Appbrain, which we here at AP use to this day thanks to a few handful features that the Play Store still hasn't implemented, analyzed 140,000 Android apps and came up with a list of the top 10 ad networks.

While they don't openly state the source of this data, I am willing to bet that it comes from analytics reported by their Ad Detector app which hit the Play Store a few months back. The app itself, much like Lookout Ad Network Detector, is very handy - it lets users figure out what ad networks, social SDKs, and even developer tools are used by apps installed on their devices.

It only makes sense that the information collected was reported back to Appbrain (hopefully after appropriately notifying users of such actions), aggregated, analyzed, and presented as a handy infographic. The sample size is 140,000 apps, which I'd say is representative of the currently active portion of the Play Store and therefore should paint an accurate picture.

Update: Appbrain confirmed my theory - the data was indeed collected by Ad Detector. They added:

Yes, the ad detector allows people to add anonymously to our ad-network database (there is also an opt-out for that).

We have also thought about whether the quality of our data is good enough and whether there may be significant biases, but our opinion is that the data is not subject to structural bias and our conclusions are valid, because of the following points:

- The ad detector was pretty much only promoted by us, it's not getting very many downloads if we're not actively promoting it on our site, so it's not 100% people that searched for "remove airpush" or something similar.

- Even if all the people that install the ad detector have a single airpush / aggressive ad network on their phone and install the ad detector, we also potentially get information about on average ~50+ other apps they have installed. We only keep apps we receive once in our database, so we very quickly collected information about the most popular apps, and even now, most of the time we have 100% of a users' installed apps in our database already. Therefore our database has very high coverage of all commonly downloaded apps.

I agree that whether the exact numbers of for instance airpush are not slightly overestimated is very hard to say. Unfortunately, it's very strongly on the rise, so even if they're actually a bit lower now than in our current data, it will soon be that way (we also looked at apps that were released in the last 30 days, and in those apps the airpush market share is 16%.)

Before we proceed, it's worth pointing out the absence of Google's AdSense on the list. It's not an omission - as Google explained last year, AdMob is used for mobile apps, while AdSense is used for mobile sites.

Without further ado, here are the top 10 ad networks - I bet you didn't think AirPush was at... #2:

  1. Admob 39.2%
  2. AirPush 6.5%
  3. Millennial Media 5.1%
  4. LeadBolt 4.1%
  5. AdWhirl 3.9%
  6. MobClix 3.5%
  7. InMobi 2.6%
  8. MobFox 1.8%
  9. Tapjoy 1.5%
  10. Startapp 1.2%

At least SellARing is nowhere to be seen... for now.

Thankfully, publishers of the top 500 Android apps are a little more sensible about aggressive, as Appbrain calls it, form of advertising (the word "intrusive" comes to mind too), and AirPush is way down at #9 on that list, with LeadBolt not even in the top 10.

Overall, however, a whopping 12% of apps use some form of aggressive advertising. Thankfully, I have none of those installed, and if I ever do by accident, Ad Network Detector will slap them right out of me.



Artem Russakovskii
Artem is a die-hard Android fan, passionate tech blogger, obsessive-compulsive editor, bug hunting programmer, and the founder of Android Police.
Most of the time, you will find Artem either hacking away at code or thinking of the next 15 blog posts.

  • Simon Boulet

    From the article: Thankfully, publishers of the top 500 Android apps are a little more sensible about aggressive advertising.
    Of course. They are in the top because they are more sensible. App that use aggressive advertising will never be popular for a big portion of user.

  • Wyngo Masala

    Do apps use more than one ad network? If not, the bottom graph makes no sense, because it totals far more than 100%.

    • http://codytoombs.wordpress.com/ Cody Toombs

      There's two common reasons for that:

      A) Like a lot of websites, many apps use a failover mechanism.  If one ad network doesn't have ads to push, then the app calls to a secondary provider.  It's a common practice when you favor an ad network for it's relatively high payout but it sometimes suffers from lack of content, just push a lower paying ad and keep going.

      B) Some ad networks use different methods for advertising.  Some are simply displaying 2 or more different kinds of ads (ex. A perpetual banner ad in the app and also a referral ad for products).  There's also apps that include multiple ad networks because each one has different aggressive methods like pushing notification tray advertisements, some are screwing with your web browsing or randomly performing pop-up ads.  Figure, if you're slimy enough to put an aggressive ad network into your app, you're slimy enough to put in 3-4 different ones.

    • http://www.androidpolice.com/ Artem Russakovskii

      They do.

  • Jjsan

    Thats why i have ad free ;)

  • John Barnard

    Maybe starting a campaign to boycott any app that uses the more aggressive forms of advertising would send a message to the developers who bundle this crapware with their programs.

    • http://www.facebook.com/profile.php?id=674791489 Loic Lacomme

      it is happening to ionroad at the moment. they keep changing the way ads work in the app after going from no ads to some to full screen ad back and forth.

      • http://twitter.com/homncruse Aaron Burke

        I just uninstalled iOnRoad yesterday after a few days of being annoyed with their notification bar ads.

        • Dannyats

          IOnRoad has removed the notification ads completely after they listened to the users, I've just reinstalled.

  • esper256

    People who are going to install an ad detector on their phone are probably doing it in response to getting aggressive ad notifications. So this data would be biased towards those type of ad networks. People who don't have any apps with aggressive ads aren't as likely to install the app and scan their phones.

    Although it could be that the stats are against apps available in the market and not instances of app installations. In which case this bias wouldn't apply. But my guess is that it is on app installation instances given that it says a "sample" of apps.

    • http://codytoombs.wordpress.com/ Cody Toombs

      Great point, it's almost certainly not going to count many people like us because we're smart enough to avoid low rent apps (sorta reminds me of the days of email viruses, it wasn't people like us opening attachments).  On the other hand, Most of the people who are installing the crapware are probably also installing most of the good apps that we use too, so the margin of error probably isn't that bad.

      • http://www.androidpolice.com/ Artem Russakovskii

        I'd say a good number of people installing ad network detectors are also doing so for prophylactic measures or just to see what's going on with apps on their devices. I know that's what I did.

        But, of course, you have a point there, esper.

        • http://codytoombs.wordpress.com/ Cody Toombs

          Oh, I wasn't even talking about the detectors, though I was assuming the majority of people installing those have either been hit by crapware or been made paranoid by companies like Norton.  I was just talking about the actual statistics, people who installed are installing the same 60-90 apps I've got installed, so the coverage seems pretty accurate.

  • crufia

    Push ad behavior in Airpush is optional, so I suspect a lot of people might have apps using it without seeing any push ads.

  • Benjamen Meiers

    The reason AirPush is 2nd is because people install the app to find what app is using AirPush so that they can delete it. Otherwise known as selection bias.

  • Not enough minerals

    i, for one, immediately delete apps that leave a running service it does not need to work. if the apps good, i'll try to disable permissions and such, even so i usually end up removing it entirely eventually. 
    i don't like useless services running on my desktop pc with tens of gigs of ram, hundreds of times more processing power and all, theres no reason i would leave them running on my poor little phone/tablet.

  • Chris

    Looks like @airpush promoted it as well :P 

  • Vitalik Derevinskiy

    Top player in the mobile market advertised Android and IOS, is


    For referrals as a gift of $ 100 to the account after the launch of at least one application with their SDK!

    Analogue methods for advertising and Leadbolta AirPusha, BUT! Dr.Web pleasant surprise on the built ad does not swear! Users install the application quietly)))

    Yield is higher than other affiliate

  • Sofía Santucho

    My experience in using ad networks has been really bad until I discovered adtomatik. The best ecpm and the highest fill rates