Archive for the ‘Data analysis’ Category

The limits of data analytics …

January 19, 2017

Team Clinton worshipped at the altar and got burned.

======

Lots of post-election articles about how the Clinton campaign got fixated on their data-rich electorate models, using them to allocate ad dollars, deploy field workers and schedule “market visits” by Hillary and her surrogates.

clip_image002

What Team Clinton seemed to have forgotten is the old Reagan adage: trust but verify.

The data models – which worked near-flawlessly for Obama – took stage as “shiny objects” that led the Clinton campaign astray.

Politico reported a case study that  illustrated the point …

(more…)

Did data analytics miss the forest for the trees?

November 17, 2016

Team Clinton’s GOTV effort got out a lot of votes … for Trump

======

huffpost-big-data-clinton

According to the Huffington Post:

As the post-election day hangover wears off, an examination of the mechanics behind the Clinton’s get out the vote efforts ― reaching out to Clinton voters in key states at the door, on the phone or by text messages ― reveals evidence of what appears to be a pretty shocking truth.

Clinton volunteers were inadvertently turning out Trump voters.

Possibly in significant numbers.

What went wrong? (more…)

Trump : “Data analytics is overrated” … could he be right?

August 2, 2016

Last week, the WSJ ran an opinion piece: Trump’s Big Data Gamble.

The punch line: “While Donald tweets to the masses, Hillary will be precisely targeting persuadable voters.”

Advantage Hillary, right?

Maybe.  Maybe not.

In an AP interview, Trump said that he “always thought that it (meaning data analytics) was overrated” and, accordingly, he’ll spend limited money on data operations to identify and track potential voters and to model various turnout scenarios that could give him the 270 Electoral College votes needed to win the presidency.

He’s moving away from the model Obama used successfully in his 2008 and 2012 wins, and the one that likely Democratic nominee Hillary Clinton is trying to replicate, including hiring many of the staff that worked for Obama in his “Victory Lab”.

clip_image002
=====

A data-light strategy may sound very old-school in the era of big data … especially coming from Trump …. but it reminded me of an opinion piece that Peggy Noonan wrote in the WSJ soon after Obama’s 2012 election win.

Noonan had a riff about predictive analytics that caught my eye.

It pointed out one of the downsides of predictive analytics … the craft of crunching big data bases to ID people, their behaviors and their hot buttons.

Here’s what Noonan had to say …

(more…)

Newsflash: Virgin brides are less likely to get divorced … but, are an endangered species.

June 10, 2016

Yep, that’s one conclusion reported in a recent Family Studies research brief.

Focusing on the most current data – the yellow line — only about 1 in 20 virgin brides end up getting divorced … rates are substantially higher for non-virgin brides.

clip_image002

=======

The interpretation of the rest of the data reported by Family Studies is downright wacky, so let’s dig a little deeper…

(more…)

The Challenger disaster: A tragic lesson in data analysis …

June 6, 2016

Well-intended engineers correctly interpreted the wrong data.

Excerpted from Everydata: The Misinformation Hidden in the Little Data You Consume Every Day

=======

I’m sure all baby-boomers have a vivid recollection, but for younger readers, here’s some background …

“On the morning of 28 January 1986, the Space Shuttle Challenger, mission 51– L, rose into the cold blue sky over the Cape. To exuberant spectators and breathless flight controllers, the launch appeared normal. Within 73 seconds after liftoff, however, the external tank ruptured, its liquid fuel exploded, and Challenger broke apart.”

clip_image002

=======

What happened?

“The specific failure,” noted the Report of the Presidential Commission on the Space Shuttle Challenger Accident, “was the destruction of the seals that are intended to prevent hot gases from leaking.…”

Investigators quickly focused their attention on a key part of the seals— the rubber O-rings that went in between two sections of the solid rocket motor— the “tang” and the “clevis.”

The O-rings on the Challenger needed to be flexible enough to compress and expand, sometimes within milliseconds.

But O-ring resiliency “is directly related to its temperature… a warm O-ring will follow the opening of the tang-to-clevis gap. A cold O-ring may not.”

In fact, investigators found that a compressed O-ring is five times more responsive at 75 degrees Fahrenheit than at 30 degrees Fahrenheit.

The air temperature at launch was 36 degrees Fahrenheit.

The commission’s report found “it is probable” that the O-rings were not compressing and expanding as needed.

The resulting gap allowed the gases to escape, destroying the Challenger.

