We announced our mobile analytics product last Wednesday, and we have been greatly encouraged by the response. It seems everyone who works in the mobile space is in need of a proper analytics solution. Site owners need to understand what is happening and “who” is doing it. Advertisers need to get actionable data on which campaigns are working and which aren’t and determine how to use their marketing dollars most effectively.

After the announcement we had over 500 invite requests and have had very positive interest from many people in the mobile space. We’re greatly looking forward to sharing all the engineering work we’ve put into device, carrier, and user detection with the broader community.

From a timing perspective, we have some partners we identified even before the announcement working on helping us iron out the kinks. We’ll be sending out invites to a small handful of additional invitees each week. We hope to allow a larger batch in during early to mid June and extend registration to accommodate anyone who has requested access to the beta a few weeks after that.

We also saw some interesting responses in the market from some alternative analytics providers. From what I can tell, it seems that they are taking issue mainly with:

A. Our intentions (considering we operate an ad network)
B. The accuracy of our data
C. Our ability to compete in this space

These issues are probably worth addressing up front.

In terms of our intentions, they are fairly straightforward. We want the mobile web to grow, and we want it grow as fast as possible. The better (and bigger) the mobile web gets, the better AdMob does as a business. We feel that in order to catalyze growth, the mobile web needs a solution that has three components:

1. Completely (unconditionally) Free
2. Easy to Use
3. Accurate & actionable

With this in mind our intentions are simple; we couldn’t find a solution in the market that met these three criteria, so we decided to build it. It’s not terribly surprising that this is the case, because in order to provide a truly free product, a company would have to have some alternative business model to subsidize the investment. While we have invested heavily in our analytics product, we don’t need to base our business on it. This enables us to give it away with no strings attached. As we have stated before, we won’t require that someone be a publisher, an advertiser, or have any business relationship with AdMob. We won’t secretly be altering data to make ourselves look better, and we won’t be trying to somehow sneak ads onto people’s sites. We expect our core business will benefit, but only indirectly. What we really want is for the mobile web to improve, and this product is needed to make it happen.

With regards to data accuracy, it’s rather premature for anyone to conjecture on how well we will or won’t do. We’ve spent two years building highly scalable systems for mobile and I can assure you we have not devoted the attention of over 30 engineers to building sub-optimal technologies. Without going into detail, it’s safe to say we’ve gotten very good at device, carrier, and user detection. That being said, the product will be in beta for some time, and that means there will certainly be bugs, problems, and inaccuracies. No analytics product, no matter how mature, can claim 100% perfection when it comes to data accuracy. Mobile analytics is a very deep and fairly complex problem; it’s reasonable to assume our analytics product (especially early on) will have its share of imperfections.

This brings us to the last point, which revolves around our ability to compete in the space. The truth is, because we have not built a business model around analytics, we don’t really consider ourselves to be competing. If we had found something that met our criteria it is unlikely that we would have spent the time and money to build our solution in the first place. Now that we have, both analytics and our core network share the same detection infrastructure, so continued investment will be a given. We believe people will probably try multiple products and make their selection after testing them out. That would be great. For us, the important thing is that everyone has access to the data in order to make better decisions. If this happens, whether through AdMob analytics or an alternative provider, we will be very happy.

Many of the alternatives have interesting capabilities and they are certainly worth a look. The ones we are aware of, in no particular order, are:

Amethon (http://www.amethon.com)
Mobilytics (http://www.mobilytics.net)
Bango (http://www.bango.com/analytics)

There may be more and if we missed you it is certainly not intentional.

The good news is that if you are an advertiser or a publisher or you just have a mobile web site and simply need access to analytics data, help is on the way. There are plenty of smart people trying to solve this problem, and that can only mean good things are in store for everyone who participates in the development of the mobile web.

- Omar