Annotated bibliography

www.elsevier.com/locate/dsw

Decision Support Systems

Managing electronic commerce retail transaction costs for

customer value

Alina M. Chircu a,*, Vijay Mahajan

b,1

a Information, Risk and Operations Management Department, McCombs Graduate School of Business, CBA 5.202 B6500,

University of Texas at Austin, USA b Department of Marketing, McCombs Graduate School of Business, University of Texas at Austin, USA

Received 24 June 2004; received in revised form 11 July 2005; accepted 20 July 2005

Available online 24 August 2005

Abstract

We investigate how electronic commerce (EC) retailers, or e-tailers, manage transaction costs and generate customer value.

We integrate information systems and marketing theories in a framework for transaction cost management based on four

contingency factors: channel, customer, product and shopping occasion characteristics. We build the framework using archival

case studies and validate it with customer interviews. We show that trying to minimize the entire cost of retail transactions is

either unsustainable or devalues the customer shopping experience. Instead, looking at transactions as a series of atomic steps

enables e-tailers to better understand and manage what really matters for consumers.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Customer value; Electronic commerce; Internet; E-tailers; Retailers; Transaction costs

1. Introduction

The information economy has created more

informed and demanding consumers than ever before.

Successful retailers are responding to the needs of

these customers by improving the tradeoff between

the customer benefits and transaction costs, thus creat-

ing superior customer value. This, in turn, enables

0167-9236/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.dss.2005.07.011

* Corresponding author. Tel.: +1 512 232 9162; fax: +1 512 471

0587.

E-mail addresses: [email protected]

(A.M. Chircu), [email protected] (V. Mahajan). 1 Tel.: +1 512 471 0840; fax: +1 512 471 4076.

retailers to attract and keep customers, increase sales

and market share, improve profits and firm value

[29,36,50–52,65].

A solution many retailers have explored in their

search for increasing customer value is electronic

commerce (EC) retailing, or e-tailing for short.

From its early days, EC has been promoted as a

way of reducing the monetary, energy, time and psy-

chological transaction costs customers incur when

shopping [2,4,5]. EC technologies allow shoppers to

search for products, receive personalized product

recommendations, evaluate and order products online,

over the Internet. Retailers that use these technologies,

operating either exclusively online or using a mixed

42 (2006) 898–914

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914 899

strategy to sell both online and offline, are generally

known as EC retailers, or e-tailers. As customers find

it more convenient and less costly to shop online

[11,49,60], e-tail sales are forecasted to reach $250

billion in the U.S. and $150 billion in Europe by 2008

[28].

However, during the tumultuous Internet boom and

bust of the recent years, relatively few e-tailers have

been able to create and appropriate enough customer

value to remain in business. We are now witnessing a

second, albeit quieter, EC revolution, with a surprising

40% of the over 200 public EC companies reporting

fourth-quarter profits in 2003 [59] and traditional

retailers profiting from blending online and offline

operations [14]. In this paper we investigate some of

the potential success factors for this revolution —

namely successful management of retail transaction

costs. We start with a review of prior literature, fol-

lowed by our methodology explanation and data ana-

lysis. We then develop and validate a contingency

framework for transaction cost management. We dis-

cuss our framework’s applicability and propose a

sample questionnaire for its implementation, and con-

clude with limitations and future research directions.

2. Transaction costs in the retail transaction chain

2.1. Transaction costs in the retail transaction chain

A retail transaction is an exchange between a

consumer and a retailer in which the two parties

obtain something from each other at a cost to each

[12]. Our focus is on the set of costs incurred by the

customer in each retail transaction, to which we will

refer, as others have done before us [18,20,35,38], as

retail transaction costs. As in this previous research,

our definition draws from transaction cost theory,

which advances the idea that transacting buyers (cus-

tomers) and sellers (retailers) experience costs related

to identifying the appropriate trading partners and

obtaining product information and prices, writing con-

tracts, purchasing, and policing and enforcing con-

tracts [18,20,38]. Because the customer is at the

center of our investigation, we focus on the demand-

side transaction costs that a customer encounters when

interacting with retailers, which capture the efficiency

of the transaction from the customer’s standpoint [20].

Transaction fees, time, effort, convenience, trouble,

and ease of use have been used to describe the trans-

action cost of customers interacting with retailers

[20,35,38]. This definition of retail transaction costs

is also consistent with papers that label the costs

customers incur while shopping as shopping costs

[10], consumer purchase costs [12], transportation

costs [6], buyer search costs [5] or customer objec-

tives related to Internet shopping [32,60]. These trans-

action costs, or shopping costs, include price-type

costs (such as parking fees, installation fees, credit

charges, taxes, travel costs, transaction fees, etc.),

time-type costs (such as travel time, waiting time,

search time, overall shopping time, delivery time,

etc.), and psychological-type costs (such as perceived

ease of use, inconvenience, frustration, annoyance,

anxiety, depression, dissatisfaction, disappointment,

personal hassle, etc., due to the store physical environ-

ment and interactions with salespeople and other cus-

tomers) [12,20,35,38]. Table 1 presents a summary of

the variety of these customer-side (demand-side)

transaction costs definitions (see Table 1).

Transaction costs occur in all steps of a consumer’s

purchase decision process: need recognition, search,

alternative evaluation, purchase, and outcome [23]. To

acquire products, or resources, customers go through a

resource lifecycle that includes several stages, each

with associated costs: establishing and specifying

requirements, identifying the source, ordering, paying

for, acquiring and testing, integrating, updating, moni-

toring and maintaining, and retiring the product [30].

Researchers have proposed that EC transactions can be

described by similar sequences of steps such as brand

search, product search, and purchase [43], forming a

consideration set, choosing a product, and buying the

product [8], need identification, evaluation of pro-

duct alternatives, evaluation of merchant alternatives,

negotiation, actual purchase and delivery, and pro-

duct service and evaluation [39], or pre-purchase

interactions (search, comparison, and negotiation),

purchase interactions (order, payment, and product

receipt), and post-purchase interactions (service and

support) [36].

This purchase process, in general, and each of its

stages, in particular, has associated costs and benefits.

In this paper, our focus is specifically on the transac-

tion costs of the purchase process. Overall transaction

costs consist of comparison, negotiation, payment,

Table 1

Various definitions of costs experienced by customers while shop-

ping

Reference Costs experienced

while shopping

Definition/operationalization

[18] Transaction costs Costs of shopping around for a

product, negotiating a price,

arranging for financing,

waiting for delivery, enforcing

and monitoring the contract

[20] Transaction costs Product price, time efficiency,

perceived ease of use

[35] Transaction costs Commissions, taxes, spread

costs, unobservable market

impact cost, transaction delays

[38] Transaction costs Time spent shopping, effort,

convenience and trouble

[12] Consumer

purchase costs

Parking fees, installation fees,

credit charges, taxes, travel

time, waiting time, search time,

frustration, annoyance, anxiety,

depression, dissatisfaction

[10] Cost of shopping Travel time (distance)

[6] Transportation costs Real cost of travel, opportunity

cost of time, implicit cost of

inconvenience

[5] Buyer search costs Driving cost, telephone calls,

computer fees, magazine

subscriptions, search time

[32] Fundamental

objectives related

to Internet commerce

Tax, shipping, Internet

connection, and travel costs,

time spent shopping, time to

receive product, convenience,

worry, disappointment, regret

[60] Fundamental

objectives related

to Internet commerce

Tax costs, queuing time, time

to select a product, payment

time, convenience, personal

hassle, ease of shopping, time

pressure, effort

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914900

delivery and service cost measures [38]. Perceived

ease of use and time efficiency [20] and the level of

support the customer receives in each phase of the

shopping decision-making process [33] have also

been used as cost measures for retail interactions. In

the context of electronic brokerages, transaction costs

are classified as direct costs (observable commissions

and taxes) and aggregated indirect costs (unobserva-

ble market impact costs due to transacting large num-

bers of shares, spread costs due to differences between

bid and ask prices, and opportunity costs due to delays

in transaction execution) [35]. Transaction costs are

also routinely modeled as travel distance from a cus-

tomer’s home to the retailer’s location, which captures

the cost of travel time or searching for products

[5,6,10]. Shopping costs, both online and offline,

also include taxes and shipping charges, travel, shop-

ping and delivery time, effort, after-sales service costs,

personal hassle, environmental impact, privacy, safety

and shopping enjoyment, and online payment choices

[32,60].

