For those of you who are sceptical about analyst forecasts eMarketer published a table which shows how different firms see the mobile advertising market. The eMarketer figures are roughly 2x the lowest estimates.

The new information for me was which companies have the leading market share and how the market breaks down between different advertising formats.

Search accounted for 45% of the market in 2011 ($653m) and the share of search is expected to rise to around 50% in the coming years. Google dominates search with around 95% market share. I’m not sure I’ve ever done a mobile search with a search engine other than Google.

Display was 31% of the market in 2011 ($445m) and the share of display is expected to rise to 37% over the next five years. Google also has the largest market share in display, but at 25% their position is not unassailable. Millennnial Media (who recently filed for IPO) and Apple’s iAd are equal second, each with around 18% market share.

If you’ve been reading this blog over the last week or so you will have seen that DFJ Esprit recently invested in StrikeAd, which plays in the display segment of the mobile advertising market. The interesting sub-trend within that market is the shift towards exchange traded media. As far as I’m aware there are no analysts forecasts for how mobile display splits between exchanges and ad networks but from our work we estimate that around 10% of mobile impressions are currently exchange traded, but we expect that to rise to around 50% of impressions over the next year or two.

The following bullets sum up the most pertinent recent news on Android:

• As an OS Android continues to do very well, with market share now over 50%
• Samsung is now the dominant Android vendor – Q4 sales were exceptionally strong
• HTC and Motorola, the other leading Android vendors are having a tough time
• The leading Android tablet, the Amazon Kindle Fire, runs a forked version of Android

Looking at this, it is easy to see a future for Android where the lions share of sales are with two vendors, one of whom (Amazon) has already forked Android and has a ‘closed’ mentality, and one (Samsung) that has a history of proprietary OS development (remember Bada). For a while Jof has been saying that he can see Samsung making its own fork of Android, and if that happens the smartphone market would be dominated by two companies with roughly equivalent proprietary OS-hardware combinations – Apple and Samsung.  The tablet market could end up in a similar place with Apple and Amazon being the lead players, although Samsung may have something to say about that.

There are of course other credible future scenarios for Android, including a resurgence of Motorola under Google’s ownership, but I think it is becoming clear that even if Android does win out against iOS it doesn’t necessarily follow that we will have an open, startup friendly, mobile ecosystem.  It may just be that we have a new set of gatekeepers.

If I’m right then developers will start developing for Android at the same time or even before they build for the iPhone – something I would like to see.  I have recently switched from iPhone to Android and my experience so far is that most startups launch first on iPhone with many not even having a date for an Android release when they first go live in the App Store.

The charts below were on Techcrunch over the weekend.

The first implication is that as tablets and mobiles continue to improve and take time and attention away from desktops and laptops video views will continue to soar.  This is good news for over-the-top TV plays and the mobile advertising ecosystem.

The second implication is that other apps and services will likely see the similar boosts in mobile usage.  Services that are ‘built for mobile from the ground up’ have been in vogue now (see Jim Breyer’s comments in this series of interviews on Techcrunch), and this tells us why.

Other interesting data in the Techcrunch post is that video plays on connected TVs tripled in Q3 and that the Android-iOS split of mobile video plays is approximately equal.

URLs and search are simply a better model for navigation, particularly as voice input is finally starting to get to ‘good enough’.

However, at this point the functionality available to apps exceeds that available via mobile browsers and so the trend is definitely still favour of apps over browsers.  In fact, I’m writing this post today after reading news that Walmart is releasing a new iPhone app.

That is actually a hassle for Walmart as well who have to worry about maintaining multiple versions of their iPhone app, and also to maintain apps on the other mobile platforms.  Once the mobile browsers are good enough it is much better for them to simply update their website.  Then they can roll out changes as quickly as they like without worrying about app store approvals and annoying their customers with too many upgrades.

My conclusion, therefore, is that both consumers and enterprises will gravitate back to the browser, and I think we will start to see that soon.

A bit of context.  There has been a lot written about app vs browser wars and as with many debates the discussion got polarised and everyone got bored before agreeing to say that there will be both apps and browsers.  I agree with that but think that the trend within that is interesting, i.e. having an opinion on whether the balance shifting in favour of apps or browsers, which is why I was careful in the title to this post to talk about the ‘balance’ between apps and browsers.

I have also limited the discussion to commerce.  I did that because the navigation point is more easily understood, and because commerce apps are relatively simple and don’t need to stretch the capabilities of the phone in ways that favour apps over browsers.  However, I think the balance will shift back to browsers does apply more generally.  Over the long run I think that even complex bleeding edge areas like games will see the balance shift back to the browser, just as we are seeing on the web.

For me the best proxy for smartphone market share is mobile ad impressions.  This metric captures internet usage of the devices which makes it the best indicator of where the mobile computing market is headed.  By focusing on internet usage it filters out low end Android handsets that are little more than feature phones and people who have smartphones because they are cool to own, but don’t use them as mobile compute devices.  My thesis is that this fashion brigade will quickly shift to the next new shiny phone when it emerges.  The trouble with looking at ad impressions is that it ignores some app activity, but all round I see it as the best metric to use.

And on an ad impression basis focused on smartphones only Android is running at a 54% market share, with iOS a distant second at 26% (courtesy of mobilemix and Digital Trends):

Using the more traditional metric of OS market share by devices sold Android is again out in front.  Gartner has them at 43% (table below) with the fast declining Symbian still holding second place ahead of iOS.  Canalys paints a similar picture with Android at a 48% market share and iOS taking the second place slot from Symbian last quarter at 19%.  All parties agree that the smartphone market continues to grow very fast and that within it Android is growing the fastest (i.e. gaining share).

The best Android phones still don’t match the iPhone (and as I said on Friday my next phone will again be an iPhone), but with the scale advantages Android now enjoys I think it will only be a matter of time before their phones are the best.

On the tablet front Android only has a 20% market share and is still some way behind Apple who has just about all of the remaining 80% (as per ABI Research), but this market is much newer and it was only 6-9 months ago that the first mass market Android tablets hit the stores.

Hat tip to Strategy Eye for the graphic and original article.