=======

So why didn’t engineers stop the launch, given the cold temperatures?

(more…)

Trump : “Data analytics is overrated” … could he be right?

May 19, 2016

In an AP interview, Trump said that he “always thought that it (meaning data analytics) was overrated” and, accordingly, he’ll spend limited money on data operations to identify and track potential voters and to model various turnout scenarios that could give him the 270 Electoral College votes needed to win the presidency.

He’s moving away from the model Obama used successfully in his 2008 and 2012 wins, and the one that likely Democratic nominee Hillary Clinton is trying to replicate, including hiring many of the staff that worked for Obama in his “Victory Lab”.

clip_image002
=====

A data-light strategy may sound very old-school in the era of big data … especially coming from Trump …. but it reminded me of an opinion piece that Peggy Noonan wrote in the WSJ soon after Obama’s 2012 election win.

Noonan had a riff about predictive analytics that caught my eye.

It pointed out one of the downsides of predictive analytics … the craft of crunching big data bases to ID people, their behaviors and their hot buttons.

Here’s what Noonan had to say …

(more…)

Hacked: Cards expose Moneyball’s strategic vulnerabilities …

June 18, 2015

Moneyball – the Oakland As use of data  and metrics to ID undervalued players —  was one of the  major catalysts for the current rage around big data and data analytics.

The Houston Astro’s  were one of the teams to adopt the Moneyball philosophy in a big way.

This week, the NY Times broke the story that the St. Louis Cardinals had hacked into Astro’s proprietary database.

Big news.

In fact, this hack seemed to get more media time than  the Chinese jacking the personal info of all government employees.

Hmmm.

image

Baseball competition aside, here’s why I think there’s a big teaching point in the story

(more…)

What does this map represent?

January 7, 2013

Take a guess …

image

No good reason for anybody to know.

It’s a mapgraphic depicting the 1,540 Walmart stores in 1990.

So what?

Here’s what makes it interesting.

For a cool, dynamic visual showing how & where Walmart has grown over the years, click the link to view FlowData.com’s Walmart growth map.

The content is interesting, and it’s a nice way to present geo-time series data over time.

* * * * *
Follow on Twitter @KenHoma         >> Latest Posts

NetTrax: If you think that you’re being followed around on the net … you’re right.

October 2, 2012

And the company doing it is probably Acxiom … recently profiled in the NY Times.

I had some weird happenings recently.

A friend who works internet marketing for a “plus sized” women’s clothes company used my computer to show me her site’s cool redesign.

For the next couple of weeks I was getting pop-up ads for big women’s clothes … even when I was on common sites like ESPN or WSJ.

Huh?

Another time, I checked the spelling of a Spanish word via Google.

Next couple of times that I went to You Tube, the lead in ads were in Spanish.

Double huh?

I was wondering how the web “knew” … now I know, courtesy of the NY Times.

Here are some highlights …

Acxiom

IT knows who you are. It knows where you live. It knows what you do.

It peers deeper into American life than the F.B.I. or the I.R.S., or those prying digital eyes at Facebook and Google.

If you are an American adult, the odds are that it knows things like your age, race, sex, weight, height, marital status, education level, politics, buying habits, household health worries, vacation dreams — and on and on.

Right now, more than 23,000 computer servers are collecting, collating and analyzing consumer data for a company …  called the Acxiom Corporation, the quiet giant of a multibillion-dollar industry known as database marketing.

Acxiom has amassed the world’s largest commercial database on consumers —  Its servers process more than 50 trillion data “transactions” a year.

Acxiom maintains its own database on about 190 million individuals and 126 million households in the United States

Its database contains information about 500 million active consumers worldwide, with about 1,500 data points per person.

Acxiom’s Consumer Data Products Catalog offers hundreds of details — called “elements” — that corporate clients can buy about individuals or households, to augment their own marketing databases. Companies can buy data to pinpoint households that are concerned, say, about allergies, diabetes or “senior needs.”

In a fast-changing digital economy, Acxiom is developing the most advanced techniques to mine and refine data.

Digital marketers already customize pitches to users, based on their past activities … think “cookies”.

Acxiom  is pursuing far more comprehensive techniques in an effort to influence consumer decisions.

It is integrating what it knows about our offline, online and even mobile selves, creating in-depth behavior portraits in pixilated detail …  Its  a “360-degree view” on consumers.

 

How it works

Scott Hughes, an up-and-coming small-business owner and Facebook denizen, is Acxiom’s ideal consumer.

