Towards a User-centered Ubiquitous Customer Feedback Elicitation

Framework

Enrico Wieck and Jasminko Novak

School of Business Studies, Stralsund University of Applied Sciences, Zur Schwedenschanze 15, Stralsund, Germany

Keywords:

Customer Feedback, Ubiquitous Computing, User-centered Design, Design Science, Reference Process

Model.

Abstract:

This paper suggests a customer-driven approach for regular elicitation of customer feedback for vendors.

Ubiquity and user-centeredness are assumed as central enablers of customer-driven feedback elicitation em-

bedded in customer’s common consumption and/or purchase environment (e.g. point of sale). Following the

design science methodology, a user-centered ubiquitous customer feedback elicitation framework is developed

as an artifact that can support vendors in developing corresponding systems. The framework is evaluated by

producing a concrete system instantiation realized with low-ﬁdelty prototyping and wizard-of-oz techniques.

Evaluation results indicate the suitability of the proposed approach: the framework was applicable for the

development of a ubiquitous customer feedback elicitation system and the resulting low-ﬁdelity prototype was

rated as suitable for ubiquitous and user-centred feedback elicitation.

1 INTRODUCTION

Current Feedback Elicitation. Customer feedback

is important for vendors for a number of reasons. For

example: (1) it can provide information for product

engineering processes and market research, (2) it of-

ten provides the opportunity to identify product faults

so that vendors may be able to ﬁx them and improve

product quality or (3) customer’s loyalty can increase

through interaction by communication to the vendor

(as part of customer-relationship management).

Considering different methods for obtaining cus-

tomer feedback, we can differentiate between vendor-

driven and customer-driven feedback elicitation.

Vendor-driven is deﬁned as feedback elicitation ini-

tiated by the vendor (e.g. surveys, interviews). In this

case, it is the vendor’s task to motivate and incentivize

the customers for providing feedback. Customer-

driven denotes the customer as the initiator of feed-

back elicitation. A vendor is not initially involved in

motivating customers for feedback communication.

We assume that customers are most likely inter-

ested in providing their feedback when actually using

a product or service. However, vendor-driven cus-

tomer feedback elicitation in many cases neither takes

place at the point of sale/use nor at the customer’s

common usage environment, e.g. at home. In addi-

tion, current approaches like interviews or competi-

tions require signiﬁcant expenditure of the vendor in

planning and conducting customer feedback elicita-

tion. Thus, vendor-driven customer feedback elici-

tation often lacks integration into the customer con-

sumption process. This can hamper customer feed-

back communication as it requires additional and ded-

icated effort of the customers not directly related to

the time-based context of their use of the product or

service. We argue that a customer driven feedback

elicitation that is better integrated into the customer

consumption process could help to better fulﬁl ven-

dors’ and customers’ feedback elicitation needs.

Solution Approach and Goals. Based on the follow-

ing assumptions, a customer driven approach is devel-

oped in this paper: (1) Feedback elicitation should be

user-centered, i.e. use customers’ intrinsic motivation

(if existing) to communicate their feedback. These

motivations are different from current feedback elic-

itation, as mainly vendors beneﬁt from feedback so

far. (2) Feedback elicitation shall be ubiquitous, i.e.

integrated into customer’s environment. This spans

customers’ purchase and usage processes. The inte-

gration is assumed to support customer’s motivation

for feedback elicitation.

These assumptions can be supported by related

work. User-centredness, i.e. a customer’s intrinsic

motivation to contribute feedback, can be seen in

269

Wieck E. and Novak J..

Towards a User-centered Ubiquitous Customer Feedback Elicitation Framework.

DOI: 10.5220/0003993302690272

In Proceedings of the 14th International Conference on Enterprise Information Systems (ICEIS-2012), pages 269-272

ISBN: 978-989-8565-12-9

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

online-communities. It is a trust-building mechanism

especially in e-commerce situations, where feedback

depicts a ﬁrst evaluation of the business partner’s

trustworthiness (cf. Dellarocas (2003) on eBay’s

feedback mechanism). Ubiquity can also be seen in

e-commerce situations as feedback is often a part of

the purchase or use processes there (e.g. Skype feed-

back function after completing a call). Different kinds

of intrinsic motivation (such as fun, sense of commu-

nity or altruism and autonomy) have also been identi-

ﬁed as the dominant factor for user-generated content

(Nov 2007, Kuznetsov 2006). Therefore, the goal is

to develop a ubiquitous customer feedback elicitation

framework based on user-centeredness and ubiquity

as main design guidelines.

