We examine the origin and the truth behind this explosive story, the prominence of headlines, and how unsubstantiated assumptions gain traction and also mainstream fist and help create myths roughly Predictive Analytics.
You are watching: How target figured out a teen girl was pregnant before her father did
By Gregory Piatetsky, might 7, 2014.
"How Target established A teen Girl was Pregnant before Her dad Did" was an explosiveheadline in a Forbesarticle by Kashmir Hill (Feb 2012) which caused a large media storm.The Forbes post only report on an short article by Charles Duhigg in the new York Times v a quieter titleHow Companies learn Your Secrets. Duhigg"s article reported on work by presented in 2010 through a Target statistician in ~ PAW - the Predictive Analytics civilization Conference. The presentation had a really technical title ("How Target gets the many out the Its Guest Data to boost Marketing ROI") and also did no cause any kind of controversy.Good lesson here on the prestige of headlines.But walk Target"s predictive analytics really identify and also reveal a teen pregnancy?
I had a opportunity to discuss this recently with Eric Siegel, who discussed that very likely this was no the case.Eric is the founder that the conference series Predictive Analytics civilization (www.pawcon.com), i m sorry is the first and top cross-vendor event that covers commercial deployment.His work-related with PAW gave him special understanding into the real scoop that the Target story and also which components of it are "real".Here"s an excerpt from chapter 2 that his book "Predictive Analytics: The strength to Predict who Will Click, Buy, Lie, or Die" (www.thepredictionbook.com). The chapter concentrates on privacy and also civil liberties involves that happen in predicting, e.g., pregnancy, job quitting, and also crime recidivism.Excerpt native "Predictive Analytics" (www.thepredictionbook.com) by Eric Siegel:In 2010, I invite an skilled at Target, Andrew Pole, come keynote in ~ Predictive analytics World, a conference because that which I serve as regime chair. Pole manages dozens the analytics professionals who run various predictive analytics (PA) jobs at Target. In October of the year, Pole ceded a mainly keynote on a wide range of PA deployments at Target. He take it the stage and dynamically engaged the audience, revealing detailed examples, amazing stories, and meaningful business results the left the audience clearly enthused. Free to view, below it is:www.pawcon.com/Target.Toward the end, Pole defines a task to predict customer pregnancy. Given that there"s a remarkable sales opportunity when a family prepares because that a newborn, you can see the marketing potential.But this was something pointedly new, and I turned mine head to scan the audience for any type of reactions. Nothing. Nada. Zilch. Normally, for marketing projects, PA predicts purchase behavior. Here, the point being suspect was no something marketers care around directly, but, rather, other that might itself be a strong predictor the a wide range of to buy needs. After all, the marketer"s project is come discover and also pounce on demand. You have the right to think the this predictive goal as a "surrogate" (sorry) because that the pertinent shopping activities a sleeve marketer is paid to care about.A couple of months ~ Pole"s presentation, new York times reporter Charles Duhigg interviewed me. Exploring, that asked because that interesting discoveries that had actually come native PA. I rattled off a couple of and contained pregnancy prediction, pointing him to the online video clip of Pole"s talk, which had thus much been receiving small attention, and introducing him come Pole. I should admit the by now the privacy question had actually left my mind nearly entirely.One year later, in February 2012, Duhigg published a front-page brand-new York times Magazine article, sparking a famous outbreak that turned the Target pregnancy prediction story into a debacle. The article, "How Companies learn Your Secrets," conveys a ton that means wrongdoing is a foregone conclusion. The punctuates this by alleging an anonymous story that a man discovering his teenage daughter is pregnant only by seeing Target"s marketing offers to her, through the unsubstantiated however tacit implication that this resulted particularly from Target"s PA project.This well-engineered splash motivated rote repetition by press, radio, and television, all of whom blindly took as gospel what had only been implied and ran with it. Not incidentally, it aided launch Duhigg"s book, "The power of Habit: Why We perform What We perform in Life and also Business," which struggle the new York Times best seller list.Beyond this, as Eric discussed to me, the brand-new York Times post itself provides one more factoid do it also less most likely the teen"s pregnancy had actually been established analytically (if "determined" through Target at every - maybe the certain teen to be simply inserted accidentally into the wrong marketing segment): Target to know consumers might not prefer to be marketed on baby-related commodities if they had actually not volunteered your pregnancy, and so proactively camouflages such activities by interspersing together product placements among other non-baby-related products. Such marketing material would by design not progressive any certain attention that the teen"s father.Target story was questioned recently in jae won Timesarticle,where Kaiser Fung offered an additional explanation:Target mixes increase its offers not due to the fact that it would certainly be weird come send an all-baby coupon-book to a woman that was pregnant but since the firm knows that numerous of those coupon books will be sent out to women that aren’t pregnant after all.You deserve to read the rest of the chapter in Eric"s publication for an ext insight ~ above this story, and also on the privacy and civil liberty concerns brought up by predictive analytics in general.Read additionally the various other chapters - as I wrote in my proof forPredictive Analytics: The power to Predict who Will Click, Buy, Lie, or Die:"Written in a lively language, full of great quotes, real-world examples, and case studies, it is a pleasure to read.
See more: How Much Does The Air Force One Plane Cost To Operate And Maintain Air Force One
The an ext technical audience will enjoy chapters ~ above The Ensemble Effect and uplift modeling - both very hot trends. I extremely recommend this book!"