Understanding what transaction costs customers

incur in each step of the transaction process can help

e-tailers better tailor their offerings to attract consu-

mers. To this end, we view a retail transaction as a

sequence of individual transaction steps that form the

retail transaction chain (RTC): store access, search,

evaluation and selection of products that meet the

customer needs, ordering and payment, order fulfill-

ment, post-sale service and returns [23,31,39].

In each step of the retail transaction chain custo-

mers perform channel-specific activities, such as

handling the products and reading labels in the off-

line channel or displaying a product’s information in

a web browser in the online channel. These activities

each generate specific monetary, time, or psycholo-

gical transaction costs. For example, offline access

involves driving and parking, taking public transpor-

tation, or walking, while online access involves turn-

ing on one’s computer, connecting to the Internet,

and navigating to a store website. Offline search

involves walking through the store and asking sales-

people for advice, while online search implies

browsing online product descriptions or running a

search by product. Offline evaluation implies touch-

ing and feeling products, reading labels, asking

salespeople or shopping partners for advice, while

online evaluation consists of evaluating product pic-

tures, virtual tours, text-based descriptions and other

customers’ online reviews. Offline selection consists

of physically placing the chosen product(s) in a

shopping cart or in one’s hand, while online selec-

tion involves placing the chosen product(s) in a

virtual shopping cart through a mouse click. Offline

ordering implies checking out at a cashier station

after waiting in line, if any, while online ordering

implies clicking the check-out button. Offline pay-

ment involves paying by cash, writing a check or

swiping a bank/credit card, while online payment

involves typing in payment (credit or gift card)

information. Offline fulfillment consists of carrying

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914 901

product(s) home while online fulfillment consists of

choosing a delivery method and waiting for pro-

duct(s) to be delivered at home. Offline service

involves obtaining product information offline, in

store or by phone, while online service involves

obtaining this information from the retailer’s web-

page. Finally, the offline returns step involves taking

product(s) back to the store (repeating the activities

from the access and ordering steps) while the online

returns step involves shipping product(s) back to the

e-tailer (in fact, going through offline access and

ordering activities with a shipping provider).

2.2. Transaction costs and customer value in EC

Customer value represents the tradeoff between the

quality, or benefits, the customer receives and the

costs, such as monetary, energy, time and psychic

transaction costs, the customer incurs by evaluating,

obtaining and using a product [36,65]. Enhancing

customer–firm interactions with technology can

reduce transaction costs, improve customers’ pro-

duct acquisition, and create competitive advantage

for the retailer [18,30,56,61]. Firms that focus on

creating customer value are able to differentiate

themselves from competitors by increasing sales,

market share, profits and market value of the firm

[29,42,45,50,51,65].

Retailers can increase customer value by lowering

transaction costs such as fees, time or inconveni-

ence, either by shifting them away from consumers

or providing incentives to offset them [12]. Using

technology to support the customer activities in the

product lifecycle enhances customer service and

creates cost savings for customers [30]. In the con-

text of e-tail, transaction costs related to ease of

access and navigation, shopping time, trust, compe-

tence, flexibility, personalization, and convenience

impact customer value, satisfaction and e-tailer

adoption [7,17,32,58,60,67]. High EC transaction

costs negatively impact intentions to adopt online

shopping [38]. Low EC channel transaction costs

increase satisfaction with the channel, which in

turn affects channel choice [20]. The level of custo-

mer service, the availability of product information,

and the speed of shopping in traditional retail outlets

do not meet customer expectations [15], possibly

enticing customers to chose online shopping instead

because product representation, product selection,

shipping and handling, on-time delivery, and ease

of ordering also impact e-tailer choice [49]. A con-

sumer’s choice of online or offline shopping chan-

nels also depends on the costs and benefits of each

channel for fulfilling the consumer’s economic

goals, quest for self-affirmation, symbolic meaning,

social interaction, and reliance on schema and scripts

for shopping [8].

E-tailers can differentiate themselves from offline

competitors by offering new value-added services

online, while leveraging their physical value chain

strengths [9,16,46,48] by emphasizing unique pro-

ducts or activities, proprietary content, superior pro-

duct knowledge, and strong personal service and

relationships [46], increasing distribution efficiency

to homes, provision of complementary assortments,

personalization, and differentiating on EC store atmo-

sphere and service [2], and making it easier to interact

with the firm by reducing customer transaction costs

[18]. Such services can create a superior shopping

experience, increase customer value and, in turn, gen-

erate a competitive edge [30] and positively affect

firm performance [52].

This suggests that offering low customer transac-

tion costs creates customer value, increases firm per-

formance and contributes to competitive advantage.

But does EC always decrease these transaction costs?

The information efficiency brought about by EC tech-

nologies can lead to lower information search costs,

lower information asymmetry among buyers and sell-

ers, higher cost transparency, and better many-to-

many communication among buyers and sellers in

an online environment [61]. But the expected effects

of increased information efficiency online – a move

towards electronic markets and minimal price disper-

sion for products sold online – have not been observed

in practice yet. Researchers hypothesize this is due to

the fact that online transaction costs are still signifi-

cant [13,19,55]. Another explanation is that even if

specific online transaction costs are lower, as theore-

tically predicted, consumers do not capitalize on this

reduction because they attach a low importance on

those costs or are hindered by other inadvertently

increased costs. In fact, allowing customers to order

online and have products delivered at home may not

always increase customer value, as suggested by the

many failures of e-tailers.

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914902

In this research, we will probe the following nas-

cent research question by focusing on the effect of EC

on the transaction costs of individual RTC steps:

How can e-tailers successfully manage transaction

costs in the retail transaction chain for improved

customer value?

3. Methodology

Our attempt in this research is to answer questions

related to the effect of EC on customer retail transac-

tion costs and effective transaction cost management

— two real-life phenomena appropriate for a case

study research investigation [66]. We use the case

analysis to build a new, testable and empirically

valid conceptual framework [22] which we validate

with consumer interviews. We collect two types of

data for this study: historical case data for conceptual

framework building, and consumer opinions for con-

ceptual framework validation, as described below.

Our study’s validity and reliability are increased

through data and investigator triangulation [21,66].

First, we define the general constructs of interest

(transaction costs in each step of the RTC and custo-

mer value) based on understanding of existing

research results while maintaining a clean theoretical

state [21,22]. Then, we analyze existing e-tail news

article databases (including personal researcher data,

Lexis-Nexis, Business Search Premier) spanning the

period 1996–2004 and identify the population of e-

tailers. We employ theoretical sampling of this popu-

lation [22] to identify five e-tailer polar cases char-

acterized by the following: firms selling physical

products evaluated through both information and sen-

sory attributes, firms selling exclusively online and

firms with offline presence or alliances (to control for

possible offline–online integration advantages), pri-

vately held and public companies with varying levels

of funding for their e-tail operations (to rule out

funding and management practices effects), bankrupt

and surviving e-tailers (to understand differences in

successful and unsuccessful management strategies),

and firms conducting operations in different geogra-

phical regions, both in the U.S. and abroad (to control

for geographical and cultural differences). The result-

ing five cases are Garden.com, Furniture.com, Web-

van, Amazon.com, and Tesco, for which we collect

historical case data [40] from financial reports, press

releases, news, the Hoovers database, web page

archives, Better Business Bureau reports, and Usenet

groups. We use this data to build a case-based con-

ceptual framework through cross–case pattern analy-

sis, overlapping of collection and analysis phases,

checks against similar and conflicting literature, until

we achieve theoretical saturation [22].