2 RESEARCH METHODOLOGY

The chosen methodological approach combines the

design science paradigm with a user-centred de-

sign methodology for its operationalization. Design

science as deﬁned in information systems research

is ”fundamentally a problem solving paradigm”

(Hevner et al. 2004) with the goal to ”develop

technology-based solutions to important and relevant

business problems” (ibid). The solutions are arti-

facts (constructs, models, methods and instantiations)

whose construction relies on ”kernel theories that are

applied, tested, modiﬁed and extended through the ex-

perience, creativity, intuition and problem solving ca-

pabilities of the researcher” (ibid). Thus, our goal is

to develop a framework for user-centred, ubiquitous

feedback systems as an artifact that can inform design

and implementation of feedback systems in practice.

The adopted research method in pursuing this goal

is depicted in ﬁgure 1. To identify the framework’s re-

quirements, a literature analysis and a user-centred re-

quirements analysis with scenario-building and focus

group discussions were performed. Two different sce-

narios were discussed: a product-based (books) and a

service-based scenario (university restaurant). Both

provided input to the requirements and to the refer-

ence process model. On the basis of the obtained re-

quirements, a user-centred, ubiquitous feedback elic-

Literature analysis

User-centered

requirements analysis

User-centered ubiquitous customer

feedback elicitation framework (reference process model)

Instantiation: low-fidelity wizard of oz prototype

First field-test evaluation

Figure 1: Research approach.

itation framework was derived. This step included

the consideration of technological constraints and en-

ablers. The derivedframework was then applied to in-

stantiate a concrete feedback system. The goal of the

instantiation was to evaluate two key questions: (1) Is

the framework suitable for the development of con-

crete feedback systems? (2) Does the derived instan-

tiation satisfy customer’s requirements for feedback

elicitation? To address these questions the derived in-

stantiation was implemented and evaluated in form of

a low-ﬁdelity prototype.

3 USER-CENTERED FEEDBACK

ELICITATION FRAMEWORK

To support vendors in developing a ubiquitous cus-

tomer feedback elicitation system, a reference process

model is derivedthat is based on the literature analysis

and user-centered requirements analysis. According

to Fettke and Loos (2005), reference process models

can be used for business engineering since they serve

as a blue-print and can be instantiated for a speciﬁc

use case. For this purpose, the framework includes

the required modules for a ubiquitous customer feed-

back elicitation system and the module relations.

The customer consumption processes as described

in Reisch and Scherhorn (2005) are the basis for

the construction of the process model. They include

pre- and post-purchase processes. Pre-purchase pro-

cesses are (1) reﬂexion of one’s needs, (2) information

searching, retrieval and evaluation and (3) purchase

decision. Post-purchase processes are (4) consump-

tion and usage as well as (5) utilization and disposal.

All requirements derived from literature and require-

ments analysis were mapped to the corresponding

consumption process phase.

The reference process model differentiates be-

tween customer driven, vendor drivenand system pro-

cesses. Customer driven processes contain all inter-

actions triggered by a customer. Vendor driven pro-

cesses are such that do need vendor’s initiation. Sys-

tem processes are those that do not require any cus-

tomer or vendor initiation, but highlight speciﬁc as-

pects, e.g. store or display existing feedback.

According to the reference process model, a cus-

tomer has only two tasks. The ﬁrst is to choose a

product using one or more methods and the second

is to communicate the corresponding feedback dur-

ing or after the consumption/usage phase. All other

processes need to be vendor or system driven. A ven-

dor should provide incentives for feedback elicitation

as part of the system design. It is also the vendors’

responsibility to realize mechanisms for the commu-

ICEIS2012-14thInternationalConferenceonEnterpriseInformationSystems

270

General processes

Pre-purchase processes Post-purchase processes

Communicate incentives and motivate customers

… community

visibility

… others (financial/

customized incentives)

Send receipt confirmation to customer

Need reflexion Information search Consumption/Usage Utilization/Disposal

Identify product

Perform survey

Purchase decision

Realize incentives through ...

