Video Ads

Hi guys. I have heard that video ads pay very well, but I never really considered using them until today.

The game that I am working on is strongly story based (cutscene, level, repeat), and is several hours long to play through. If I used video ads at a low frequency, maybe players would be willing to watch.

I’d love to get your feedback. Do you think players would revolt? Do you think that I would earn more revenue with simple banner ads? Has anyone already tried this? If so, any recommendations for video ad networks?

Thanks!

Where is the video AD?

The idea would be to let the player play 20-30 minutes worth of the game ad-free, then, before they start a new level, show a full screen video ad before they can continue playing. Repeat.

I tried MobFox but it couldn’t find a video ad in the USA. Maybe I was just doing it wrong. Their video ads SDK is in beta, though.

If anyone has had success loading MobFox videos, please let me know.

Adcolony has great ecpm’s but terrible fill rates !!! I think in the long-term this will be a great option for developers but there doesn’t seem to be enough advertisers yet so it’s a shame !

I’m afraid that is against the new Google Policy…

Video advertising is one element of today’s advertising scene that is still powerfully under-appreciated and under-utilized. As time passes, we’re going to see a huge investment of capital there. Unlike the not-too-distant past, when there weren’t many tools and resources available to advertisers and developers in the video advertising market, that reality continues to change. Just this past week, as a perfect example, Airpush released its SDK 5.0, and guess what came with it? Something called “Smartwall ads,” which optimize between different media including video. It’s all about maximizing both potential AND earnings!

Hi Androider. Thanks for letting me know!
Can you tell me what rule it violates?

I don’t consider this a violation of rule 3:
Forcing the user to click on ads or submit personal information for advertising purposes in order to fully use an app provides a poor user experience and is prohibited. Users must be able to dismiss the ad without penalty.

The player is not required to click on the ad, and the ad “dismisses” itself without penalty to the user. If we interpreted the second sentence as strictly independent of the first, it would imply that we would be required to put Close buttons on banner ads as well. So I don’t think that is the intended reading.

That said, I understand how there might be room for interpretation here. Is there an authority I can check with? Is there an Android Developers Forum that Google staff monitors and responds to?

If it does break that rule then YouTube also violates it.

I would be VERY careful about assuming that just because Youtube can do something… so can you.

That is exactly the rule I’m refering to. I have not seen video ads, but banners are not full screen and so there is nothing to dismiss. Anyway, I’m no expert in ads, even the guys who base their business on ads don’t seem to have a clue on what is ok and what is prohibited…

Come on guys, what’s not clear?

It is prohibited to have ads that force users to click on them or submit some personal info in order to continue using the app. With most banner ads you can continue without clicking them so it is fine. If someone would implement it in a way that to continue using you’d have to click on the banner that would be against the rules too.

Among other ambiguous parts of the new policy this one’s crystal clear.

@max: I think the only question wrt the new policy is: do I have to provide a mechanism for the user to be able to dismiss the video ad without watching it all the way through? I expect the answer is no, but given a particular interpretation of the second sentence one might read it otherwise.

I’d say the answer is yes. Watching the video ad all the way through is as much consuming the ad as clicking a banner.

awesome review on video ops on another thread by Gordon here -
http://makingmoneywithandroid.com/forum/showthread.php?tid=660&highlight=airpush

have been using AppBrain’s appwall for a while and have recently switched to Airpush SmartWall product, and here is my comparison of the products. I have also tried Leadbolt and Tapjoy’s appwall products extensively but they are not at all competitive with Airpush and AppBrain so i have left them out:

  1. Airpush SmartWall CPM = $7 - $8
    AppBrain AppWall CPM = ~$3

(I think this is because AppBrain’s product only includes AppWall, whereas Airpush’s includes other full-page formats combined into 1)

  1. Airpush Fill Rate = 89%
    AppBrain Fill Rate = 95% (sometimes higher)

  2. Airpush Support = so-so
    AppBrain Support = AWESOME

  3. Airpush Payments = weekly with no fees (WOW)
    AppBrain Payments = shortly after month end (Still pretty good compared to other networks, but obviously airpush payments are the best)


Overall, the Airpush product is greatly superior so far HOWEVER I notice an interesting trend where during the Asian timezone the AppBrain wall performs a bit better. To try and capitalize on this, I am currently trying to write a home-made mediation SDK which will switch between Airpush and AppBrain SDK’s in real-time depending on rules I set – I think that will increase the yields tremendously. PM me if you’d like to try it when I’m done !