Finally, we collect and analyze primary data on

consumer opinions to validate the emerging concep-

tual framework, following a strategy previous research

deems appropriate for the conceptual model building

phase [32]. To understand the nature of the issues

under investigation, we solicit formal and informal,

structured and unstructured, face-to-face and written,

and individual and group feedback from consumers

who were asked to discuss their current and past

online shopping experiences and the advantages and

disadvantages of online and offline shopping. The

sample consists of over 80 U.S. consumers engaged

in full-time graduate studies, part-time studying and

working, or full-time work. While not a random

sample, this pool of respondents provides variation

on age, gender, professional background, interna-

tional experience, technology skills, and online shop-

ping experience. Specifically for external validation,

the subjects report variety in their e-tailer experiences,

incomes, shopping needs, and past and present shop-

ping environment (including geographical location

across the US and in Europe, Asia, and Latin Amer-

ica, access to personal and public transportation, dis-

tance from retailers, etc.).

4. Understanding e-tailer transaction cost

management strategies

As described earlier, our first step was an in-depth

analysis of five retailers selected through theoretical

sampling — Garden.com, Furniture.com, Webvan,

Amazon.com and Tesco. These seemingly unrelated

e-tailers have one thing in common: the promise of

improving the shopping experience by lowering the

costs associated with retail transactions. Not all of

them succeeded in keeping this promise — only Ama-

zon.com and Tesco have survived while Garden.com,

Furniture.com and Webvan have gone bankrupt. A

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914 903

summary of their characteristics is presented in Table 2

and an analysis of their management of transaction

costs is presented in Table 3 (see Tables 2 and 3).

The common patterns we identify across the five

cases suggest that online transactions still have dis-

advantages and are not always preferable over offline

transactions. This finding qualifies the predictions of

transaction cost theory and the associated recent lite-

rature regarding lower transaction costs for consumers

(as well as firms) online. Based on an inspection of

Table 3, the following comments can be made about

the case studies.

Table 2

E-tailer case summaries

Garden.com (1)

Furniture.com (2)

Web

Ownership Public Private Pub

Established 1995 1998 199

Ceased operations 2000 2000 200

Annual sales $15 million $80 million $17

Profit (loss) ($9.9 million) ($46.5 million) ($45

Status Bankrupt Bankrupt Ban

Operations U.S. U.S. U.S

met

Business model Takes online orders

for bulbs and plants

from customers and

submits them to its

suppliers (growers).

Growers ship

products directly to

customers by Federal

Express.

Takes online orders

for furniture from

consumers and

submits them to

manufacturers.

Manufacturers build

products to order and

ship them to

customers through a

special delivery

service.

Tak

for

cons

prod

war

met

hom

truc

Product

characteristics

[2,43]

Low price, frequently

purchased, hobby

goods, tangible/

experience attributes.

High price,

infrequently

purchased goods,

tangible/experience

attributes.

Low

purc

cons

tang

attri

Channel Online Online Onl

References [41,44,27,26] [24,25,53] [63,

Notes: (1)

The Garden.com domain name was purchased in 2001 by W. Atlee Bu

evolution of the original Garden.com company only. (2)

The Furniture.com domain name was purchased in 2001 by a group of

traditional retailer, to launch a new Furniture.com business model in 2002

available at local furniture stores, which manage the delivery, service an

company.

First, even when EC technologies lower some

costs such as access, search, evaluation or ordering,

consumers may not appreciate these reductions. In the

Garden.com case, traditional search and evaluation

costs create shopping enjoyment for hobby garden

purchases and their reduction is not important for

consumers. In the case of Furniture.com, search and

evaluation costs are insignificant again in the consu-

mer’s mind because of the large monetary stake (fur-

niture is a relatively expensive product) and the low

frequency of incurring such costs. Because access

costs for grocery purchases in the U.S. are generally

van Tesco Amazon.com

lic Private Public

6 1996 1995

1 N/A N/A

8 million $420 million $5.26 billion

3 million) $22 million $35 million

(pro-forma)

krupt Surviving, profitable Surviving, profitable

in 2003

. (select

ropolitan areas)

U.K. US, Canada, U.K.,

Germany, Japan,

France

es online orders

groceries from

umers. Delivers

ucts from its own

ehouse in each

ropolitan area at

e using its own

ks.

Takes online orders

for groceries from

consumers. Picks

products from its own

traditional stores and

either keeps them

from in-store pick-up

or delivers them at

home using its own

trucks.

Takes online orders

for media, electronics,

other products. Ships

products from own

warehouses or

submits order to third

parties. Manages

online stores for

others (Target, Toys

bRQ Us).

price, frequently

hased,

umable goods,

ible/experience

butes.

Low price, frequently

purchased,

consumable goods,

tangible/experience

attributes.

Combination

ine Online and offline Online (with offline

partnerships)

62] [47,63,57] [5,54]

rpee and Co, a garden catalog retailer. This analysis focuses on the

former employees, who partnered with Levitz Home Furnishings, a

. The new company allows customers to search online for products

d returns. This analysis is restricted to the original Furniture.com

Table 3

Transaction cost management at five e-tailers

Company Transaction costs management analysis

Garden.com [41,44,27,26] !It reduced transaction costs with low importance for avid gardeners who purchase garden products as a hobby (access, search, ordering, payment, service). These costs contribute to shopping enjoyment [8,31,64] and their

elimination decreases customer value.

!It increased transaction costs with high importance (evaluation, selection, fulfillment). Products purchased – especially plants and bulbs – were high in experience attributes, and channel characteristics (which involve

displaying product information and pictures rather than allowing direct inspection of the actual products)

increased evaluation costs. The product search and evaluation tools available in the online channel created

customer expectations that were not met later, during the fulfillment step, and customer value was lowered.

Fulfillment transaction costs increased because products were delivered directly from suppliers, without

Garden.com control or customer ability to check order status until actual FedEx shipping, sometimes delayed

for months.

!It reduced transaction costs with high importance (returns) — at a high cost for the e-tailer by simply replacing products without asking for actual returns.

Furniture.com [24,25] !It reduced transaction costs with low importance (access, search, ordering, payment). The company assumed that customers find furniture shopping stressful and dislike interacting with salespeople — hassles that the online

channel eliminated. But customers – even if time-strapped or averse to shopping – are willing to incur such

transaction costs because furniture is relatively expensive and purchased infrequently.

!It increased transaction costs with high importance (evaluation, selection, returns, fulfillment). Testing the furniture, feeling the texture or seeing the exact color of upholstery are important for customers — and these are

the exact activities for which the online channel characteristics increase transaction costs significantly. Even with

features such as sending fabric and leather swatches upon request, furniture evaluation is more difficult online.

Furniture.com tried to offset the increase in fulfillment, service and returns transaction costs inherent in an online

channel by offering free delivery, optional extended warranties, and free product exchanges, but the incentives

offered by Furniture.com were not enough to offset these transaction costs.

Webvan [63,62] !It reduced transaction costs with low importance (access, search, payment, service, returns). These costs were already low in major U.S. markets for most consumers due to offline 24 h shopping, ample parking space,

infrequent grocery items returns and consumers who prefer to browse aisles rather than make shopping lists. The

online evaluation and selection of products has relatively the same costs as the offline version — and higher

costs for fresh produce, which is evaluated primarily on experience attributes that cannot be perfectly conveyed

online.