… trust-building

rating profiles

… manual

selection

… recomm.

system

… search in

listings

… search for

features

Communicate feedback ...

… as product

rating

… as fault

report

… as problem

solution request

Select product through ...

… automatic

selection

… as product

idea

Display existing feedback

Vendor driven processes

outside consumption process

Key:

Vendor driven processes

inside consumption process

System processesConsumption process phase

Customer driven processes

inside consumption process

Store feedback

Analyse feedback

Figure 2: A ubiquitous customer feedback elicitation reference process model.

nication of incentives to the customer, for example

through the customers’ feedback visible in a commu-

nity or through a rating proﬁle. However, though the

vendor should formulate an appropriate incentive sys-

tem, its design should build on intrinsic motivational

factors (such as fun, community reputation etc. (Nov

2007, Kuznetsov 2006)). In this way, the incentive

system is customer-oriented and provided at design

time as part of the system design. Thus, it’s activation

is actually not vendor- but customer-driven (in con-

trast to other approaches such as surveys etc.).

The feedback system is then responsible for prod-

uct identiﬁcation, displaying existing feedback, send-

ing a feedback receipt conﬁrmation to the customer

as well as for storing and analysing the feedback for

further use. The resulting reference process model is

shown in ﬁgure 2.

4 FIRST EVALUATION

The goal of the evaluation was twofold: (1) to test if

the framework is applicable for instantiation, i.e. if

it is suitable to generate a concrete version of a feed-

back system. By demonstrating this we can prove the

internal validity of the developed framework accord-

ing to the design science method. (2) To test if the

derived concrete instantiation, e.g. a prototype, satis-

ﬁes the criteria of user-centered and ubiquitous cus-

tomer feedback elicitation. By proving these aspects

of the instantiated prototype we can validate whether

the design goals of the instantiation were reached.

Prototype System and Evaluation Design. A

scenario-based procedure was chosen to develop a

concrete instantiation. The instantiation itself was re-

alized as a low-ﬁdelity wizard-of-oz simulated proto-

type. A university restaurant scenario was chosen for

the validation setting since it was one of the scenarios

addressed in the requirements analysis and as it repre-

sents a recurring service that commonly lacks an es-

tablished solution for ubiquitous, customer-centered

regular feedback elicitation. Most often customers

can only use an online feedback form or talk to the

staff that usually has limited time for conversation

only. Furthermore, enabling technical infrastructure

such as AutoID technology RFID may be readily

available if RFID chip cards are used for the payment

process (as increasingly common in university restau-

rants). This makes it a realistic scenario both in terms

of a concrete need and the realistic possibilities for

eventual technical realisation.

The prototypesystem was realized as follows. The

feedback design follows a trafﬁc-light metaphor (red,

yellow, green). A customer meal tray is supposed to

have a built-in RFID chip. The customer’s chosen

meal is registered at the cash desk and stored onto the

meal tray RFID chip there. After consumption, a cus-

tomer evaluatesthe lunch with the meal tray by touch-

ing the feedback light on a separate table (cf. ﬁg-

ure 3). A notebook display was used to send a conﬁr-

mation receipt to the user. This display was controlled

Figure 3: A customer submits feedback using the prototype.

TowardsaUser-centeredUbiquitousCustomerFeedbackElicitationFramework

271

by an assistant who was responsible for displaying

the conﬁrmation. Participants were not told that the

assistant operates the display and thus had the im-

pression of a functioning system (wizard-of-oz simu-

lation). Restaurant staff was briefed for co-operation

in advance for a more realistic simulation. Partici-

pants were asked to perform a complete consumption

process, i.e. choose and pay a meal (sweets), eat and

small talk (as in a typical lunch situation), and return

the meal tray including feedback elicitation.

In total, 14 participants took part in the evaluation,

57% male and 43% female. All were students having

lunch regularly in the university restaurant. Partici-

pants were divided into three groups with each four

to ﬁve participants as smaller groups are more real-

istic and easier to observe. Their impressions from

the test were obtained through Likert-scale question-

naires and a group discussion.

