Friday, 1 February 2008

Further Observations on Mobile Search

Justin did a good post last time on Mobile Search and this week I found myself commenting on a post from Peggy-Ann Salz here, so I thought I'd pen some thoughts on our findings regarding mobile search and discovery.

The first thing to say is that these terms mean different things to different people depending on their objectives and that for observations on the relative merits of each system, one really does need to isolate from which perspective the observation is made.

Mippin itself interacts with these search and discovery engines in different ways:

(a) as a receiver of natural search traffic. Our universe is constantly being indexed by bots from search engine providers  and as a result our chances of being found by someone in response to a search on their phone is growing. In this regard much has been made of the immature way that Google's algorithm over-emphasizes the importance of web presence in its calculations - and this certainly is to our disadvantage - but the fact of the matter is we receive 30 times more traffic in this way from Google than the next best search engine (which is just Yahoo! but it changes by the week). If Yahoo!'s clickthrough rate is higher it really is immaterial at this point because for us it is the traffic that matters and we do not pay for this of course.

(b) the second element which is important to us is in paid for search terms in which we are returned as a link in response to a user query. We define a list and bid an amount for these terms. Justin may be able to shed more light on this in due course but again we find it very easy to drive momentum through Google and almost impossible to drive it through Yahoo! in any market. Again, clickthrough rate is irrelevant really (it is relevant for conversion as higher context means more effective acquisition but that is for another day) - as we pay only for what we see. In this category, AdMob also plays and is an effective distribution channel which is improving all of the time. Outside of the US, it is our empirical finding that Google is already driving more traffic here, though AdMob is a good and complementary second play. Yahoo! is not a player yet - or at least - Yahoo! has not been able to educate us sufficiently to be able to drive any meaningful traffic.

Much can be said on how Google and AdMob could improve visibility and reporting and I am happy to respond to any questions on this.

(c) A third way in which we interact with these engines is on the revenue side. We are integrated with both Google and AdMob to be able to ingest adverts from them and serve them to our users. On the positive side, it works and AdMob is our preferred partner currently because Google tends to serve up a lot of its own inventory to its other mobile services which gets a bit tedious for the user. On the downside, there is very little targeting capability in either engine but here I think we are at the forefront of what can be done and will always be in a position of trying to cajole the ad servers to be better.

Observations that AdMob's ad inventory mainly relates to mobile content suppliers is true - though a number of premium brands are beginning to use it - but to be a rounded discussion one should also point out that Google is certainly no better at this point.

Labels: , ,


Blogger Pedro Sousa said...

Very interesting post! I loved to read it, keep it up! ;)

Pedro Sousa

7 March 2011 at 17:51  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home