!It increased transaction costs with high importance (evaluation, selection, fulfillment). For most customers, fulfillment costs were generally higher, since products were not delivered on the same day and customers had to

be home for the 30-min delivery window to receive the products. Only for a small proportion of customers —

those unable or too busy to physically go to the store or carry the products back home, such as families with

kids, seniors, or people with disabilities, fulfillment costs were indeed lower. Fulfillment costs were also kept

high by Webvan’s occasional operational problems, which translated into high proportions of late deliveries and

order inaccuracies. Since groceries are rarely returned, the reduction in return costs was also insignificant.

!It reduced transaction costs with high importance (ordering) — but not enough to offset the other increases. Tesco [47,63,57] !It reduced transaction costs with low importance (payment, service, returns), as costs were already low

(payment) or infrequent (service, returns).

!It reduced transaction costs with high importance (access, search, evaluation, selection, fulfillment). The online channel characteristics created more value for Tesco’s customers than in Webvan’s case because different

environmental conditions make offline transaction costs higher in the U.K. Also, it reduced evaluation costs

through online–offline integration. The vast majority of Tesco’s 1 million registered customers still like to

examine fresh produce and learn about new products by browsing local store aisles. As a result, Tesco integrated

its online store with local offline supermarkets, allowing shoppers to buy from their local store they liked and

trusted instead of operating the online store from a central warehouse. This also enabled Tesco to combine the

information from their customersT online and offline purchases, allowing customers to seamlessly reuse past shopping lists and decreasing their online search and selection transaction costs. While operationally more

difficult, the solution turned out to provide additional benefits for Tesco and their customers as well: in-store

order pick-up, quick, reliable delivery from local stores and consistent online and offline regional pricing

strategies.

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914904

Table 3 (continued)

Company Transaction costs management analysis

Amazon.com [3,54] !It reduced transaction costs with low importance (access, payment). Costs were already low for most consumers due to availability of offline stores and payment options.

!It maintained transaction costs with high importance at a similar level (returns). The trip to an offline store was replaced with a trip to a shipper.

!It reduced transaction costs with high importance (search, evaluation, selection, fulfillment). Amazon.com reduced search and evaluation costs for both search attributes (through online search and product information

features) and experiential attributes (through customer product reviews, purchase circles, and product sampling

features such as CD track samples or bLook Inside the BookQ virtual browsing of table of contents and selected pages). Ordering costs were reduced with the b1-Click orderingQ feature, which eliminates tedious tasks such as entering the payment and delivery information. Fulfillment costs for most products have remained high since the

products have to be shipped to a customer’s home, but Amazon.com has been trying to offset these costs though

free shipping promotions and a variety of for-fee shipping options with a large range of delivery times and

annual unlimited rapid shipping subscriptions to match customer delivery delay sensitivity. Amazon.com has

also experimented with keeping fulfillment costs for electronics at the offline level by allowing customers with

an immediate need to buy products from one of their affiliates, such as Circuit City, and pick them up in a local

store.

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914 905

low due to the prevalence of grocery and convenience

stores opened 24 h a day, 7 days a week, and avail-

ability of private transportation options, Webvan’s

attempt to lower these costs even further did not create

value for many consumers.

Second, EC technologies can also raise some RTC

transaction costs that are important in customers’

minds. Delivery costs are important for emotional

involvement products, such as hobby goods, or expen-

sive products, such as furniture. Garden.com and

Furniture.com inadvertently increased these costs by

eliminating the social and sensory aspect of shopping,

and by establishing inadequate online connections to

suppliers that made it impossible to speed up deliv-

eries or at least track purchases. Webvan also

increased transaction costs that matter, such as those

for product evaluation and delivery.

Third, successful companies such as Amazon.com

and Tesco manage a delicate balance between the

transaction cost level and the transaction cost impor-

tance: the levels of transaction costs with high impor-

tance are lowered, while the levels of transaction costs

with low importance for consumers are not affected

much. The reduction of access and fulfillment costs

seems to be valuable for Tesco’s U.K. consumers, for

which these costs are higher to start with. Similarly,

Amazon.com’s customers seem to appreciate the abi-

lity to quickly search for products and evaluate their

quality based on many other customers’ recommenda-

tions and reviews — features not available in a tradi-

tional store. Tesco and Amazon.com also offset the

increase in evaluation costs by offering well-known

products from existing offline grocery stores and

allowing digital sampling of books and CDs, respec-

tively. On the other hand, Garden.com and Furniture.-

com lowered only unimportant transaction costs. They

increased evaluation and fulfillment costs for

unknown, infrequently purchased products for which

quality was important, due to emotional involvement

(hobby) or high prices. Similarly, Webvan lowered

transaction costs that were already low enough for

U.S. consumers and increased important costs, such

as evaluation and fulfillment, for some consumers, by

selling products they could not inspect beforehand and

requiring customers to be at home to accept deliveries.

5. A contingency framework for managing

transaction costs in EC

In case study research, generalization can be

achieved by developing empirically testable, concep-

tual models [21,22]. To attain this goal, we further

integrate the case study findings with existing theory

to develop a contingency framework of e-tail transac-

tion costs.

Our analysis suggests that consumer preferences

for technology-assisted shopping are highly situa-

tional, depending on consumer segments and product

categories. This confirms previous findings that show

consumers shopping for groceries value convenience

and shopping speed, while those shopping for major

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914906

appliances and furniture are more interested in product

information and service, and those shopping for toys

or apparel need privacy and fun of shopping to be

satisfied [15]. Consumer preferences can be explained

in part by understanding the specific affordances of

each channel regarding the shopping experience. For

example, a tactile offline channel experience cannot

be fully replicated online, and thus the lower sensory

affordance of the online channel will matter if the

customer wants to feel the texture of the product [37].

A better understanding of customer’s needs in

online and offline shopping can only be achieved by

jointly evaluating the impacts of channel technology,

product, user and shopping occasion characteristics on

the shopping experience [1]. The previous sections

point out that advanced online channel features that

replicate the offline store experience through recom-

mendations, blook inside the bookQ and 3D virtual tours have helped Amazon.com lower search and

evaluation costs. Because sitting on a couch is not

an experience replicable online, Furniture.com’s eva-

luation costs were increased, pointing out the impact

of product type on transaction cost levels. The type of

user (busy and with large disposable income versus

with more free time but less income) and his environ-

ment (distance from stores, availability of transporta-

tion means, etc.) also affected transaction costs in

Webvan and Tesco’s case. Finally, shopping for a

hobby rather than for a utilitarian need, resulted in

higher online transaction costs for Garden.com.

This suggests four contingency factors impact

transaction costs in the RTC: channel characteristics

(bhow is the transaction performed?Q), customer char- acteristics (bwho performs the transaction?Q), product characteristics (bwhat is being transacted?Q), and shopping occasion characteristics (bwhy is the trans- action performed?Q). These contingency factors can affect the level of individual transaction costs, the

importance of individual transaction costs, or both.

Our case analysis further suggests that the level and

importance of transaction costs affect customer value

and retailer competitive advantage. We summarize

this contingency framework in Fig. 1 (see Fig. 1).

We validate the framework with comments

obtained from consumers, as described in the Metho-

dology section. The results of this process suggest that

the transaction costs are situation-specific and depend

on the characteristics of the online and traditional

channel, customer, product and shopping occasion,

which interact with each other. In the words of one

consumer, bBoth traditional and online shopping have their merits in certain situationsQ. Next, we discuss each of the four contingency factors in detail, provid-

ing examples of supporting theories and consumer

opinions for each factor.