In fact,  Acxiom created him.  Mr. Hughes is a fictional character who appeared in an Acxiom investor presentation in 2010.

A frequent shopper, he was designed to show the power of Acxiom’s multichannel approach.

In the presentation, he logs on to Facebook and sees that his friend Ella has just become a fan of Bryce Computers, an imaginary electronics retailer and Acxiom client.

Ella’s update prompts Mr. Hughes to check out Bryce’s fan page and do some digital window-shopping for a fast inkjet printer.

Such browsing seems innocuous — hardly data mining. But it cues an Acxiom system designed to recognize consumers, remember their actions, classify their behaviors and influence them with tailored marketing.

When Mr. Hughes follows a link to Bryce’s retail site, for example, the system recognizes him from his Facebook activity and shows him a printer to match his interest.

He registers on the site, but doesn’t buy the printer right away, so the system tracks him online.

Lo and behold, the next morning, while he scans baseball news on ESPN.com, an ad for the printer pops up again.

That evening, he returns to the Bryce site where, the presentation says, “he is instantly recognized” as having registered.

It then offers a sweeter deal: a $10 rebate and free shipping.

It’s not a random offer.

Acxiom has its own classification system, PersonicX, which assigns consumers to one of 70 detailed socioeconomic clusters and markets to them accordingly.

In this situation, it pegs Mr. Hughes as a “savvy single” — meaning he’s in a cluster of mobile, upper-middle-class people who do their banking online, attend pro sports events, are sensitive to prices — and respond to free-shipping offers.

Correctly typecast, Mr. Hughes buys the printer.

But the multichannel system of Acxiom and its online partners is just revving up.

Later, it sends him coupons for ink and paper, to be redeemed via his cellphone, and a personalized snail-mail postcard suggesting that he donate his old printer to a nearby school.

 

Waste”

There is a fine line between customization and stalking.

While many people welcome the convenience of personalized offers, others may see the surveillance engines behind them as intrusive or even manipulative.

Privacy advocates say they are more troubled by data brokers’ ranking systems, which classify some people as high-value prospects, to be offered marketing deals and discounts regularly, while dismissing others as low-value — known in industry slang as “waste.”

Exclusion from a vacation offer may not matter much …  but if marketing algorithms judge certain people as not worthy of receiving promotions for higher education or health services, they could have a serious impact.

“Over time, that can really turn into a mountain of pathways not offered, not seen and not known about.”

A bit creepy, right?

>> Latest Posts

Behavioral analytics … bad when Target does it … OK for political campaigns?

September 19, 2012

A couple of months ago Target got some bad press when it was revealed that the company was mining customers’ purchase histories to slot them into behavioral groups susceptible to tailored promotional pitches.

For example, Target identified purchases that mothers-to-be made early in their pregnancies – sometimes before they even knew they were pregnant.  Think bigger jeans, skin care lotions.

Many folks railed that it was an example of big brother invasion of privacy.

Well, guess what?

Political campaigns are using the same methods that Target was using

The modern science of politics is increasingly based on principles from behavioral psychology and data analytics.

image

Campaigns today mine large data bases with mathematical algorithms that slot folks into categories and provide the basis for how people should be approached (or ignored).

According to the WSJ:

Perhaps the most valuable data in modern campaigns comes from statistical “microtargeting” models—the political world’s version of credit scores.

Campaigns gather thousands of data points on voters, culled from what they put on their registration forms, what they have told canvassers who have visited their homes in the past, and information on their buying and lifestyle habits collected by commercial data warehouses.

The campaigns then run algorithms trawling for patterns linking those demographic characteristics to the political attitudes measured in their polling.

Financial institutions run such statistical models to generate predictions about whether a given individual will demonstrate a certain behavior, like paying a bill on time or defaulting on a loan.

Campaigns do the same, only they are interested in predicting political behavior.

So it’s typical now to generate individual scores, presented as a percentage likelihood, that a voter will cast a ballot, support one party or the other, be pro-choice or antiabortion, or respond to a request to volunteer.

These scores now stick to voters as indelibly as credit scores.

And just as a bank officer won’t sign off on a loan without requesting one, a field director for a campaign won’t send a volunteer to a voter’s door without knowing the relevant number.

BTW: It’s Team Obama that’s doing most of this stuff.

Bad for Target … but OK for Obama.

Hmmm

* * * * *

WSJ source: “The Victory Lab: The Secret Science of Winning Campaigns” by Sasha Issenberg

>> Latest Posts