Stephen Baker

The Boost
Home - posts tagged as Tribes

How to battle shopping barnacles
October 21, 2008Tribes

Barnacles. They're the shoppers who look for bargain deals and try never to buy anything at full price. Many retailers want to locate these people and "fire" them. How to do that? For starts, they can stop spending money to send them catalogs.

On MineThatData, Kevin Hillstrom offers 43 tips, including many to deal with pesky barnacles. Here are three:

Returns: Customers who return at least sixty percent of the merchandise they purchased are likely to return merchandise in the future. If you're looking to trim marketing expense, here's a place to do so without negatively impacting profitability. (ie. don't send them catalogs)

Fulfillment: Customers who failed to receive items in their last order are often less likely to order in the future than are customers who receive what they wanted to purchase.
(ie. drop them)

Channel: Customers who order from catalogs over the telephone "age slowly", meaning these customers can be profitably targeted for a long time. Customers who order online age faster, meaning that after "x" months they become unresponsive. Customers who order via stores age very fast, becoming unresponsive soon after a purchase.

add comment link to post share:

What would you infer about the book based on this reader?
September 19, 2008Tribes

I just got this screen shot from my blog designer at InfinetDesign. He was doing a search on Amazon, and his recent history popped up, along with the behavior of people like him. (I cropped the page to make it easier to see.)

Now I worked really hard to make The Numerati a general-interest book. But if screen shots like this get around, people will come to pretty drastic conclusions about the Numerati "tribe"--and our marketing efforts with the mainstream crowd, I fear, are doomed.

add comment link to post share:

Cell phone data: Is six degrees passe?
September 3, 2008Tribes

TechCrunch points to a new study from the French branch of cell-phone carrier, 02. It shows that the familar notion of Six Degrees of Separation (which led to a John Guare play
and provided the title for an excellent book on networking and complexity by Duncan Watts) may be obsolete.

From the O2 release:

The term was coined by US psychologist Stanley Milgram following a 1967 experiment. The six degrees theory was upheld in a 2006 Microsoft study of instant messenger conversations. However, the O2 study reveals that within a shared ‘interest’ network (i.e. hobbies, sport, music, religion, sexuality etc), the average person is connected by just three degrees.

[Social organisational specialist Jeff] Rodrigues finds that we are usually part of three main networks based on family, friendship and work. Outside of these we are, on average, part of five main shared ‘interest’ networks based on a range of personal interests from hobbies, sport, music and the neighbourhood we live in, to religion, sexuality and politics. It is the growth of these shared interest networks and the influence of technology on them that has led to the reduction in the number of degrees of separation.

Not to belabor the point, but the Numerati in marketing, media and politics are pouring over these networks of ours--and trying to figure out which tribes to place us in.

add comment link to post share:

NewsCred responds: algorithms to filter out bias
August 26, 2008Tribes

The other day, I asked how NewsCred could adjust its algorithms to weed out political partisanship. I got a response on Blogspotting from Shafqat Islam, a co-founder. He says that they've been working hard from the get-go on the very question I raised. They're looking at different ways to weigh votes, credits and discredits. And they're using statistical models developed in studies of public health epidemiology. (This is one of the key points about the Numerati--and what I find so fascinating in the subject. A breakthrough in one area can illuminate another. In BusinessWeek, I wrote about a Microsoft researcher who redirected a spam-fighting algorithm against HIV/Aids. In NewsCred's case epidemiologists may have important lessons for next-gen editors.)

Shafqat writes:: "...[P]olarizing crowds do pose a problem to any algorithm that depends on explicit inputs/votes from the community. But contrary to Steve's first comment, we did think of this from the very beginning and are continuously working on improving the algo to cater for this.

"First and foremost is building technical barriers towards gaming and herd mentality (for example, when 100 people sign up just to discredit a journalist). Preventing that behavior is fairly straightforward given tools available today.

"But to really handle polarized crowds, we've turned to public health epidemiology and looking at how bias plays a role in voluntary surveys. Thats essentially the same issue we face on a 'voluntary' voting site. Without giving too much away as we still work on the algorithm, perhaps treating a read without a vote as an explicit input (with a smaller weight than a credit/discredit) can help. There are other options, all around weighing votes that we are exploring. I think the most important point is that we're always working on it, and will make all our trials and conclusions transparent once fully implemented."

On the same subject, Mathew Ingram suggests that readers, given the right tools, could dig into the credibility rankings by themselves. If you read the comments on his post, Shafqat responds at some length.

add comment link to post share:

What's the relative value of an Obama bumpersticker?
August 4, 2008Tribes

A hypothetical. Let's say I have a very close friend in the liberal enclave of Montclair who just bought a Mini Cooper. She wants to put an Obama bumpersticker on it. How much is that bumpersticker worth to the Obama campaign? How many Mini lovers are torn between the two candidates, see the bumpersticker, and conclude: Someone with my values is voting for Obama?I would say very few. In fact, I think it would confirm for many SUV and truck drivers that Obama is somewhat alien, a candidate loved by wealthy small-car lovers. The only car worth less to Obama than a Mini would be a Prius.

