Exploring the world of Google Metrics
You may remember an early post Let’s call it the Google in which we explored the Morgan Stanley’s Mobile Internet Report that revealed the World Wide Online Advertising ARPU was estimated in 2009 at $46.41 and that $20.06 of that belongs to Google.
SAI produced a chart in March this year based on similar figures that revealed the Average Revenues per Unique for each of the major web sites. In that chartGoogle achieved $18.44 per unique, AOL (thanks to its subscription business) gets $12, Yahoo gets $6, Microsoft $4.42, Facebook $3.09, and Twitter $0.62. Like most commentators SAI used these figures to proclaim Search Advertising is more lucrative than Display advertising.
The error in SAI analysis of course is the simple fact Google’s revenues are not generated by search advertising (i.e. putting Ads on Google’s search pages) but by putting “Google Search Ads” on web sites scattered across the web. The success of Google’s business model is to reverse engineer the search engine logic to create the world’s most powerful aggregated advertising network. The deal that Google offers the world isn’t “optimize your site for Google and we’ll send you lots of traffic” but “if you have lots of traffic we can help you to turn those eye-balls into pennies”.
In previous posts I have examined the stats I have generated here at excapite that prove a Top 10 Ranking on Google is of negligible value to the publisher. Today I was going to continue on with my Google theme by exploring the mathematics that explained why a Top 10 ranking on Google for a popular topic covered by over 100 Million + search listings in Google would deliver only average 1 or 2 hits per day to the publisher. Or put another way: Why is the atmosphere of the web so thin?
I’d already canvassed some of those issues in a previous post when I suggested that the problem of building a search engine can best be explained by thinking of a number between 0 and 3 Trillion and then asking a friend to guess the right answer. We discovered in that post it takes 39-40 guess to find the answer. So what does this guessing game tell us about the challenge of trying to write a post that attracts the largest possible volume of search engine traffic to your site?
Thankfully Bradd Libby provides us with an insight into how we may discover the clues to finding the right answer in his recent post on The Lies of Efficient Frontier. In this post Bradd provides us with an exploration of the economics of the AdWords Auction process in which he explains that there are 100,000,000 (one hundred million) possible ‘combinations of bids’ for an account with 2 keywords. In Bradd’s example he was explaining the permutations on the $ bid value however the insight he provides is simply this. If, as in the example from the previous post, it takes an average of 39-40 guess to discover the right number, how much more complex is the challenge if we have to guess 2, 3 or more numbers using the same technique?
This then is the Google Search Ranking challenge. Not only do you have to achieve a high page ranking but you have to achieve that high page ranking consistently across all the possible permutations of those key words. basically if your key word is iPad App you need to be on the front page of every possible permutation of a search string including the 2 words “iPad” and “Apps”. Best of luck…
I’ve become a big fan of Bradd’s work. His excellent articles on the market dynamics of the Paid Search Auction, the ins and out’s of click (PPC) advertising campaigns and CTR Metrics (See Ad auctions are not auctions, Lessons From Google’s Own Paid Search Campaigns: Forget The “Rules” and Statistical Significance: Not Just For Geeks Anymore) are highly recommended reading for anyone interested in discovering more about the subtle art of measuring the impact of Google.
I also highly recommend Matt Shanahan’s blog on the metrics of survival for online publishers. Matt has amassed a worthy body of work examining individual case studies in the application of Average Revenue per User (ARPU) as a measure of profitability and viability across a number of leading web brands. Exploring how the scale of revenue is a more accurate predictor of success than the amount of traffic. It is of course a simple concept and one that I have explored before in posts like More proof that traffic doesn’t translate into customers, Search me or Trust me?, Pay Wall Economics 101, The Business of Blogging, Why newspapers need to get sticky, Why the NY Times 10% Problem is not a problem but a significant growth opportunity and Here’s why Analog Dollars equal Digital Pennies.
The problem I have with Google of course is simply that the old mass media advertising model was Browse With Us, Buy From Them. The new model online has become Browse With Us, Buy From Us. In this context Google’s whole business model is “so yesterday”. Fortunately for Google and the rest of the web the urban mythology surrounding Search Engine Marketing has prevented the market from unmasking the great pretender – so far.
