Assessing impacts of introducing ship‐to‐store service on sales and returns in omnichannel retailing: A data analytics study

• Author: Gregory R. Heim,Michael Ketzenberg,M. Serkan Akturk
• DOI: http://doi.org/10.1016/j.jom.2018.06.004
• Published date: 01 July 2018
• Date: 01 July 2018

Contents lists available at ScienceDirect

Journal of Operations Management

journal homepage: www.elsevier.com/locate/jom

Assessing impacts of introducing ship-to-store service on sales and returns in

omnichannel retailing: A data analytics study

M. Serkan Akturk

a,∗

, Michael Ketzenberg

b

, Gregory R. Heim

c

a

College of Business, Clemson University, 100 Sirrine Hall, Clemson, SC, 29634-1305, United States

b

Mays Business School, Texas A&M University, 4217 TAMU, College Station, TX, 77843-4217, United States

c

Mays Business School, Texas A&M University, 4217 TAMU, College Station, TX, 77843-4217, United States

ARTICLE INFO

Accepted by: T. Browning

Keywords:

Omnichannel retailing

Ship-to-store

Channel integration

Consumer returns

ABSTRACT

Omnichannel retailing features, such as ship-to-store (STS) service, are designed to deliver a seamless shopping

experience for customers. For a retailer, introducing omnichannel capabilities requires major investments to

integrate physical stores and online marketplaces, yet holds a promise of potentially enhancing revenue streams

from both brick-and-mortar (BM) stores and online store channels. We assess the promise of ship-to-store cap-

abilities by analyzing transactional data from a national jewelry retailer to study impacts of introducing ship-to-

store on a retailer's operating performance, in terms of sales and customer returns. Contrary to expectations, the

findings show that online sales decreased after ship-to-store was introduced, although BM store sales increased.

Detailed analysis of the transactional data suggests that, after STS implementation, some customers switched

from the online channel to the brick-and-mortar channel. This switch occurred mainly for high-value purchases.

The customers who actually remained with and fully completed a sale using the ship-to-store service typically

were those that bought low-value items. Our findings also suggest that introducing ship-to-store increased cross-

channel customer returns of online purchases to physical stores. Concurrently, these new ship-to-store returns

generated additional BM store sales. The paper contributes by showing how introducing ship-to-store service can

have different impacts in terms of sales and returns across a retailer's channels.

1. Introduction

We assess impacts of introducing ship-to-store service on retailer

sales and customer returns across multiple retailing channels. With the

advent of technology-enabled shopping alternatives, retailers began

augmenting their retailing channels with many new service processes,

the collection of which has evolved into what today is known as om-

nichannel retailing. The main focus of omnichannel retailing is to offer

consumers a seamless shopping experience, no matter which channel

they use (Rigby, 2011;Brynjolfsson et al., 2013;Bell et al., 2014). With

omnichannel retailing, customers can buy online, buy in stores, or buy

via several other shopping modes (e.g., catalogs, mobile devices).

Among the many models of omnichannel purchasing and order fulfill-

ment, major retailers today offer buy-online and pick-up-in-store,ship-to-

store,ship-from-store, and reserve-online and pick-up-in-store services to

meet customer expectations.

Many omnichannel service processes are designed to draw custo-

mers into physical stores (RIS, 2012;MA, 2014) and thereby increase

store traffic(

Yantra, 2005;Lieb, 2015). Retailers do this via purchase

options such as ship-to-store or buy-online and pick-up-in-store, as well as

via return options such as buy-online-return- to-store (Zhang et al., 2010).

Store traffic is essential to increase sales (Gulati and Garino, 2000;Bell

et al., 2014), either through impulse purchases or through the assis-

tance of store employees (Fisher and Raman, 2010;Mani et al., 2015).

For example, at the national jewelry retailer we study, employee

guidelines and training materials indicate that ship-to-store and buy-

online-return-to-store events are important selling opportunities. For

ship-to-store, sales associates are directed to use the occasion to cross-

sell accessories and attendant items, along with profitable services like

warranties. For returns, the selling prescription for the salesperson is to

convert the return into an exchange or to up-sell to a more expensive

item.

