Surviving Google Slaps
Online search is getting increasingly crowded, making attention getting differentiation even more difficult, as well as expensive. Here are a few things we’ve found that help a little. Try this at home if you like, but note that results may vary.
The sheer number of people doing online searches attracts ever more adword campaigns. In many ways, this escalation is akin to arms races, driven by the often unstated assumption about the huge number of Chinese who need FillInProductNameHere that occasionally has been in vogue over the last 150 years or so.
Statistically, the large number of adword campaigns has to occasionally result in some being successful, not on their merits or crafted designs, but on probability distributions at the edges of Dirac’s Large Number Hypothesis. This did occasionally happen in search’s early days, giving rise to an army of online pitchmen with claims of getting rich quickly using their methods. Most of these are gone now, and hitting search jackpots only results in the rules being changed, and very quickly at that. No one really knows, and in fact it may be impossible for any individual or group of individuals to know, the ‘formula’; it may resemble something like a really complicated Fourier transform with articulated coefficients that change based on unpredictable search patterns.
But a little differentiation can make a secular difference given the huge number of searches. One that we’ve found is content density: a web page’s ratio of characters outside of html tags after whitespace is eliminated, to the number of characters within the tags themselves. We noticed this when we were testing one of our products that isn’t related to search at all. It was intriguing enough that we built an automated test that looks at the content density for the first 20 results in random searches.
The content density is not an absolute number, but relative to the average of a search sequence, as related to the position within the search results. The dispersion cloud is large and thin, and the content density as related to position for a given search category varies considerably. But, overall, the slope of content density from the first to the 20th position in a search result set is negative, but the variation between the first and 20th position seldom exceeds 0.3% on the several thousand test runs. About 1% of the tests also looked at a search result’s 50th position, which trended lower that that of the 20th position, but by a much smaller amount: 0.05% using the same averaging techniques.
Every little bit helps, and this is certainly little. Increasing content density may help rankings, easily done by moving as much overhead out of the webpage itself: explicit css directives (use external files), javascript (ditto), explicit internal links instead of relative ones, short titles and alternates, and minimizing tag usage generally.
Whether content density is a significant factor is conjecture. Other things may have more weight, things we didn’t look for in our tests. Such as spelling, phrase usage, and the boring details of diagramming sentences that had to be endured in grade school (perhaps now we know of what use it is). The keyword clouds are going to vary over time, sometimes with amazing rapidity depending on the mood of the world at any given instant.
While adwords may be helpful in getting visitors, it should not be the only advertising and marketing tool in any company’s promotional activities.
© Copyright 2010 Chuck Brooks for FutureWare SCG
A Word From Our Sponser
Finally, a fast way to fix 
html scripts without having to rebuild them!
Tags: adwords, seo, web design