Channel characteristics describe how the retail

transaction steps can be performed, and consist of all

the features available for transacting, such as product

search, side-by-side product comparison, quick check-

out, personalized product recommendations, other cus-

tomers’ ratings and comments about the products, vir-

tual tours, virtual models (in an online channel) or

product displays, salespeople, fitting rooms, and self-

check-out machines (in an offline channel). The online

channel characteristics can lower access costs (by elim-

inating the need to drive to the store), search, evaluation

and selection costs (by offering instant access to rich

product information) and ordering costs (by effortlessly

placing products in shopping cart and allowing instant

check-out with no waiting). As one consumer points

out, the transaction is bquick and easy. I didn’t have to go anywhere to shop, I didn’t have to stand in lines, I

didn’t have any problem finding [a product]Q. It is important to note however that the channel

characteristics will reduce transaction costs only if the

alternatives have higher costs. Accessing a store

online instead of driving is valuable only for those

customers who live in a crowded neighborhood with

limited parking or retailer diversity, or who are located

far away from retailers and have no reliable transpor-

tation means. One consumer points out: bI live in the only place in the metro area without a major book-

seller within 5 miles — so traditional book shopping

is not that convenient although I enjoy itQ. Another consumer who lives in a rural area confesses that it is

inconvenient to bdrive to a store far away from homeQ. Another consumer complains about the difficulty of

finding a parking space when shopping offline,

bespecially on weekendsQ. And a consumer living in a large U.S. city decided not to use online grocery

shopping although it was available, since it was easier

for this consumer to pick up groceries from the corner

store.

The channel characteristics may also have the

unwanted effect of increasing, instead of decreasing,

search and evaluation transaction costs due to lack of

Channel (HOW?)

Customer (WHO?)

Product (WHAT?)

Shopping Occasion (WHY?)

Transaction Cost Contingency

Factors

Individual Transaction Cost Importance

(A, Sr, E, Se, O, P, F, Sv, R) Low High

L o w

Low Customer

Value

No Competitive Advantage

Impact

High Customer

Value

Increased Competitive Advantage

In d

iv id

u a

l T

ra n

s a

c ti

o n

C o

s t

L e v e l

(A ,

S r,

E ,

S e

, O

, P

, F

, S

v, R

)

H i g h

Low Customer

Value

No Competitive Advantage

Impact

Low Customer

Value

Decreased Competitive Advantage

Fig. 1. A contingency model of transaction cost impacts on customer value and e-tailer competitive advantage. Legend: A=access, Sr=search,

E=evaluation, Se=selection, O=ordering, P=payment, F=fulfillment, Sv=service and R=returns.

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914 907

expert human contact, fulfillment costs due to delays

in delivery, and return costs due to having to mail

back products instead of returning them to a local

store. On the one hand, customers remark: bthe large shipping costs, as well as the delay in receiving items

purchased online is a definite downsideQ, and bI some- times prefer the help of a human instead of a

computerQ. On the other hand, others feel that the higher delivery cost is offset by improved search

cost: bJust walk into the typical bookseller looking for a specific book. Unless it is, or it used to be, a New

York Times bestseller, you have to be lucky to find it.

It doesn’t take too many fruitless trips to a local

bookstore during busy weekends to realize that a

few extra dollars shipping might often be worth itQ. Customer characteristics refer to attributes of the

customer, such as income and time availability, and

general attitudes related to shopping. Researchers

have proposed that customers value their time propor-

tionally to the amount of money they could earn by

working instead of performing shopping activities.

Becker [9] posits that consumer non-working time is

intrinsically valuable, as households can combine it

with market goods to produce commodities for con-

sumption, and choose the set of commodities that

maximize their utility function. Dual-income house-

holds interested in consuming a commodity such as

bmeals at homeQ, for example, tend to purchase more expensive goods like frozen pre-cooked meals instead

of less expensive basic food ingredients that require

longer cooking time, since their time is more valuably

spent working than cooking [9]. The cost of time also

affects consumer preferences towards in-store shop-

ping versus delivery services, with consumers choos-

ing delivery services if their cost of shopping time is

high relative to the cost of delivery [9]. As a consumer

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914908

working full-time but with little disposable income

remarks: bThis past Thanksgiving I waited 4 hours in line to save about $20 on some trinkets, but it was just

for the sake of getting something cheaper. If I have to

drive to a store just to find a good deal for a digital

camera, for example, I’ll do it in a minuteQ. The convenience and time savings offered by e-tailers

are appealing to households with little discretionary

time, and especially to dual-income households.

Online shoppers value more the time savings possible

in online shopping than the potential price savings

[11]. Reducing the time spent shopping, and therefore

the transaction costs, by moving access, search, eva-

luation, selection and ordering transaction steps online

will create more value for high-income, time-strapped

consumers. As one consumer remarked, bI don’t have a lot of free time [. . .] so I’m happy to pay a little extra money [when shopping online]Q.

Studies also confirm that time starvation and a

wired lifestyle are a strong predictor of e-tailer shop-

ping behavior [11]. In the words of the customers we

interviewed, bMany of us do not have time to visit stores even if they are open until 9pmQ. Another consumer who did not use online shopping explained

that buying from bricks and mortar stores is easier for

her because she has more disposable time (while

studying full-time), but buying online would make

sense if she had time constraints. As a result, a cus-

tomer will prefer online shopping bwhen time is lim- ited or travel is not preferredQ, or bwhen I am short on time and I cannot go shopping at a retail outletQ. These comments express the time constraints of the broader

Internet population as captured in our sample by

adults referring to their past and current behavior

while working full-time or part-time.

Customers who dislike shopping in offline stores

and interacting with salespeople will also place a

higher importance on reducing their transaction

costs and will find performing the corresponding retail

transaction steps online more attractive. As one con-

sumer remarked, bI really hate menial chores like shoppingQ. bWhen I shop at the mall, more often than not I receive bad service from individuals who

have extremely little knowledge of the products or

services they are sellingQ, explained another EC fan. Another consumer explains why she does not like

offline shopping: bMostly I dislike things related to the salespeople — i.e. when they don’t know the

products, the prices, the product location in the

store, when they are not polite or helpfulQ. Clearly, different customers will value the reduction in trans-

action costs due to online access, search, evaluation,

selection and ordering differently.

Product characteristics include the product price,

type of product purchased (the mix of search and

experience attributes of the product), weight, and the

availability of the product (mass-market vs. niche/

specialty). The importance of all transaction costs is

likely to decrease as the product price increases, since

consumers will be less sensitive to additional costs if

these costs are necessary to make sure their purchase

fits their preferences [12,43]. Our interviewees con-

firmed they prefer e-tailers bespecially for small-ticket items,Q but shop at offline retailers bfor high-end itemsQ that are extremely customized such as bbusiness suits, furniture and antiquesQ.

The importance of reducing transaction costs is

also different for different product types, as defined

by a mixture of search, intangible or informational

attributes (such as color and size) and experience,

tangible or physical attributes (such as fit and fabric

feel) [2,43]. Search attributes can be easily evaluated

online. Product reviews and other buyer’s product

evaluations could make experience attribute evalua-

tion easier online. However, experience attributes

could also be harder to evaluate directly online,

since the ability to touch, see or try on products can

only be partially, if at all, replicated online. bIt is a totally different experience to touch a book, read the

back cover, and even smell itQ, remarked one consu- mer. Others we interviewed also pointed out that the

inability bto feel, touch and try on a productQ is a major deterrent from using e-tailers.

Searching for products online is also better when

consumers lack product quality information. One con-

sumer points out she used Webvan to buy groceries

and trusted them to pick produce better than her.

Product evaluations based on other customers’ experi-

ences bdefinitely add to the whole online shopping experienceQ. As another consumer recalls: bI was haphazardly looking for a self-help book and started

to search by zeroing in on an old favorite, dWhat color is your parachute?’ The dPeople who bought XX also bought ZZ’ feature led me to the selected choice.

After skimming some reviews, I was convinced this

was the perfect book to chooseQ. On the other hand,

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914 909

when consumers have the required product knowl-

edge, online reviews are not valuable: bI believe I could make a better selection myselfQ, commented one consumer.

