Performance marketers have been multiplying their arsenal over the past few years and mobile advertising has really come into its own over the past 2 years or so. There are many ad network for mobile ads, but without a doubt the largest one is owned and operated by Google – combining search ads, display ads, and ads for mobile apps served through their AdMob product.
For many of our clients, GDN mobile ads alone account for 20 to 30 percent of their conversions. But what about AdMob?
A few months ago, I suggested that AdMob was not a good fit for performance marketing. At the time, the network wasn’t integrated into AdWords and the level of control was ridiculously limited.
For example, targeting options were basically country, carrier, and device. From a performance marketing perspective, putting an ad in front of everyone who owns an iPhone in the United States limits the possibilities for success quite a bit.
However as Google integrated the offering into AdWords this past June, more flexibility and many more targeting options became available that would suggest that it’s at least conceivable performance marketers can find success within AdMob, even if they’re not promoting app downloads.
AdMob now even supports ad extensions for location and for call, enabling support for marketers who need to promote local business. Because geotargeting for mobile means in some cases the device location is the source for the geolocation, it is now even possible to reach some mobile users while they are near your store and using an app.
Apps and Categories
While there were certain adjustments to geotargeting, which made it easier to be much more accurate in ad delivery than just targeting countries, carriers, and devices/platforms, the most interesting new feature is the ability to target specific apps and the categories and subcategories they belong to.
It’s now possible to run ads across dozens of categories and sub-categories of apps, and to monitor which specific apps are delivering results. You can also target apps by searching for them using keywords and adding them one at a time, there is even an option to enter them in bulk. Marketers can then decide how much to bid for clicks or conversions for the entire ad group or for each app.
The difficulty for the moment is the lack of an API for technology vendors and tech savvy marketers to develop applications that will mine the opportunities for success automatically. With the ever growing number of available apps, which currently sits at about 300,000, figuring out which apps are working can quickly become a nightmare.
Preliminary Tests and Results
We have been running a test for a few weeks in collaboration with a client. The goal of this test was to try to generate leads at or below a specific CPA using algorithms to make decisions on which apps to target and how much to pay for each click based on the observed volume of conversions and CPA. The results are promising though not conclusive yet, and the client has been able to generate a significant number of conversions (free signups for a publisher).
While we are now fairly close to the target CPA, though we have not yet quite reached it, it is clear that there is a lot of volume available on AdMob. We are seeing large volumes of cheap clicks, which makes it likely that with continuous optimization efforts success is a likely outcome.
I’m curious how many of you have been testing AdMob and who has seen the kinds of results we have? If you’re one of those who has tested it, have you been successfully generating leads at acceptable CPAs with AdMob? Leave a comment below!