I have some experience with GetJar Rewards.
Firstly their SDK is difficult to incorporate - there is no easy front end like AppBrain etc.
It is nitty gritty callback upon callback - first for user authentication then for licensing etc. etc.
Some of that complexity is because they have put out demo app/code that portrays a maximal use of the SDK - so it is overly complicated.
I had complained about some of the issues with GetJar in this thread:
http://forums.makingmoneywithandroid.com/advertising-networks/1013-10-000-full-screen-interstitials-per-day-what-shouldi-earning.html
Thread: 10,000 full screen interstitials per day, what shouldI be earning?
Joe at GetJar had posted in that thread that they were making it faster (I can confirm it is slightly faster than before). And to improve on the demo code.
Bottom line - from my experience is - that I have spend a huge time on GetJar which I often wonder if would have been better spent on something like Tapjoy.
The thing that I found attractive about GetJar was that they were a clean model. However that is not matched by the execution.
In addition, the GetJar SDK is slow, complicated and needlessly (as I said even the UI is maximally designed - too big graphics etc. which needlessly slows things down - or perhaps they need to upgrade server end).
To make matters worse - it looks much worse than Tapjoy - every time I have asked testers if they prefer the GetJar way of paying vs. the Tapjoy - they almost universally say they like the Tapjoy (it is perhaps the colors that Tapjoy employs).
So GetJar really needs to get it’s act in order - they are throwing money on other things it seems but not on the basics - perhaps they are improving the server side, but they need to bring in some new ideas on the UI and developer code side.
So in summary:
- GetJar pays you $0.9 for every 100 coins that a user “pays” you (GetJar keeps the $0.10)
- the SDK is a PAIN to implement
- the UI is universally seen as inferior to Tapjoy
I don’t know how complicated Tapjoy integration is - but I have been seriously thinking about dumping GetJar (even now I was at a deadend as I was trying to revamp the GetJar UI I am presenting to the user - but I BALK at approaching or modifying that code !). I spent literally 3 weeks trying to clean up their demo code to some form that I could understand - I was hoping to make it simpler - but even now I balk at modifying the code (the reason is that in those 2-3 weeks I got burned numerous times as I modified the code and it wouldn’t work - eventually it did work and I now understand the mechanics better - but STILL I psychologically am wary of changing things too much). The model is asynchronous with callbacks for user authentication (which can timeout - so you have to ensure that you do it again next time you want to license etc. and so on). They have tacked on recommended pricing etc. stuff as another callback (which needlessly complicates the demo code for a new user). I think after following complaints they have removed the recommended prices callback stuff. In earlier versions they also had a please wait dialog which prevented doing user authentication in advance - but now that is possible - but again as I said the user authentication can time out also (!) so when you go to check licensing (if product been bought before) it may still take time to user authenticate etc.
The only thing which prevents me from dumping GetJar is that I’ve invested so much time in it - and my initial liking for their clean model (plus the company is generally above board i.e. is not tacky like Tapjoy).
But I fear they are not putting the time into these developer-centric issues. They probably need to get some new blood in there to shake up this side of the business.
Either that, or they are planning for the next few years - and so are not bothered by using big graphics etc. (thinking the internet will get fast for everyone around the world) or something …
In practice
I have been using GetJar - and the revenue was initially 1/10 of the banner ads (!).
Now it is approaching the banner ad revenue.
However, I feel this is primarily a weakness at my end - i.e. it seems in-app billing cannot just be bolted on - i.e. people expected to just get the free version of the app.
The app has to be designed with that in mind - i.e. some essential features should be kept for future unlocking, OR (and perhaps better for the developer) a more pay-as-you-go model where the user gets some initial “coins” which get used up and then they wind up having to buy more (in order not to frustrate users, can offer a daily reward of coins just for coming back to the app). This model can be seen in the CoinDozer app.
For this reason, I am sticking a bit longer with GetJar (but also thinking about Tapjoy) - and hoping to use such a model in next app.
And in THAT context it MAY work.
Even then, I have a feeling that Tapjoy may still outperform GetJar purely from the user reaction i.e. they don’t mind waiting for the Tapjoy and it’s “offers” - which can be many compared to the options GetJar gives initially (though if the user has the GetJar Rewards app they can get credits for just about any app they download - i.e. GetJar it seems is willing to pay out of pocket to jumpstart their model). But in order for this to work they should put more options on the screen the user sees (in a way the earlier GetJar model was better as it had a better selection of apps).
The reduced set of apps on main screen may be a sign of the limited number of advertisers that GetJar has - and this may wind up hurting their viability.