However, our respondents pointed out that the

current search tools offered by e-tailers are still limited

in their ability to find products if the customer does

not know the exact product description or does not

know what product she wants altogether. bWhen I know specifically what item I want online shopping

is advantageous. But if I was going for a backpacking

trip in Greece and I wanted to buy a few guidebooks I

would prefer to look at the books to determine which

best fit my preferences — which would be difficult to

accomplish onlineQ. Consumers also dislike the lack of detailed online product information (such as pic-

tures, 3D, virtual tours), online search tools that return

a very poor match and are bhard to use if you don’t know exactly what you are looking forQ, and the complexity of online search. For some consumers

buying expensive items such as electronics, this is a

matter of knowledge and trust of the offline retailer,

where they have access to magazines such as Con-

sumer Reports right in the store and to sales associates

who bask the right questions and help with product selectionQ.

The home delivery, which is offered on most

online purchases, is an advantage only if it is hard

to carry the products to one’s home. A former Web-

van customer recalls he liked the delivery service

because he lived on the 4th floor in a building with

no elevator, and the delivery person carried bulky

grocery purchases up the stairs for him. Finally,

transaction costs for accessing retailers and searching

for readily available, mass-market products offline

will be lower than those for niche or specialty pro-

ducts. Lowering access and search costs by transfer-

ring them online will therefore create more customer

value for specialty products than for mass-market

products, since transaction costs for the latter are

already low.

Shopping occasion characteristics describe the

need for making the purchase (personal use vs. gift,

utilitarian vs. hobby purchase, or shopping alone vs.

shopping with friends or family), the frequency of

purchase (frequent versus infrequent), the need for

the product (immediate versus in the future) and the

aggregation of purchases (buying single vs. multiple

products in the same shopping occasion). The shop-

ping occasion defines the amount of hedonic gains

related to psychological benefits (such as enjoyment)

and social benefits (such as prestige and social class

membership) that customers derive from the transac-

tion [8,34]. More hedonic benefits decrease the impor-

tance of transaction costs and increase customer value.

Many of our interviewees pointed out that they blike to go shopping and look around [for products]Q. In fact, the inefficiencies embedded in the search, eva-

luation and selection retail transaction steps can posi-

tively impact the value of the shopping experience

when purchasing gifts, hobby items, or when shop-

ping with friends or family [8,12,43,64]. One consu-

mer remarked that he will never shop online for

groceries because he liked bthe experience of shop- ping with the entire family over the weekendQ, while another added that for him shopping bis a social, family activity we do every weekendQ.

Infrequent shopping occasions translate in lack of

product experience and thus increased evaluation

transaction costs [2,43]. The level and importance of

access and fulfillment costs will be increased for

frequent purchases, but decreased when customers

purchase multiple products during the same shopping

occasion. And the importance of all transaction costs

will be increased if customers have an immediate need

for the product: bWhen I buy a book, I am anxious to start reading it right away. [The e-tailer] makes me

wait and furthermore makes me pay for shipping and

deliveryQ, remarked one consumer. E-tailer transaction costs are increased for gift products with an immediate

need, as another consumer explained: bI only buy books for my children in a traditional store. The

immediate gratification of buying a book binds them

to the book, and they often start reading it in the car

on the way homeQ.

6. Discussion

The complexity of the factors affecting the cost of

each retail transaction step suggests that e-tailers have

to carefully consider their strategic focus: focused

(niche) or mass-market. A focused strategic orienta-

tion requires the e-tailer to identify the specific com-

bination of channel(s), product(s), customer(s), and

shopping occasion(s) for which it can offer value.

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914910

An industry-wide, or mass-market, strategic orienta-

tion can only be achieved if the retailer can respond to

all contingencies involving the retail transaction

chain. In many cases, this requires retailers to provide

both online and offline shopping, and allow customers

to mix and match online and offline retail transaction

steps as needed.

For example, Garden.com, Furniture.com and

Webvan pursued a mass-market strategy. However,

their retail transaction chain design – border mass- market products online and have them delivered at

homeQ – was in fact appealing only in a limited number of specific situations, restricting these firms

to de facto niche strategies. This was not enough to

sustain these companies’ aggressive spending, tar-

geted at building the online infrastructure and adver-

tising for a large consumer audience. The mismatch is

avoided by Tesco, which offers a mix of shopping

options in an environment where online ordering and

home delivery are valuable for most customers. Ama-

zon.com also successfully manages its transaction

costs by offering channel features that match a variety

of product, customer and shopping occasion charac-

teristics, such as the ability to blook insideQ books, view 3D product images, choose delivery times based

on tiered shipping options (and, in some cases,

through in-store product pick-up through alliances

with offline retailers) and see customer-uploaded

reviews and product pictures. By replicating and

even enhancing the essential capabilities for physical

product inspection and rapid delivery capabilities

available in the offline channel makes Amazon.com

attractive to a variety of customer segments.

An analysis of transaction costs using our con-

tingency framework can reveal to any e-tailer exactly

what customer segments are likely to purchase their

products and how often they will do it. This will

help the e-tailer understand if it can attract a suffi-

cient number of customers to stay in business, and

how to minimize its advertising budget by targeting

its most likely buyers. Because our model shows that

product search, online or offline, is not costless due

to a variety of transaction costs, e-tailers can also use

our insights to avoid price competition within a

channel and across channels (a possibility suggested

by others before us as well [13,19,43]). The frame-

work can also help e-tailers understand the impact of

their particular business model inner-workings on

customer value by answering questions such as:

when and what type of customers will wait for

products to be delivered at home (and how long)?

For what customers, products and shopping occa-

sions will it be appropriate to have variable product

quality from different suppliers, if any? What types

of customers, products and shopping occasions are

amenable to higher evaluation costs online? The

framework could thus be the starting point for both

online and offline segmentation and targeting strate-

gies by any new or existing e-tailer.

To this end, we develop a sample consumer ques-

tionnaire that can help companies understand the four

interacting contingencies and their impact on transac-

tion costs (see Table 4). Instead of making assump-

tions, sometimes unrealistic, about EC transaction

costs and the factors affecting them, e-tailers can use

the questionnaire to learn what customers really think

about the combination of factors affecting transaction

costs. Thus, they can surface relevant combinations of

channel, customer, product and shopping occasion

characteristics that they should focus on to create

customer value. The questionnaire can also be used

to further test our framework.

Because the four contingency factors – channel,

customer, product and shopping occasion – interact,

the mean scores across the questionnaire items are not

directly interpretable. Instead, we suggest e-tailers use

the questionnaire primarily as a segmentation tool.

First, the responses to the channel, customer, product

and shopping occasion can be used to understand how

many distinct segments exist in each category. Addi-

tional analyses can reveal if different customer seg-

ments (as defined by customer items) value the EC

technologies available in each transaction step (as

defined by channel items) differently. Similar diffe-

rences could be surfaced in perceptions about products

purchased (as defined by product items) or shopping

occasions (as defined by shopping occasion items).

The results can be used to determine if the current or

planned e-tailer channel, customer, product and shop-

ping occasion combinations can sustain a mass-market

strategy or are more amenable to a niche strategy. The

results can also suggest if and how the e-tailer should

make IT and marketing investments to better match its

strategy. The e-tailer will thus better understand the

tradeoffs among improving its channel features,

expanding its customer base, changing its product

Table 4

Sample e-tailer survey for diagnosing and managing consumer transaction costs

Item Strongly

disagree

Neither agree

nor disagree

Strongly

agree

(Channel) I can obtain costless, instant access to products. 1 2 3 4 5

(Channel) I have access to a significant amount of search (informational) product information. 1 2 3 4 5

(Channel) I have access to a significant amount of sensory (look, feel, smell, fit, etc.) product

information.

