judging AppWall effectiveness over time

I was using Leadbolt AppWall HTML ads - but recently added some logic so DIFFERENT urls are shown on different days.

I was wondering what the effectiveness of the AppWall is over time - within the same session, and over subsequent sessions, and over subsequent days.

For now I have just implemented logic so a different URL (since Leadbolt allows you to create different additional AppWalls) is shown on day 1, 2, 3, and beyond 5 all lumped into one URL.

It will take some days to get data for this - but should be interesting.

My guess is that the AppWall MAY lose it’s effectiveness when presented beyond the first day (and possibly the first session ?).

Which would suggest that instead showing a full screen ad maybe preferable at that point (i.e. for jaded users).

Now, I could be wrong - as the AppWalls may change over time - but from my location I usually see the same apps in the AppWall - so if a user has ignored it once they may ignore it again.

Another alternative is to alternate between two ad networks for the AppWall - for example Leadbolt and AppBrain (however in my testing my app was crashing occasionally when I did the AppBrain AppWall - the same one I show at the end of the app - but showing it early may have been a problem - or maybe it was some obscure bug with the Leadbolt AppWall fancy prefetch stuff).

The stats should indicate if it is true that the majority of revenue from AppWall is derived the first day.

With finer granularity analysis (if one does URLs changing everytime the AppWall is shown) - one could perhaps identify if the MAJORITY of the AppWall revenue is actually from LEAVING users - i.e. users who are NOT long-term users of your app - but are those who were app-surfing i.e. looking around for apps - the next suggestion for the app they received took them to that new app download.

If this is true - it could raise the intruiging possibility that the app and it’s long term users are providing LESSER revenue than the users who are NOT interested in your app.

I have included GetJar so users can potentially purchase features etc.

However I am seeing very low interest - the revenue is lower than Admob banner ads even (from 1500 new installs a day there is hardly even 1 feature unlock using GetJar).

Now there could be other reasons - the app is bad, the features are not desired, GetJar is slow (new SDK is faster).

But there is a question of competition for user attention also - if I were to put up a full screen interstitial reminding users of the Free Upgrade option, it would be competing with the Leadbolt AppWall.

The question to ask is - do you gain more by showing the advertisement for free upgrade/getjar feature ?

And perhaps knowing how well an AppWall performs, this could be split up this way:

If the AppWall starts underperforming (say it drops to 1/2 or 1/4 it’s revenue per impression or eCPM) then it suddenly becomes less attractive than say Greystripe full screen ads (which are paying $0.8 eCPM). If the Leadbolt full screen ads don’t require EULA (?) then one could show those.

One can also show a FreeUpgrade/GetJar advertisement at that point - and highlighting that then may give better revenue from the GetJar.

Perhaps the more detailed analysis of same-day performance of AppWall - i.e. how it’s performance goes down by repeated presentation - could also be used to decide whether to start showing full screen ads (instead of AppWalls) … ?