Apple Squash – Realistic Ad Revenue (part 2)

Observations and Recommendations

First off, let’s get straight to the numbers. I’ve only included PubCenter revenue this time around because its fill rate has been fairly high and MobFox hasn’t been serving much of the leftovers. Ad Duplex has been picking up roughly an additional 200 impressions a day. On the 29th of Feb (highlighted below) I released a content update that appeared to have a minimal impact on impressions.

If you’d like to try the game these revenue numbers come from you can download Apple Squash from the Windows Phone marketplace here.

So what are my observations based on these numbers and the numbers from my previous 2 games?

Impressions not filled by your primary network are unlikely to be filled by any network
In this experiment my secondary network (MobFox) filled around 5% of the impressions PubCenter didn’t take. I’ve seen similar results with other combinations of providers. My assumption is that these unfilled ad requests are likely to be coming from a country that no provider cares much about. This is why Ad Duplex is the perfect secondary network since it doesn’t care about region at all.

1 (paid) ad provider is enough
If your app is extremely popular then maybe the small percentage a secondary network will give you will be worthwhile, but for most I don’t think it would generate the minimum $50 most networks require before paying you. So just pick your favourite since they’re all very similar. I have personally tried MobFox, PubCenter and AdMob and for me the best was PubCenter.

Use Ad Duplex as your fall-back
It’s silly to waste impressions your paid network can’t fill, but chances are other paid networks can’t fill them either. Ad Duplex fills almost all requests and will convert them into free advertising for you.

Ad networks need volume
This one might just be my experience but it would seem that to get a decent ecpm (and therefore decent revenue) your app needs to be generating enough volume for your ad provider to care. Without volume you will likely get a poor ecpm and this is one of the reasons I recommend 1 ad network is enough. The 2nd ad network is unlikely to get enough volume to make the impressions worth anything.

ECPM is crazy volatile
My ecpm started at 0.12c and reached as high as $4.37. That’s 36 times higher than my starting point! This makes it very hard to predict revenue with an ecpm that can fluctuate so wildly and is really just something you’ll have to accept. If your ecpm stays consistently low it might be time to consider switching networks. I actually think the recent ecpm numbers I’ve been getting with PubCenter are abnormally high and I’m just waiting for the bubble to burst…

Keeping users engaged is the key to Ad-based revenue
Apps that keep users coming back for more are the ones that make money. Games like Apple Squash that are don’t have a lot of lasting content will eventually fade away into obscurity. The numbers above show this is already starting to happen. So if you don’t plan to follow the path of lots of new content and updates, then the only way to prevent the inevitable decline is through good marketing and getting your app featured in enough places to keep a steady stream of new users coming in… If anyone has tips for me on this I’d love to hear them!

Apple Squash - Realistic Ad Revenue Expectations

Last week I released a new game to the Windows Phone marketplace. It’s a simple action game that took around 2 weeks to create in my spare time. The game is called “Apple Squash” and you can find it on the Windows Phone marketplace here.

For this game I built a custom ad rotator that displays ads from PubCenter, MobFox and AdDuplex in that order. If ads were not available in the first network it would try the next and so on. PubCenter is Microsoft’s own advertising network and is regarded as one of the best for Windows Phone, particularly for ads served in the US. MobFox seems to perform better in Europe while AdDuplex generates no revenue but instead generates impressions to advertise your app in other apps that have the same control. 

Apple Squash has been on the marketplace for just over a week and I’d estimate has had around 2,000 downloads so far. In my opinion this is a fairly typical example of your average performing app. Below are the results of its first week live.

Take from these numbers what you will but the purpose of this blog post is not to dissuade you from building apps or from using the ad-based model, it’s purely just to share one developer’s experience so that you can add it to the experiences of others to come up with your own measure of success. There are plenty of examples of success stories out there for Windows Phone (see Taptitude or Krashlander), but for me this is the more realistic example of what you can and should expect.

Stay tuned for more blog posts on how Apple Squash performs after the "honeymoon" period.

Where to from here?

Now that the Nokia challenge is done and dusted it’s back to my original project that I’ve been working on here and there for the last six months. It’s a spin on the Tower Defence genre where instead of having your enemies follow a set path you actually need to dig deep into the ground to release them and they will follow the path that you created. I honestly underestimated the amount of work involved, particularly in the graphics area. I’m only now starting to see a light at the end of the tunnel. Maybe in another month I’ll have something good enough to release as a public Beta? 

Yeah I’ve been saying that for a while…

Here’s a couple of screenshots of what it looks like now. Maybe releasing these will spur me on to finally finish?

 

How much is possible in a weekend?

Microsoft and Nokia recently put out a challenge to Australian developers: produce apps for the Windows Phone marketplace and score a free Nokia Lumia. 3 apps would get you a Lumia 710 and 4 apps would get you a Lumia 800. This challenge is still running and details can be found here.

Now let’s be honest, I don’t know why anyone would go for the 710. It looks like a great phone but if you’re already developing 3 apps, developing a 4th for the awesome Lumia 800 is not much of a stretch.

With only 50 phones available this competition was all about quantity not quality. I quickly dumped all of my original ambitious ideas in favour of anything I could build, test and submit in half a day. In fact all of my apps could have been done in less but I couldn’t bring myself to publish something that looks like sample code stolen from the web.

Below is the end result. The challenge was great fun and I’m pretty proud of what I was able produce inside of a weekend. It's a testament to the great developer tools and submission process available for Windows Phone that you can go from design to submission in around 6 hours. The only sour experience I had was when Fast Dice failed certification first time around (a rookie mistake with the White Theme!) but the resubmission took 6 days to be retested!

So please check out my submissions for the competition. They’re all free and don’t contain ads. I actually find Score Keeper and Fast Dice to be really useful and I reckon they’re as good as any other similar app out there.

Score Keeper

(marketplace link)

Reaction Timer

(marketplace link)

Fast Dice

(marketplace link)

Flash Memory

(marketplace link)

More than just code

Chances are, when you first start out as a mobile game developer you’ll quickly realise there’s a lot more to a game than just code. It seems obvious, but you forget that all the extra components like graphics, fonts, leaderboards, music and sound effects that go into make a polished product can easily take longer to put together than the actual code.

Thankfully you don’t need to reinvent the wheel and there are a lot of great resources out there already. Daniel Spruce over at Escape Development has put together a great list of free developer resources that you can check out here:


http://escapedevelopment.blogspot.com/2011/12/free-developer-resources.html


While I generally always do my own graphics and art, I’ll freely admit I’m hopeless at music and developing your own sounds effects just isn’t practical. I’ve used almost all the resources on this great list.

One resource in particular I’d like to highlight is Font Squirrel.


http://www.fontsquirrel.com/


The difference between using an appropriate, quality font in your game and using the default font is massive. And in terms of effort it’s one of the simplest and easiest ways to make your game look professional. I didn’t use special fonts in my first game Eclipse since I was worried about font licensing but now that I’ve found a quality set of free fonts I’d never use standard font again (unless I was creating a Metro UI of course!).