You are here
Interventional Targeting? "Get into people's heads" -- Part IX of Privacy-Publicacy Faultline Series
Submitted by Scott Cleland on Tue, 2009-05-19 11:24
"Google's algorithm helps the company "get inside people's heads even before they know they might leave," said Laszlo Bock, who runs human resources for the company." See the Wall Street Journal article: "Google Searches for Staffing Answers."
- The Journal article explains that in an attempt to retain employees, Google crunches employee data in a mathmatical formula to determine which employees are at most risk of leaving. As is Google's standard practice, "Google officials are reluctant to share details of the formula..."
What intrigued me was the bold claim above by Google's head of personnel is that the company has endeavored to, and succeeded in, systematically getting in "peoples heads" to learn their intimate concerns before someone even figures out what they are feeling.
Why this incident is so illustrative of the increasing tension between people's expectation of privacy -- and the publicacy business model to make money off of people's private information -- is that it spotlights how far publicacy companies may delve into this privacy grey area while at the same time not disclosing the details necessary to determine if the practice creates a privacy problem.
Let's connect some dots.
- Google's search advertising business is based on intention-based collective intelligence, i.e. knowing private information on what people may want to buy so they can serve them ads.
- Algorithms are only as good as the data they calculate.
- Google's core algorithms are driven off of words, because words are the key data that Google's crunches to determine intentions.
- Google tracks words in search, gmail and other Google apps to determine user intentions.
- So where are the best sources of words that Google employees may disclose that signal what they may be feeling even before they are consciously aware of it?
- The best data sources to get into Google employees heads are their searches, their gmail incoming and outgoing, their phone patterns via GrandCentral, where they go via Google Maps and Google Android's Latitude location tracking program, etc.
- In other words, the most efficient alogrithm to get into a Google emloyee's head is to algorithmically track the employees interactions with others -- i.e. if recruiters, competitors or former employees are searching on or contacting Google employees... etc.
My point is that if Google aspires to get into people's heads to the point of knowing what they will do before even they know it... that is either a pretty troublesome aspiration or one that requires a whole lot of internal controls to ensure that this intimate knowledge and responsibility is not abused.
- Google's publicacy capabilities and plans apparently are moving beyond behavioral advertising, interest-based targeting, attitudinal targetting -- to the ultimate influence -- interventional targeting.
- Contemplate how much more privacy-publicacy tension this trend could cause.
- What if other corporations wanted Google to monitor or crunch data on their employees so they could target "retention interventions?"
- What if political candidates, parties or governmental entities asked Google for profiles of voters whom the alogrithm indicates might in the future consider switching candidates, party affiliations, or positions on issues, but don't yet know it -- for the purposes of "political intervention?"
- What if groups on either side of the question asked to target individuals who might have latent or unresolved questions about their either their sexual orientation or their religious beliefs -- for the purposes of "decisional intervention?"
If one does not see the slippery slope here between privacy and publicacy one is not paying attention. What should set off alarm bells is when this type of slippery slope publicacy behavior is not fully transparent or fully vetted in public debate.
- Wouldn't most people want to know if they are the "target" of a publicacy intervention to influence their life path?
Privacy-Publicacy Faultline Series here:
- Part I: The Growing Privacy-Publicacy Fault-line -- The Tension Underneath World Data Privacy Day
- Part II: Implications of User Location Tracking
- Part III: Extreme Publicacy -- Does Privacy Stand a Chance?
- Part VI: Why FTC’s Behavioral-Ad Principles Are a Big Deal
- Part V: Privacy prevailed in Facebook's privacy-publicacy earthquake
- Part VI: Do People Own Their Private Information Online?
- Part VII: Where is the line between privacy and publicacy?
- Part VIII: "Privacy is Over"