@jan75
Thanks for the very quick reply! I wanted to respond to each of your comments one at a time, since I think it’s important we flesh this out clearly.
"If you got $0.02 revenue without installs it was a CPC campaigns running and not CPI. Bear in mind RevMob has a few CPC campaigns running too."
I understand that RevMob runs some CPC campaigns, but I think it’s very unlikely CPC campaigns explain why so many of the developers in this forum are seeing $0.02 revenue and zero installs in their dashboards.
First I’ll point to my own experience. As I mentioned above, I received and clicked on several ads, and the impressions and clicks appeared in the dashboard immediately and correctly. The revenue remained $0.00, which indicated to me that the campaigns I was seeing were CPI, not CPC. I then clicked another ad, installed the advertised game, and ran it for several minutes. The new click showed up in the dashboard along with $0.02 of revenue, but with still zero installs. I repeated this process on another device with the same results. This suggests to me that the install reporting was not working. Of course, it’s possible that by pure coincidence the ads I originally clicked on where CPI (which is why I saw no revenue for those clicks), and the two that I actually installed were CPC (which is why the amount of revenue I saw was tiny). But that strikes me as pretty unlikely, since as you yourself noted, only a few RevMob campaigns are CPC, and I would have to have been pretty unlucky if both of the apps I installed were from CPC campaigns, while none of the ads I clicked on without installing were CPC. It also doesn’t explain why the install number itself remained zero in the dashboard.
Second, since so many other developers in this thread are experiencing the same issue, it’s extremely unlikely that the apps everyone installed happened to be CPC campaigns.
(Both of the devices I tested on were iOS devices running iOS 6.1.x).
"Regarding the transition you have to take into consideration your users must update their apps in order for the tracking method work properly."
I performed all of my testing in a private, pre-release version of my app using Corona 1101 with RevMob 5.2. Because it’s a pre-release version and I’m the only user, I can see very clearly whether each action I take in the app does or doesn’t show up in the dashboard. I think many other developers on this thread are doing the same thing to test their integration. So, I don’t think your comment could explain the issue.
"If Corona is the only engine that did not prepare for this transition that was announced more than a year ago by Apple there is nothing RevMob can do now. This is the result of Corona limitations and not RevMob."
I completely agree that it was very unfortunate that Corona didn’t implement IDFA support until just a few days before Apple’s drop-dead deadline.
However, now that IDFA is accessible in Corona, and the RevMob SDK versions 5.2 and 5.2.1 use it, why are installs still not being reported, and the only revenue so many developers are seeing on this thread is $0.02?
You said that “there is nothing RevMob can do now,” which is a very strange comment. In fact, RevMob already did exactly what it needed to do – it released a new version of the RevMob SDK for Corona that accessed IDFA through the new API Corona exposed. The problem is that this new version of the SDK, with the most recent builds of Corona, still doesn’t seem to be working. and we haven’t received any convincing technical explanation of why.
It’s actually not clear to me if this is an issue specific to Corona, or if other engines are experiencing the same issue.
EDIT: I do know that the RevMob SDK uses not just the IDFA, but also the “iosAdvertisingTrackingEnabled” flag, which until Corona build 1106 was apparently not working correctly. I suppose it’s possible that this explains the issues everyone is experiencing, but I’ll have to test it to find out (right now, in fact).
EDIT #2: Actually, it seems the RevMob SDK uses “iosAdvertisingTrackingEnabled” only for diagnostic printing, so I don’t think the fact that the flag wasn’t working correctly until build 1106 will explain the issue we’re discussing.
Thanks!