Would love to hear the experiences others are having.

I am about to release my app - but as a developer I find impression-based to be more sensible for developers (since you are committing to space for the ad - the ad agency should commit to something as well).

Why should developers have to bear the brunt of low click-through rate caused by bad ad graphic, bad ad message etc. and so on.

It intuitively seems that video ads fit the “impression-based revenue” model.

Although here also I suppose the revenues could be different - according to:

  • whether user skipped the ad after seeing initial second or so (assuming was a cached ad pre-downloaded - as it probably SHOULD be so there is no wait - this also means pre-roll video ads are problematic unless have a very fast connection)

  • whether user interrupted flow of video ad in middle so only saw PART of it

  • user saw whole video ad

I would be interested in hearing of video ad sources as well - and as folks say here this could be the future of ads (esp. if you can pre-load/cache the video ad could be workable on slower connections also - though folks’ data plans may run awry of such internet use ??).

As I understand it - a pre-roll video ad MAY be acceptable to users - at least that is what many of the ad networks say about their interstitial full screen ads.

However pre-roll does not give much time for buffering/caching - so user may be impatient - especially if video is not playing smoothly at start (if it does they may even consider watching it - assuming it is a new video ad).

Video ads could be shown in the middle of a game during a natural pause.

I have been experimenting with showing interstitial ads using a sort of timed mechanism - so the ad CANNOT appear in areas where you don’t want to bother the user.

And is only triggered by some events - and even then is cancellable.

The end result is that the ad is usually shown (I am talking about in-game now not pre-roll) at a time when the user pauses for a little while.

And at least in my testing I have seen this to be a relatively non-intrusive way to show such ads.

After I publish my app and see the results (though I will not have anything else to compare with) I can better estimate the value of such an approach.

AdColony pays by impression :wink: as I have written though, they give great CPM’s but terrible fill rates.

Thanks.

What is a terrible fill rate ? Never or like 10% ?

What are their CPMs - are they high enough that CPM x fill rate - giving an estimate of total earnings possible - is still competitive ?

Even so (with a low fill rate) - there could be value if integrated as first option - with fallback to some of the other options.

I wonder what is limiting the growth of video ads - it makes sense that this is where things will go and because of it’s simpler business model (plus continuity with TV advertising).

The limiting factors maybe that users may get angry at high data costs.

Secondly maybe an impression at the ad networks that sufficient bandwidth is not available - or there is no support for video ad yet in games (i.e. chicken and egg problem) - or the higher cost to deliver video (essentially infrastructure like youtube required at the ad agency ?).

However, I suspect small videos - like 30 seconds may be totally doable - even on a slow connection if there is capability to cache the ad during game play etc. - and then show it when appropriate (and video will run seamlessly without wait - and may even be watched by the user).

adcolony’s fill rate was ~20% when I tried it, with cpm’s around $15. and their SDK is beautiful, it does caching and other cool stuff… $15 cpm x 20% fill rate = $3 ecpm which is around what I saw with AppBrain but much better than Tapjoy or LeadBolt. as I posted recently, with Airpush SmartWall i am getting $13+ eCPM’s so that is my current interstitial solution and they are adding Video Ads to smartwall within a few weeks supposedly. but I have only been using it for around 1 week so I can’t vouch for it fully yet.

adcolony is AWESOME cpm, but weak fill rate. IMO that will just get better over time, so I think AdColony is a great company and potential contender for the long term but not quite ready. the problem with video, there simply isn’t many mobile video ad campaigns out there yet since the format is so new.

re: buffering / bandwidth
I was thinking of only serving video ads if they user has an active wireless connection. If that was not available, I would deliver the player the normal banner ads.

One concern with this setup is that some players might not understand what was going on, and think “What a greedy app! It gives me video ads AND banner ads!”

So, possibly, before rolling a video ad I could present a screen as follows:

This <chapter> of the game is presented ad-free by:

  • Continue to Video (countdown for N seconds) -
  • No thanks, I like banner ads -
  • Upgrade to ad-free version -

which offers the user information, a choice, and an upsell.
[hr]

I feel like there is a factor missing from this math, which is impressions/unit time. You might present a banner ad (impression) every 1-3 minutes, but if you were presenting a video ad only every 10-30 minutes, your video ad revenue would need to be divided by 10 to do a fair comparison.

Is that right?