Participants fulﬁlled most tasks without notable

problems. There was little uncertainty with the pro-

totype’s feedback lights as some participants were not

sure about the correct meal tray positioning for touch-

ing the light. In such a case, either other participants

gave assistance or the persons interacted with the pro-

totype, e.g. by trying out other positions, until the

notebook display showed a conﬁrmation.

Results and Lessons Learned. In summary, the re-

sults indicate that a majority of the participants would

like to have the opportunity to contribute their feed-

back daily (71 % completely agree (ca), 29 % partly

agree (pa)), would daily provide their feedback (50 %

ca, 43 % pa) and would like to know the other’s feed-

back (14 % ca, 57 % pa). These results indicate that

the design was successful in building on participant’s

intrinsic motivation. Prototype experience was rated

as fun (43 % ca, 57 % pa), simple to use (57 % ca,

43 % pa) and feedback elicitation was rated practical

(64 % ca, 36 % pa). Immediate receipt conﬁrmation

of the submitted feedback on a screen was also agreed

with as a good idea, but it was noted that other cus-

tomers should not see one’s own rating. This points

to the importance of anonymity and implies need for

further research on this aspect of the framework de-

sign. Participants also favored the prototype system

over traditional feedback methods and would want to

have it implemented in the university restaurant.

In the discussion, participants suggested using a

lamp indicating successful feedback contribution or

an acoustic signal instead of showing the selected

feedback light itself.

Overall, the prototype evaluation indicates that

the derived ubiquitous customer feedback elicitation

framework is suitable for instantiation of a concrete

prototype. The prototype was evaluated as appropri-

ate for feedback elicitation by the participants and was

perceived as satisfying user needs (user-centredness).

5 CONCLUSIONS

We have presented an approach to elicit customer

feedback based on user-centeredness and ubiquity as

principal design guidelines. A framework for user-

centered, ubiquitous customer feedback elicitation

has been proposed as an artifact supporting vendors in

the development of such feedback systems. Through

an instantiation of a concrete prototype and its eval-

uation by a wizard-of-oz simulation in a ﬁeld-test,

the internal validity and design goals of the proposed

framework were validated.

The evaluation represents a ﬁrst test with limi-

tations regarding prototype implementation, partici-

pant representativeness and other inﬂuencing circum-

stances. Thus, the results allow only limited gener-

alization and depict therefore a ﬁrst indicator of the

suitability of the proposed solution. Further evalua-

tions with repeated tests and more participants over a

longer period of time are recommended. In addition,

other use cases should be included for producing dif-

ferent instantiations and validating more comprehen-

sively the different parts of the proposed framework.

Future research should also address the question

of feedback conﬁrmation in more detail, as well as the

integration of mobile devices (e.g. smartphones) and

the integration with feedback in online-communities.

REFERENCES

Dellarocas, C. (2003). The Digitization of Word-of-Mouth:

Promise and Challenges of Online Reputation Sys-

tems. Management Science, 49(10):1407–1424.

Fettke, P. and Loos, P. (2005). Der Beitrag der Referenz-

modellierung zum Business Engineering. HMD -

Praxis der Wirtschaftsinformatik, 241:18–26.

Hevner, A. R., March, S. T., Park, J., and Ram, S. (2004).

Design science in Information Systems Research. MIS

Quarterly, 28(1):75–106.

Kuznetsov, S. (2006). Motivations of contributors to

wikipedia. ACM SIGCAS Computers and Society,

36(2).

Nov, O. (2007). What motivates Wikipedians? Communi-

cations of the ACM, 50(11):60–64.

Reisch, L. A. and Scherhorn, G. (2005). Kauf- und Kon-

sumverhalten. In Frey, D., von Rosenstiel, L., and

Hoyos, C. G., editors, Wirtschaftspsychologie, pages

180–187. Beltz Verlag.

Wieck, E. (2011). Konzept und Architektur eines Feed-

backsystems f¨ur Produkte und Dienstleistungen im In-

ternet der Dinge. Master’s thesis, Stralsund University

of Applied Sciences.

ICEIS2012-14thInternationalConferenceonEnterpriseInformationSystems

272