"InShorts: A Game-Changer in the Digital News Age"
How Much Do We Pay for Things Numbersense by Kaiser Fung
2. NUMBER
SENSEHOW TO USE BIG DATA
TO YOUR ADVANTAGE
K AISER FUNG
New York Chicago San Francisco Athens London Madrid Mexico City
Milan New Delhi Singapore Sydney Toronto
3. Contents
Acknowledgments vii
List of Figures ix
Prologue 1
PART 1 SOCIAL DATA 17
1 Why Do Law School Deans
Send Each Other Junk Mail? 19
2 Can a New Statistic Make Us
Less Fat? 54
PART 2 MARKETING DATA 75
3 How Can Sellouts
Ruin a Business? 77
4 Will Personalizing Deals
Save Groupon? 95
5 Why Do Marketers
Send You Mixed Messages? 112
4. vi CONTENTS
PART 3 ECONOMIC DATA 127
6 Are They New Jobs
If No One Can Apply? 129
7 How Much Did You Pay
for the Eggs? 153
PART 4 SPORTING DATA 173
8 Are You a Better Coach
or Manager? 175
EPILOGUE 201
References 211
Index 219
6. 1 2 3 4 5 6 7 8
How Much Did You Pay
for the Eggs?
Recall when you last shopped for groceries. Do you re-
member what you purchased, and how much you paid
for each item in your shopping cart? If you bought a carton
of milk or juice, do you know if the price you paid was over
or under the average? When you answered the previous ques-
tion, what did you mean by the average? Was it the normal
price at the specific store, or the median price among several
stores in your neighborhood, or something else? Do you re-
member if the milk was one of the store’s weekly specials?
Do you remember if you redeemed a clipped coupon? Do
you remember if you picked up a new brand of juice because
of a promotional offer? Did you switch from Tropicana to
Odwalla, or from Minute Maid to SunnyD?
If you are like the average shopper, you’ll have difficulty
coming up with these answers. When it comes to remember-
ing prices, we are hopeless.
Businesses have long known—and exploited—our price
amnesia. In the late 1980s, two marketing professors, Peter
Dickson and Alan Sawyer, collaborated with a large super-
market chain to measure just how clueless consumers were
7. 154 NUMBERSENSE
about purchases they made only 30 seconds or fewer ago. Re-
searchers intercepted shoppers immediately after they placed
certain target products—such as coffee, toothpaste, and mar-
garine—into their shopping carts. Almost everyone consented
to answering a few questions when offered a $1 gratuity for
participation. To raise the chance of finding price-conscious
shoppers, part of the study was conducted in late January,
when household budgets were stretched after the winter holi-
days. Did people know how much the stuff in their shopping
carts cost? Were they aware of any special deals that applied to
those items? The results of the 800 or so interviews conducted
at four different branches of the chain were disturbing.
After arriving at the display, the average shopper moved
along within 12 seconds, but the majority could not name
the correct price of the item they just took off the shelves.
The average error was 15 percent of the real price. One out
of five shoppers could not even offer a guess of the price.
Their awareness of special discounts was even poorer. This
supermarket chain heavily advertised specials in newspapers
and on television, using the phrase “Cost Cutter Bonus Buy”
accompanied by a scissors symbol. In addition, the manage-
ment placed bright-yellow labels with the slogan and scissors
right next to the standard black-and-white price labels on the
store shelves. And yet, three out of five had no clue if the item
in their shopping cart was on special or not; estimates of the
price reduction given by those who could suffered an average
error of 47 percent.
The jaw-dropping findings didn’t stop here. The research-
ers learned that people who shopped frequently for an item
were equally as hopeless as the rest. Finally, the professors
performed an aided brand awareness test, similar to the one
mentioned in Chapter 1, on the hunch that some shoppers
could recognize the special price label even if they could not
recall the exact price. In yet another surprise, only 54 percent
of the participants managed to pick out the correct price label
from a choice of three.
8. ECONOMIC DATA 155
This line of research asks deep questions of the founda-
tions of modern economics. In a market economy, prices are
supposed to capture all there is to know about supply and
demand. Producers and consumers are predicted to respond
to these prices. When a half or more of the population are
blatantly inattentive to price tags, we wonder if the economic
profession has gotten this core assumption wrong. Dickson
and Sawyer thought consumers with stronger motivation to
consider prices would perform better in their study, but it
turned out those who shopped at the inner-city store were
even more clueless about the amount they spent on groceries.
Marketing experts have long ago abandoned many economic
principles that are at odds with reality. Behavioral econo-
mists are now tackling this kind of challenge, and their in-
sights may well modernize the foundations of the discipline.
Now, take the side of the store manager. For a gallon of
milk, we require a target price of $3.50 over the next four
weeks. We can, unimaginatively, set a fixed price of $3.50.
But a good portion of our customers love the game of cou-
pons and deals. We can, for example, charge a normal price
of $3.60, and one day a month, run an irresistible bargain
price of $1.50. Alternatively, we can advertise a weekly spe-
cial of $3.00 on a regular price of $3.60. All three pricing
schemes produce an average price of $3.50. Which strategy
would yield the most revenues? The winner depends on how
our customers respond to discounting. That in turn depends
on how they process the prices. Here are a number of pos-
sibilities to consider:
• Availability: People take what comes to mind first. Behav-
ioral psychologists Daniel Kahneman and Amos Tversky,
whom we met in Chapter 5, are champions of this theory.
• Recency: Perception is affected by the most recent price
encountered.
• Frequency: Customers remember the price that appears
most often.
9. 156 NUMBERSENSE
• Average: Customers have a mental image of the average
price. This suggests that they intuitively sense the aver-
age value of a set of numbers.
• Median: Customers have a mental image of the median
price. This requires that they spontaneously discard ex-
treme values.
• Extremes: Perception is swayed by unusually large or
small numbers.
• Losses: Customers pay undue attention to price increases
because they regard price increases as financial losses.
• Numerosity: Customers perceive a better deal when sav-
ings are divided into numerous small installments rather
than applied in total to a single purchase.
There is as yet no definitive research on how consum-
ers perceive prices. It’s not even clear that everyone favors
the same set of heuristics. The decision criteria may vary by
the type of purchase. For durable goods not replaced often,
like stoves and ovens, it’s irrelevant to talk about frequency,
average, median, or numerosity. Big-ticket items and petty
spending surely are not given equal consideration. Perhaps
Kahneman and Tversky’s perspective is the broadest: All the
other criteria pinpoint which price becomes “available.”