From a customer's perspective, the benefit of omnichannel in-

tegration is an increase in the value proposition offered by retailers

(Gallino and Moreno, 2014;Gao and Su, 2017a) due to lower trans-

action costs, higher service quality, and lower perceived risk

(Herhausen et al., 2015). With ship-to-store service, another customer

benefit is a perceived increase in product variety because retailers can

augment physical store inventory with virtual inventory offered on the

Internet (Radial, 2016).

https://doi.org/10.1016/j.jom.2018.06.004

Received 26 October 2016; Received in revised form 20 June 2018; Accepted 22 June 2018

∗

Corresponding author.

E-mail addresses: makturk@clemson.edu (M. Serkan Akturk), mketzenberg@mays.tamu.edu (M. Ketzenberg), gheim@mays.tamu.edu (G.R. Heim).

Journal of Operations Management 61 (2018) 15–45

Available online 09 July 2018

0272-6963/ © 2018 Elsevier B.V. All rights reserved.

T

The envisioned benefits of omnichannel retailing for retailers and

customers have not been lost on practitioners, as evidenced by the sheer

number of retailers pursuing omnichannel strategies. Even so, retail

executives still worry about introducing effective omnichannel pro-

cesses. To date, little academic research has studied the efficacy of

omnichannel retailing tactics to stimulate demand, drive store traffic,

or enable customer returns, which we address.

While omnichannel retailing provides benefits to customers, and

ostensibly to retailers, implementation by necessity involves the adop-

tion of costly and difficult-to-implement information and material

handling technologies that can generate new operational challenges

(Davis, 2008;Zhang et al., 2010). Omnichannel retailing requires in-

tegrating promotion campaigns, assortment planning for online and

offline channels, inventory systems, and warehouses (Gallino and

Moreno, 2014). It also can require multi-channel order management

systems, integration to third-party partners, and many other internal or

outsourced retailing systems (Perdikaki et al., 2015). Historically, re-

tailers have had a hard enough time accurately tracking their store

inventory in the first place (DeHoratius and Raman, 2008), let alone

having the capability to offer inventory visibility across multiple

channels in a real-time manner. As a case in point, while 60% of re-

tailers in a recent survey claim they have implemented inventory visi-

bility across channels, 80% of them report that their systems need

improvement due to implementation issues (BRP, 2016). Thus, there is

a clear tradeoffbetween the costs and challenges associated with im-

plementing omnichannel retailing and the benefits that may arise from

such systems. Complicating matters is that there is a variety of omni-

channel process alternatives, each having different operational com-

plexities and distinctive value propositions.

This paper focuses on implications of introducing ship-to-store ser-

vice. Although some people may use the terms ship-to-store (STS) and

buy-online and pick-up-in-store (BOPS) interchangeably, the two are in

fact largely different service processes with different fulfillment trade-

offs(

Acimovic and Graves, 2015). In short, BOPS provides customers

with real-time store-level product availability information, lets custo-

mers complete transactions online, and allows customers very soon

thereafter to pick up the items in a store at their convenience (Gao and

Su, 2017a). BOPS reduces shopping transaction costs for customers

since items are picked and packed by store employees prior to customer

pick-up. In contrast, with STS, customers complete a purchase trans-

action online, and then wait for a notification about delivery of pur-

chased items to their local stores, free of charge. With STS, shipping of

items from a central distribution center (DC) generally occurs, even if

the item is already available at the store, as is the case with the retailer

we study. Hence, BOPS uses in-store inventory to fulfill customer de-

mand, while STS uses centralized fulfillment.

In practice, we observe many variations of BOPS and STS processes,

including hybrids of the two, with some involving local store-to-store

inventory transshipment. Table 1 compares eight national retailers in

terms of the BOPS or STS features that they offer to customers. Clearly,

there is no single omnichannel strategy that retailers are pursuing. We

thus are left to wonder why a retailer chooses to offer a variant of such

services. Academic research adds little clarity. We are aware of only

two research works that address BOPS (Gallino and Moreno, 2014;Gao

and Su, 2017a) and only one that addresses STS (Gallino et al., 2017).

