Greg Sterling notes that Microsoft’s $150MM advertising campaign has increased their share of the search market share by 1.4% (from 12.8% to 14.2%), and states (more than asks):
“Is that any kind of ROI?”
A somewhat difficult question to answer, outside of the right context…
If Microsoft could continue to buy market share at that rate, it would acquire 100% of the search market for a lowly cost of $10.7B.
But of course, things aren’t that simple. A one month change in habit (as is the case here) doesn’t mean long term success. And more importantly, that first percentage point is going to be (I’d wager) an order of magnitude (or two) cheaper than the last percentage point, unless you have a clearly superior offering. (The fact that Microsoft doesn’t have a superior offering may well be evidenced by the fact that they didn’t take share from Yahoo! or Google.)
Clearly a bad buy then, right? Not so fast.
Remember, also, that Microsoft is building its own paid listing bidded marketplace (which will lead to further first-tier fragmentation). At some point (soon), they are going to need to seed that marketplace with active advertisers and drive individual marketplaces to a liquid state (defined as all served bidded listings being separated by a $0.01 increment).
No problem, right? Heck, I’ve often said that any rational marketer would buy any ROI positive flow that meets her hurdle rate and which she can service, advertising “budget” be damned.
Unfortunately, not everyone “gets” this yet. There are a lot of advertisers who advertise with either Google or Overture, citing level of effort, ease of use, etc as reasons not to use both.
So, if you’re a (potential) marketplace owner, you’re very concerned about offering a critical mass of search volume to your search advertisers, lest they not make the effort to bid in your marketplace. In that case, $150MM for a 1.4% gain (or, said another way, a 10% increase in volume) looks cheap.
Hmm, now that I’ve come this far, let’s dig a bit deeper together, eh?
SiliconBeat broke the story last night that Google is buying Urchin, a marketing analytics company.
On the surface, the play here is obvious: if you hold ROI constant and increase conversion rates, CPC prices will rise (proportionally). Said another way, as long as all of the ROI doesn’t go into the advertiser’s pocket — and it won’t, because their competition will eventually wise up and use the same solution, thereby increasing competition — Google will profit.
Said yet another way… what Google giveth, Google taketh.
But again, that’s easy and obvious… so let’s return to that notion of fragmentation and the problem of managing listings and bids across marketplaces… then add an ever increasing variety of advertising options, channels (platforms), formats, etc.
If you can imagine such a scenario (not hard, because it’s happening right now)… wouldn’t it make sense to want to control the interface through which all of that data is reported… and ultimately, the interface through which many of the decisions on that data are acted upon?
Update: As if all of the above isn’t reason enough, this forum thread says that Urchin has one issued patent and one assigned: System and method for monitoring and analyzing internet traffic (earlier version here) and System and method for tracking unique visitors to a website. Interesting. Can’t imagine there aren’t prior art claims for this stuff.