thomas Carter February 25, 2012


    That’s when I released an update with the new icon, along with an updated in-app purchase. That moved my app up the charts on “recently updated” which is why you see the spike.


  • Tim February 27, 2012

    hello carter, can you tell me how to make an in app purchase model for my own app? thanks

  • Carter Thomas Carter February 27, 2012


    Can you explain further? I’m not sure what you mean by an in-app purchase model.

    Basically, you have three options for in-app purchasing best practices:

    1. Buy additional features (more guns, more coins, etc)
    2. Unlock parts of the app (levels, etc)
    3. Upgrade to premium version

    Determine which of those fits best with your app and go from there.


  • david March 1, 2012

    carter, i have a question. i made an weather app that users can interact with the weather. and i name it as god of weather. can i put the app in both game and utility catogory?

    if not, how can i put my app in the game catogory?
    please answer me asap.

  • Carter Thomas Carter March 1, 2012

    Yes – you can put it in both. With Games, you’ll be asked to put it into two sub categories as well. This will happen when you are in iTunes Connect and are getting ready to publish your app.

    I outline this in my book that I’ll have in a few days – make sure you sign up for my newsletter so that I can send it to you.



  • Serge March 20, 2012

    Hi Carter,

    First off: this is great info. Thanks for sharing this on your blog. (I will be a regular visitor now). I am developing my own app and I’m already thinking about the marketing. Here is a question: I read, during my research, that Apple does not allow app names as keywords for your app.

    If I read your blog, I conclude this is not true? I do understand that Apple won’t approve Angry Birds as a keyword for your app Alpha Combat because those games are not simular..

    Thanks again man!


  • Carter Thomas Carter March 20, 2012

    Hey serge,

    That’s a great point and I should have been clearer about that. Apple does not like (they say they don’t allow) keywords to be app names, but when I use them, it still works. In fact, the alpha combat using metalstorm keyword has worked for over 4 months now. Plus, the latest case study I did shows how big of a boost it provides.

    So – yes it may not be exactly what Apple wants, but for now it works and Ive got the proof to show for it.

    Thanks for reading,


  • Serge March 20, 2012


    Thanks for your quick response. That sounds great! I will also try this. The release of my app is still at least a month away, though.



Leave a Comment