The third in a series of We Heart Log Data posts from Google brings back an old friend and a controversial topic.
Fresh off his appearance at SES New York, Google’s Shuman Ghosemajumder, Business Product Manager for Trust & Safety, dropped a little blog post goodness on the Internet with his take on logfiles and their importance. As one of the many people working against click fraud at the search ad company, we’re always delighted to see him provide some commentary on the subject.
His post at the official Google blog on logfiles and click fraud fell into line with the two previous posts about Google’s use of the data it captures. Logfiles are good, Google has the best of intentions in using that data, and they’ll protect it from all threats.
On its applicability to combating click fraud, Ghosemajumder wrote that logs hold the clues to detecting click fraud. “Logs provide us with the repository of data which are used to detect patterns, anomalous behavior, and other signals indicative of click fraud,” he said.
Google looks at the AdWords clicks as they arrive, checking for details like IP addresses and other factors Ghosemajumder declined to mention. As per their new policy, Google anonymizes logfiles after 18 months.
Google’s nigh-obsessive need to talk about logfiles started on the cusp of European Union approval of their DoubleClick acquisition. Complaints about privacy arose almost immediately when Google first disclosed the deal last year.
And well they should have. Google will have an immense repository of Internet user data available to it, once all the database connections have been put in place (if they haven’t already.)
Convincing people the company won’t do bad things with this access isn’t the crux of the concerns. Google wants to portray its role as a virtually druidic custodian of information, which it will care for and nuture. That data plays a critical role in its business, and how well it can address the always thorny issue of click fraud for its advertisers.
