Google Website Tracker Has Ironic Flaw

A new Google tool to track the comparative popularity of websites has a notable hitch: it doesn't measure Google itself. The service also doesn't cover sites owned by Google, such as YouTube or Blogger.

Google Trends for Websites is an extension of the existing service which measures the phrases people are looking for most often and how popularity changes over time.

The new service tracks websites, making it a competitor to established trackers such as Alexa.com. Any user can get a graph showing traffic over time, while registered users can get further details such as estimated numbers and a site's popularity in different countries. (Source: webware.com)

Why Google-owned sites are blocked from the service is disputed. Officially, the company claims it isn't allowed to give out traffic details outside of official financial filings. But independent commentators say it's more likely a flaw in the way the system works.

It appears Google is getting its traffic data by tracking what people with the Google toolbar installed are doing and then extrapolating the results. In general, this should be reasonably accurate as Google has a large customer base and there's no reason to think its users aren't fairly representative of Internet users as a whole.

The problem is that virtually everyone who uses the Google toolbar will by definition be a regular visitor to Google's own website. That means the estimated traffic levels for the Google site would be misleadingly high. That wouldn't just make the service as a whole appear less credible, but it could conceivably cause regulatory problems if it seemed Google (a public company) was overstating its own success. (Source: techcrunch.com)

To be fair, rival services like Alexa.com face the same problem. But Google's high profile means they are the first company to face so much scrutiny for this ironic, but technically unavoidable, problem.