1 2 3 4 5

(Channel) I can compare products easily. 1 2 3 4 5

(Channel) I can check-out easily. 1 2 3 4 5

(Channel) I receive purchased products right away. 1 2 3 4 5

(Channel) The return process is hassle-free. 1 2 3 4 5

(Customer) I am frustrated (intimidated/bored/embarrassed) by offline shopping. 1 2 3 4 5

(Customer) I am a time-strapped person. 1 2 3 4 5

(Customer) I have a large disposable income. 1 2 3 4 5

(Product) The products I buy are hard to handle/transport. 1 2 3 4 5

(Product) To evaluate products, I only need descriptive information about them. 1 2 3 4 5

(Product) I buy mostly niche/specialty products. 1 2 3 4 5

(Product) I buy mostly small-ticket items. 1 2 3 4 5

(Shopping occasion) I buy the same product repeatedly. 1 2 3 4 5

(Shopping occasion) I have an immediate need for the products I buy. 1 2 3 4 5

(Shopping occasion) I shop together with family or friends. 1 2 3 4 5

(Shopping occasion) I buy the products available in your store mainly as gifts for others. 1 2 3 4 5

(Shopping occasion) I buy the products available in your store mainly for my hobby. 1 2 3 4 5

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914 911

mix, or conducting targeted advertising for specific

shopping occasions.

7. Conclusions

Our framework proposes a finer lens for analyzing

how retailers can manage customer transaction costs

to create customer value, online and offline. It pro-

vides a richer perspective on individual transaction

costs in the retail transaction chain and shows how

four interacting contingency factors – channel, custo-

mer, product and shopping occasion – affect each one

of the individual costs and contribute to customer

value creation. The takeaway from our framework is

that the online channel may be more appropriate (i.e.

have lower transaction costs) for performing some,

but not all, retail transaction steps for some, but not all

customers, products or shopping situations. If e-tailers

were to look for a decreased overall transaction cost

they would have a practical problem with quantifying

and then minimizing the aggregate transaction cost.

What they have to do instead, in order to create

customer value, is to start with the individual transac-

tion steps and then determine which ones are impor-

tant for customers. This should happen before

attempting to minimize the cost of each step, while

being careful not to destroy any benefits such as

enjoyment and socialization that may be associated

with a seemingly high transaction cost. Purchase data

analysis or customer surveys and questionnaires such

as the one we suggested in Table 4 (see Table 4) can

be used for this purpose. A successful retail strategy

has to provide the right mix of technology-supported

transaction steps for each customer. Our analysis

shows that it is hard, if not impossible, to maximize

customer value by statically decreasing each indivi-

dual transaction cost. Retailers have to allow dynamic

segmentation of customers by providing them the

option to choose among a range of channels and

associated transaction costs in each transaction step.

Our research is exploratory, in that it proposes a

new way of understanding how availability of online

technologies, customer preferences, product features

and shopping occasion characteristics influences the

transaction costs of retail customers. The framework

we propose in this paper needs to be further tested to

obtain additional proof of its applicability. Future

research can also focus on understanding the nature

of transaction costs in more detail. This paper focuses

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914912

only on the impact of transaction costs on customer

value; future studies should also investigate the med-

iating role of customer value on the link between

transaction cost management and firm performance

measures [51] such as market share, profits, and mar-

ket value of the firm.

Acknowledgements

The authors would like to thank the Editor-in-Chief,

three anonymous reviewers, Prabhudev Konana and

Rajashri Srinivasan of the University of Texas at Aus-

tin, and Rajiv Kohli of the College of William and

Mary for the feedback provided on earlier versions of

the paper.

References

[1] R. Agarwal, V. Venkatesh, Assessing a firm’s web presence: a

heuristic evaluation procedure for the measurement of usabil-

ity, Information Systems Research 13 (2) (2002).

[2] J. Alba, J. Lynch, B. Weitz, C. Janiszewski, R. Lutz, A.

Sawyer, S. Wood, Interactive home shopping: consumer, retai-

ler, and manufacturer incentives to participate in electronic

marketplaces, Journal of Marketing 61 (3) (1997).

[3] Amazon.com, Form 10-K for the fiscal year ended December

31, 2003, United States Securities and Exchange Commission

File No. 000-22513 (February 19, 2003).

[4] Y. Bakos, A strategic analysis of electronic marketplaces, MIS

Quarterly 15 (4) (1991).

[5] Y. Bakos, Reducing buyer search costs: implications for elec-

tronic marketplaces, Management Science 43 (12) (1997).

[6] S. Balasubramanian, Mail versus mall: a strategic analysis of

competition between direct marketers and conventional retai-

lers, Marketing Science 17 (3) (1998).

[7] S. Balasubramanian, P. Konana, N.M. Menon, Customer satis-

faction in virtual environments: a study of online investing,

Management Science 49 (7) (2003).

[8] S. Balasubramanian, Sridhar, Raj Raghunathan, Vijay Maha-

jan, Consumers in a multiple channel environment: product

utility, process utility, and channel choice, Journal of Inter-

active Marketing 19 (2) (2005).

[9] G.S. Becker, A theory of the allocation of time, Economic

Journal 75 (299) (1965).

[10] D.R. Bell, T.H. Ho, C.S. Tang, Determining where to shop:

fixed and variable costs of shopping, Journal of Marketing

Research 35 (3) (1998).

[11] S. Bellman, G.L. Lohse, E.J. Johnson, Predictors of online

buying behavior, Communications of the ACM 42 (12) (1999).

[12] W.C. Bender, Consumer purchase-costs — do retailers recog-

nize them? Journal of Retailing 40 (1) (1964).

[13] E. Brynjolfsson, M.D. Smith, Frictionless commerce? A com-

parison of Internet and conventional retailers, Management

Science 46 (4) (2000).

[14] N. Buckley, Internet shopping — the sequel, Financial Times

(2004 (November 29)).

[15] R.A. Burke, Technology and the customer interface: what

consumers want in the physical and virtual store, Journal of

the Academy of Marketing Science 30 (4) (2002).

[16] A.M. Chircu, R.J. Kauffman, Reintermediation strategies in

business-to-business electronic commerce, International Jour-

nal of Electronic Commerce 4 (4) (2000).

[17] A.M. Chircu, G.B. Davis, R.J. Kauffman, Trust, expertise and

e-commerce intermediary adoption, in: H.M. Chung (Ed.),

Proceedings of the Americas Conference on Information Sys-

tems, 2000.

[18] E.K. Clemons, S.O. Kimbrough, Information systems, tele-

communications, and their effects on industrial organization,

in: L. Maggi, R. Zmud, J. Wetherbe (Eds.), Proceedings of the

International Conference on Information Systems, 1986.

[19] E.K. Clemons, I.H. Hann, L.M. Hitt, Price dispersion and

differentiation in online travel: an empirical investigation,

Management Science 48 (4) (2002).

[20] S. Devaraj, M. Fan, R. Kohli, Antecedents of B2C channel

satisfaction and preference: validating e-commerce metrics,

Information Systems Research 13 (3) (2002).

[21] L. Dubé, G. Paré, Rigor in information systems positivist case

research: current practices, trends, and recommendations, MIS

Quarterly 27 (4) (2003).

[22] K.M. Eisenhardt, Building theories from case study research,

Academy of Management Review 14 (4) (1989).

[23] J.F. Engel, D.T. Kollat, R.D. Blackwell, Consumer Behavior,

Holt, Rinehart and Winston, New York, NY, 1978.

[24] M.A. Farmer, G. Sandoval, Furniture.Com Closes Doors,

Lays Off Most of Staff, 2000 (November 6) (News.com,

available at: http://news.com.com/2100-1017-248145.html?

legacy=cnet).

[25] Furniture.com. About us, available at http://www.furniture.

com (2003).

