Traditional and IS-Enabled

Customer Acquisition
on the Internet
Choi J.; Bell D.R.; Lodish L.M.;
Management science v.58 no.4 , pp.754-769 , 2012

20081351 Fahad
20123373 신정연

1
Introduction (Case)
• Why Wal-Mart didn’t have to worry
about the online mall (e.g. Amazon)
5 years ago.
• Today, however, half of Wal-Mart
customers say they’ve shopped at
both merchants.
• Because the discounter’s traditional
customers are getting more tech-
savvy.
• In last fiscal year, Amazon posted
41% revenue growth vs 8% at Wal-
Mart.
• So, Wal-Mart has increased its
investment in its online business.
2
Basic concept
• Customer benefits from using online retailers
include
– Lower prices
– Greater convenience
• Convenience is determined by
– Time distance: Shipping time proximity of an
Internet purchase to a customer
– Travel distance: physical travel distance from the
customer’s location to the nearest offline stores

3
Basic concept
• Target customer density
– Interdependency among target customers
induced through density creates a synergistic
effect
– generates offline & online shopping demand
– amplifies customer’s propensity to communicate
or observe each other’s behavior
→ additional effect on offline and online WOM

4
Spatiotemporal evolution
of Netgrocer.com buyers
in Pennsylvania

5
Basic concept
• What is WOM(Word of mouth)?
– Oral or written form of promotion in which
satisfied customers tell other people how much
they like a business, product, service, or event.
• Advantages of WOM
– Effectiveness of recommendation
– Big effect with low price
– Encourages better products and services
– Strengthen brand image

6
Basic concept
• WOM is
– 7 times as effective as
magazine advertising
– Twice as effective as
radio advertising
– Engendered by “benefit matching”
• Benefit matching: the recipient of a WOM
recommendation experiences a positive fit with the
information conveyed and the product or service
recommended.
7
Data and Measures
• Zip Code-level Cumulative numbers of
new buyers at Childcorp.com
– Childcorp.com is leading internet retailer selling a
large selection of children’s necessities
– ask shoppers from childcorp.com “How did you
hear about our website?”

8
Data and Measures
– Answers are classified into the four mutually
exclusive categories
• Offline WOM: personal referrals from
acquaintances and accidental referrals from
unacquainted people in local regions
• Online WOM: referrals through online message
boards, blogs and online communities
• Online search: paid and organic keyword search
from search engines
• Magazine advertising: ads in an affiliated
magazine targeted at the customer group
9
Data and Measures
– Model zip code-level counts of new buyers
acquired through each of the four processes
– Why zip code-level?
• Practical in terms of data requirements and
managerially useful as many retailers collect sales
information at the zip code level

10
Data and Measures
• Focal variables:
– Target customer density
– Location-based convenience benefits
• Control variables:
– Online price benefit
– Magazine circulation
– High-speed internet access
– Geodemographic characteristics
• Spatial clustering of zip codes

11
< Summary statistics for model covariates > 12
Empirical model
• 𝑦𝑘,𝑧 𝑚 ~𝑃𝑃𝑃𝑃𝑃𝑃𝑃(λ𝑘,𝑧 𝑚 )
• 𝑦𝑘,𝑧 𝑚 : the number of new buyers
acquired by process k in zip code z in
regional cluster m

13
Empirical model
• λ𝑘,𝑧(𝑚) is modeled as a function of
– target customer density
– location-based benefits
– the number of target customers
– a set of control variables that capture observed
heterogeneity
– unobserved baseline by regional cluster
– zip code-level measurement error
log(λk ,z (m ) ) = x k′ ,z (m )βk + ε k ,z (m )
x k′ ,z (m )βk = ϕk ⋅ (Target customer density)z (m )
+ ∆k ⋅ (Location - based benefits)z(m)
+ log(n z (m ) ) + ψ k ⋅ Controls z(m)
+ α k ,0 + α k ,m
14
< Parameter estimates from the model >

15
Empirical findings
• Target customer density
Offline Online Online Magazine Total
Variable
WOM WOM search ads buyers
𝜑 (Target
0.071 0.064 0.034 0.033 0.048
customer density)

– Target customer density has positive and

significant effect on total new buyer acquisitions
and on all four acquisition modes.
– The effect of target customer density is large for
WOM acquisition compared to independent
acquisitions.

16
Empirical findings

– As the firm penetrates locations with a higher target

density, WOM will be the most effective acquisition
mode.
– Conversely, as the firm penetrates into rather sparse
areas of lower target density, online search and
magazine advertising acquisitions will be more
effective.
17
Empirical findings
• Location-based convenience benefits and
Offline vs Online WOM acquisitions
Time Offline Online
Total buyers
distance WOM WOM
1-day 1.189 0.733 0.889
2-day 0.555 0.308 0.431
3-day 0.326 0.218 0.290

– Fast shipping has an positive and significant

effect on online demand.
– Fast shipping is more effective in generating new
buyers through offline WOM than online WOM.
18
Empirical findings
Travel Offline Online
Total buyers
distance WOM WOM
1-day -0.076 -0.044 -0.074
2-day 0.268 0.179 0.230
3-day 0.126 0.073 0.098

– Offline WOM acquisitions are more sensitive to

offline travel distance.
– The greater the expected distance a shopper in a
given location must travel to an offline store, the
greater the online demand.

19
Empirical findings
• Control variables
– The effect of saving on sales tax
• Positive and statistically significant for WOM
buyers and for search buyers
• But, no effect on acquisitions through magazine
advertising
– Magazine circulation
• Increases in buyers via online search and
magazine advertising

20
Empirical findings
• Control variables
– Geodemographic control variables
• Online demand is higher in zip codes with higher
population growth rates, more college educated
and wealthy individuals, etc.

21
Empirical findings
• New managerial insights
– IS-enabled methods of acquisition are
important in the new Internet retail economy,
but traditional methods remain vital in a
complementary manner.

22
23
Conclusion
• Acquisitions in general and WOM acquisitions in
particular benefit from physical proximity among
target customers.
• Location-based benefits enhance offline WOM
acquisitions more than they enhance online
WOM acquisitions.
• Acquisition modes are complementary and gains
from geotargeting are possible.

24
References
• Choi, J., S. K. Hui, D. R. Bell 2010. Spatio-temporal analysis of
imitation behavior across new buyers at online grocery retailer. J.
Marketing Res. 47(1) 75-89.
• Anderson, E. T., N. M. Fong, D. I. Simester, C. E. Tucker. 2010 How
sales taxes affect customer and firm behavior: The role of search
on the Internet. J. Marketing Res. 47(2) 229-239
• http://en.wikipedia.org/wiki/Word-of-mouth_marketing
• http://www.ehow.com/info_8430120_advantages-wordofmouth-
marketing.html
• http://www.businessweek.com/articles/2012-03-29/why-wal-mart-
is-worried-about-amazon

25