Commentary

Google Shares Analytics Data

Google-Analytics

Sharing data with Google or Microsoft isn't new. Install a Web browser and the respective company likely asks for help in identifying bugs and glitches by agreeing to share data collected from an event.

In Google Analytics, the benefit of sharing data is the promise of having access to benchmarks. In the past Google provided the data, but not in a clean and comprehensive format, according to marketers. They had to navigate through the user interface to dig out the data. Now Google provides that data in a newsletter. Yes -- benchmark data delivered into an email inbox.

The first volume of the Analytics Benchmarking Newsletter pushed out earlier this month replaces the standard benchmarking report for Google Analytics account holders with data shared in this newsletter. Google will use the newsletter as an experiment to "surface more useful or interesting data to analytics users," the newsletter reads. Data contained in the letter comes from all Web sites that have opted in anonymous data-sharing with Google Analytics. The newsletter explains that only those Web site administrators who have enabled the anonymous data sharing will receive this benchmarking newsletter.

The newsletter is the first time that Google has pushed the data to marketers. The breakdowns are by geography, rather than industry. It groups all Web sites in the U.S. together. For example, the average number of pages in the U.S. stands at 4.7. The newsletter shows bounce rates and time, but benchmarks all Web sites in the U.S. that use Google Analytics and opted in to share the data. It should be segmented, at the very least, by vertical.

Lumping in all Web sites in the U.S. is not a helpful metric for Kenshoo CMO Aaron Goldman, although Google does segment geography by traffic source and conversion rate. It also calls out stats by operating system. Goldman said the data is interesting but not useful, and he hopes Google would segment it by verticals such as retail, travel, financial services and automotive.

Slice the time and bounce rates by vertical so marketers can identify and compare. For example, visitors who come to one retail site might get five page views, but the average is eight. It would tell this retailer they might not have correctly optimized product listings.

"If the conversion rate is higher than the average, I don't care how many pages they visit clicks on because if it only takes three pages for them to convert rather than eight, I'm doing something right," Goldman said.

A clear opt-in process provides marketers with an option. But while the newsletter aims to bring transparency to data, some might wonder whether it opens privacy issues by pushing out the data, rather than having marketers dig through the site to find it.

Next story loading loading..