[26] Garden.com. Form 10-K annual report for the fiscal year

ended June 30, 2000, United States Securities and Exchange

Commission File Number 0-26265 (October 13, 2000).

[27] Garden.Com. Gardenescape, The garden watchdog, available

at: http://gardenwatchdog.com/c/2172 (2004).

[28] Going global: from shopping to search, America’s online

giants have their sights set on International expansion, Finan-

cial Times, 2004 (June 13).

[29] S. Gupta, D.R. Lehmann, J.A. Stuart, Valuing customers,

Journal of Marketing Research 41 (1) (2004).

[30] B. Ives, G.P. Learmont, The information system as a compe-

titive weapon, Communications of the ACM 27 (12) (1984).

[31] R. Kalakota, A.B. Whinston, Electronic Commerce: A Man-

ager’s Guide, Addison-Wesley, Reading, MA, 1997.

[32] R.L. Keeney, The value of Internet commerce to the customer,

Management Science 45 (4) (1999).

[33] R. Kohli, S. Devaraj, M.A. Mahmood, Understanding deter-

minants of online consumer satisfaction: a decision process

perspective, Working Paper, University of Notre Dame, 2003.

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914 913

[34] P. Konana, S. Balasubramanian, The social–economic–psy-

chological model of technology adoption and usage: an appli-

cation to online investing, Decision Support Systems 39 (3)

(2005).

[35] P. Konana, N.M. Menon, S. Balasubramanian, The implica-

tions of online investing, Communications of the ACM 43 (1)

(2000).

[36] P. Kotler, Marketing Management. Analysis, Planning, Imple-

mentation and Control, Prentice-Hall, Upple Saddle River, NJ,

1997.

[37] H. Li, T. Daugherty, F. Biocca, Characteristics of virtual

experience in electronic commerce: a protocol analysis, Jour-

nal of Interactive Marketing 15 (3) (2001).

[38] T.P. Liang, J.S. Huang, An empirical study on consumer

acceptance of products in electronic markets: a transaction

cost model, Decision Support Systems 24 (1) (1998).

[39] P. Maes, R.H. Guttman, A.G. Moukas, Agents that buy and

sell, Communications of the ACM 42 (3) (1999).

[40] R.O. Mason, J.L. McKenney, A historical method for MIS

research: steps and assumptions, MIS Quarterly 21 (3) (1997).

[41] No Garden Variety Website, Catalog Age 16 (4) (1999).

[42] A. Parasuraman, Reflections on gaining competitive advantage

through customer value, Journal of the Academy of Marketing

Science 25 (2) (1997).

[43] R.A. Peterson, S. Balasubramanian, B.J. Bronnenberg, Explor-

ing the implications of the Internet for consumer marketing,

Journal of the Academy of Marketing Science 25 (4) (1997).

[44] Plants by Mail — FAQ’s Consumer Commentary Regarding:

Garden.Com by Garden Escape, Rec.Gardens Usenet Group

(accessed Feb. 15, 2001), 2001.

[45] M.E. Porter, Competitive Advantage: Creating and Sustaining

Superior Performance, The Free Press, New York, NY, 1985.

[46] M.E. Porter, Strategy and the Internet, Harvard Business

Review 79 (3) (2001).

[47] Q&A with Tesco.Com’s John Browett, BusinessWeek (Octo-

ber 1, 2001).

[48] J.F. Rayport, J.J. Sviokla, Exploiting the virtual value chain,

Harvard Business Review 73 (6) (1995).

[49] D.J. Reibstein, What attracts customers to online stores, and

what keeps them coming back? Journal of the Academy of

Marketing Science 30 (2) (2002).

[50] R.T. Rust, P.J. Danaher, S. Varki, Using service quality data for

competitive marketing decisions, International Journal of Ser-

vice Industry Management 11 (5) (2000).

[51] R.T. Rust, T. Ambler, G.S. Carpenter, V. Kumar, R.K. Srivas-

tava, Measuring marketing productivity: current knowledge

and future directions, Journal of Marketing 68 (4) (2004).

[52] K.A. Saeed, Y. Hwang, V. Grover, Investigating the impact of

web site value and advertising on firm performance in electro-

nic commerce, International Journal of Electronic Commerce 7

(2) (2002–3).

[53] G. Sandoval, Furniture.Com a Case Study in E-tail Problems,

2000 (July 31) (CNET News.com, available at: http://news.

com.com/2009-1017-243395.html?legacy=cnet).

[54] G. Sandoval, How Lean Can Amazon Get? 2002 (April 19)

(CNET News.com, available at: http://news.com.com/2100-

1017-886784.html).

[55] M.B. Sarkar, B. Butler, C. Steinfeild, Intermediaries and

cybermediaries: a continuing role for mediating players in

the electronic marketplace, Journal of Computer-Mediated

Communication 1 (3) (1995).

[56] V. Sethi, W.R. King, Development of measures to assess the

extent to which an information technology application pro-

vides competitive advantage, Management Science 40 (12)

(1994).

[57] P.B. Seybold, Get inside the lives of your customers, Harvard

Business Review 79 (5) (2001).

[58] D.M. Szymanski, R.T. Hise, E-satisfaction: an initial examina-

tion, Journal of Retailing 76 (3) (2000).

[59] The E-Biz Surprise, BusinessWeek, 2003 (May 12) (available

at: http://www.businessweek.com/magazine/content/03_19/

b3832601.htm).

[60] G. Torkzadeh, G. Dhillon, Measuring factors that influence the

success of Internet commerce, Information Systems Research

13 (2) (2002).

[61] P.R. Varadarajan, M.S. Yadav, Marketing strategy and the

Internet: an organizing framework, Journal of the Academy

of Marketing Science 30 (4) (2002)1984.

[62] WebvanGroup, Form 10-K for the Fiscal Year Ended Decem-

ber 31, 2000, United States Securities and Exchange Commis-

sion File Number 0-27541, 2001 (April 2).

[63] What Webvan could have learned from Tesco, Knowledge at

Wharton, 2001 (available at: http://knowledge.wharton.upenn.

edu).

[64] Y. Wind, V. Mahajan, Convergence Marketing: Strategies for

Reaching the New Hybrid Consumer, Prentice Hall, Upper

Saddle River, NJ, 2001.

[65] R.B. Woodruff, Customer value: the next source for competi-

tive advantage, Journal of the Academy of Marketing Science

25 (2) (1997).

[66] R. Yin, Case Study Research, Sage Publications, Beverly

Hills, CA, 1984.

[67] V.A. Zeithaml, A. Parasuraman, A. Malhotra, A conceptual

framework for understanding e-service quality: implications

for future research and managerial practice, Working Paper,

University of North Carolina at Chapel Hill, 2002.

Alina M. Chircu is an Assistant Professor

in the Information, Risk and Operations

Management Department at the McCombs

School of Business, The University of Texas

at Austin. She holds BS and MS degrees in

Computer Science from the University

bPolitehnicaQ of Bucharest, Romania, and a Ph.D. in Management Information Systems

from the University of Minnesota. Her cur-

rent research interests focus on business

value of information technology, e-com-

merce, e-business and e-retailing strategy, electronic government,

and technology adoption. Her work has been published in Commu-

nications of the ACM, Journal of Management Information Systems,

International Journal of Electronic Commerce, Electronic Markets,

and Electronic Government.

A.M. Chircu, V. Mahajan / Decision Support Systems 42 (2006) 898–914914

Vijay Mahajan holds the John P. Harbin

Centennial Chair in Business in the

McCombs School of Business, The Uni-

versity of Texas at Austin. He received

his BTech at the Indian Institute of Tech-

nology at Kanpur and his MS in Chemical

Engineering and Ph.D. in Management

from The University of Texas at Austin.

His work has been published in Journal

of Marketing Research, Journal of Market-

ing, Marketing Science, Management

Science, and Harvard Business Review.