Clearly, this gap in the literature, compared against the extensive of-

ferings of these omnichannel services in the marketplace, signals a

significant research opportunity.

In this paper, we empirically investigate the impact of introducing

STS service on a retailer's operations by using a series of quasi experi-

ments. We use the difference-in-difference (DID) econometric metho-

dology to compare the pre and post periods of STS introduction. To do

so, we collected a proprietary data set from a national jewelry retailer

that implemented STS service. As with virtually all fashion items,

jewelry is an experience good, and because of this, there is a significant

risk of an item in this product category getting returned, since the

customer is only able to assess its quality and fit after it is received.

The data set in its entirety spans four years with more than 20

million customer-level purchase and return transactions. The jewelry

retailer operates more than 1000 stores in the U.S. and Canada. Using a

subset of this data set that corresponds to a two-year time window (one

year before and one year after) surrounding the point in time of the STS

introduction, we are able to observe how purchase and return activities

of customers change after the retailer introduced the new STS service

and how these changes affect retailer performance.

Given the nascent state of research, we are positioned to make

several contributions. Building on firm-level transaction cost theory and

consumer utility maximization, we develop research hypotheses re-

garding the STS service. Theoretical reasoning leads us to hypothesize

that both online and brick-and-mortar (BM) channel sales should in-

crease after introduction of STS. Overlapping the findings of Gallino

and Moreno (2014) for BOPS in furniture/housewares retailing, we do

find that BM store sales increase after the jewelry retailer's STS rollout,

while at the same time online sales actually decline post-STS. Moreover,

the increase in BM sales is larger than the decrease in online sales.

Plainly, there is more to the story than a simple channel shift of de-

mand. A key point to note as well is that the theoretical underpinning

for an increase in BM store sales resides with an increase in online sales.

Ostensibly, the new availability of STS should directly stimulate online

demand, and through the process of store pick-ups, generate store

traffic and hence store sales. Yet, we find that online sales decline.

Clearly, with this retailer's introduction of STS, a more nuanced ex-

planation is needed.

In short, after digging into details within customer transaction data,

we find that while customers may be drawn via STS service to make

purchases online, many customers subsequently decide not to wait for

store delivery, and instead opt to go directly to a store and immediately

buy similar merchandise. This activity occurs mainly for high-value

items. What is striking here is that there is no way for customers to

know with certainty that the specific items they are interested in are

even available at the store, since store inventory information is not

made available online. In fact, only roughly half of the items available

via STS are also available in stores, so long as they are not out-of-stock.

We speculate that the research that customers conduct online makes

them comfortable enough to believe that the selection offered in local

stores will satisfy their needs, whether or not the online item is im-

mediately available. In contrast, the online customers who do end up

using STS are customers that mainly purchase low value items, for

which the relative savings in shipping cost afforded by STS is dis-

proportionately greater. This finding contributes by showing how a new

STS service can lead to non-uniform customer actions across retail

channels.

Another facet regarding STS introduction is its effect on product

returns. We find that cross-channel customer returns increase. We also

find that product returns of sales made at BM stores decrease, while

returns for online purchases remain unchanged. Customers who switch

from pure online shopping to STS service via the BM channel have

conducted prior research online, thus they should be more knowl-

edgeable about their purchases compared to regular BM customers. This

Table 1

BOPS vs. STS.

Firms Offers BOPS BOPS lead time Offers STS STS lead time

Walmart Yes 4 h Yes 7-10 business days

Best Buy Yes 45 min Yes 3-7 business days

Lowe's Yes 20min No –

Kohl's Yes 4 h No –

Macy's Yes 4h No –

REI No –Yes 7-10 business days

Michaels No –Yes 5-7 business days

Kirkland's No –Yes 7-10 business days

M. Serkan Akturk et al. Journal of Operations Management 61 (2018) 15–45

16