If this were an advertising marketplace, like Google's adwords, how much more would it be worth for Obama to have his bumpersticker on a rusted pick-up, ideally equipped with a gun rack? A hundred times more valuable? At least, I'd say.

add comment link to post share:

What does my lawn say about my values?
July 28, 2008Tribes

Reading The New Yorker, I finally found the right name for my lapsed lawn. It's a "freedom lawn." My question: If we look at a barely-tended lawn as a verdant, weed-strewn stretch of personal data, what does it say about me? Could The Numerati draw conclusions about me as a voter or a shopper by looking at the horticultural mix in my yard?

Here's how Elizabeth Kolbert describes it:

...[T]he simplest alternative to the modern, industrialized lawn may be a lawn that functions more or less as it did in the eighteen-forties, before herbicides or even sprinklers had been invented. In “Redesigning the American Lawn” (1993), F. Herbert Bormann, Diana Balmori, and Gordon T. Geballe dub such a lawn the Freedom Lawn. The Freedom Lawn consists of grass mixed with whatever else happens to seed itself, which, the authors note, might include:

dandelion, violets, bluets, spurrey, chickweed, chrysanthemum, brown-eyed Susan, partridge berry, Canada mayflower, various clovers, plantains, evening primrose, rushes, and wood rush, as well as grasses not usually associated with the well-manicured lawn, such as broomsedge, sweet vernal grass, timothy, quack grass, oat grass, crabgrass, and foxtail grass.
The Freedom Lawn is still mowed—preferably with a push-mower—but it is watered infrequently, if at all, and receives no chemical “inputs.” If a brown spot develops, it is likely soon to be filled by what some might call weeds, but which Bormann, Balmori, and Geballe would rather refer to as “low growing broad-leaved plants.”

This has been an interest of mine for years. I'm probably the only person at BusinessWeek who's ever written an advocacy piece for push mowers.

I was thinking about lawn/politics correlations as I biked through Montclair this weekend. I was on a very beautiful street called Gordonhurst, where all the lawns were wonderfully tended. And I thought: Would I ever hate to live here! We're talking about an enviable street, one where the houses cost 2 or 3 times more than mine. And yet I didn't want to live there, because I'd feel pressure to do work to fit in with people whose values--in that one grassy domain--were not mine. So, the question: Are people with neat yards more likely to vote in a certain way? I'll have to ask some of the political Numerati...

link to post share:

How about another Facebook page for real friends?
July 26, 2008Tribes

I don't know what to make of my Facebook account. My mix of friends includes lots of colleagues, a handful of business contacts, and a high school classmate who swore up and down at a reunion 10 years ago that he didn't remember me. (Now, either his memory has improved or he's invited everyone on our high school list.) But if you look at the people who traveled to my wedding long ago (in Mexico) as my closest friends in the world (in addition to a few friendships I've forged since), not one of that group is following my sporadic activity on Facebook. (I stay much busier on Twitter)

So should I set up another Facebook account for my friends in the old-fashioned sense of the word? I probably won't do it, because most of those old analog friends of mine don't bother with such things (as far as I know). But it might be interesting to manage two profiles and two different worlds. The question for the new Me: Is it worth friending that rather shallow, promotional and scatter-brained me with the eccentric friend list? Should I trust him?

add comment link to post share:

Should we provide airlines with our dating questionnaires?
July 19, 2008Tribes

Take a look at all those people waiting to get on the same flight from Boston to Seattle. That's what Scott Brinker does. They have one thing in common--a destination. But they have hundreds or even thousands of different preferences when it comes to music, wine, food, movies, air-conditioning and whether or not to chat with loudmouths in the next seat. (Some of them, of course, are loudmouths and have no hang-up in this area.) Brinker analyses this as a search marketing challenge. How to figure out what people want from only one piece of data?

But just imagine if the airlines had questionnaires similar to those people fill out for online dating. They could figure out what kind of travelers we are and provide us with stupendous service. This isn't the greatest example, of course, because airlines have no money to provide much more than cattle car service. But I'm thinking that we'll soon be in control of our master profiles. They'll contain our preferences and maybe even the patterns of our Web wanderings. They will describe everything from our allergies to our taste in music. We'll entrust them to certain companies in exchange for customized service. And when they betray our trust and sell these profiles, as some will, I guess we'll sue.


add comment link to post share:

©2018 Stephen Baker Media, All rights reserved.     Site by Infinet Design

Kirkus Reviews -

LibraryJournal - Library Journal

Booklist Reviews - David Pitt

Locus - Paul di Filippo

read more reviews

Prequel to The Boost: Dark Site
- December 3, 2014

The Boost: an excerpt
- April 15, 2014

My horrible Superbowl weekend, in perspective
- February 3, 2014

My coming novel: Boosting human cognition
- May 30, 2013

Why Nate Silver is never wrong
- November 8, 2012

The psychology behind bankers' hatred for Obama
- September 10, 2012

"Corporations are People": an op-ed
- August 16, 2011

Wall Street Journal excerpt: Final Jeopardy
- February 4, 2011

Why IBM's Watson is Smarter than Google
- January 9, 2011

Rethinking books
- October 3, 2010

The coming privacy boom
- August 17, 2010

The appeal of virtual
- May 18, 2010