Process and Challenges in Designing Data Registration via File Import in

an Administrative e-Commerce System

Pedro Victor de Sousa Silva

, Anna Julia Abreu Lima de Souza

and Ingrid Teixeira Monteiro

Federal University of Ceara, Quixada, CE, 63900-000, Brazil

Keywords:

e-Commerce, Import Data, Double Diamond, Usability Test, Design Techniques.

Abstract:

This article presents the full account of a case study within VTEX, a Brazilian e-Commerce company. It re-

ported the design challenge to conceive the functionality for importing records via a spreadsheet ﬁle. Several

techniques of Human-Computer Interaction (HCI) and User Experience (UX) were applied within the Double

Diamond design process, such as CSD matrix, ideation techniques, user stories, interaction diagrams, proto-

types. In this document, we report how each technique was applied and present the results obtained from them.

The results of the prototype usability test with user representatives are also presented. Some lessons learned,

from this case study, were: to keep the design process iterative, exchanging techniques whenever necessary;

to involve team members so that everyone is on the same page regarding the problem and the solution; to test

the solution as soon as possible while still in prototype, even if it is not with end users but with their represen-

tatives.

1 INTRODUCTION

In general, an electronic commerce (e-Commerce)

system allows any transaction using electronic means

Assunc¸

ao (2019). This type of technology aims

to support buying and selling over the internet Loh

(2014). For an e-Commerce system to function prop-

erly, at least for large companies, robust platforms are

necessary to store, process, and exchange informa-

tion at scale Assunc¸

ao (2019). VTEX

is a Brazilian

multinational company specialized in e-Commerce,

offering solutions in the areas of marketplace, om-

nichannel, payments, conversations, and extensions to

streamline integrations and development. These so-

lutions enable its customers to manage their online

stores and tap into new growth opportunities through

the various products and services they provide.

Among the various products offered by VTEX,

there is Sales App. This product is the primary so-

lution for uniﬁed commerce. It allows for the integra-

tion of online and physical sales channels, placing a

greater emphasis on customers by putting them at the

center of the purchasing process

https://orcid.org/0009-0003-2361-2747

https://orcid.org/0009-0003-4346-2037

https://orcid.org/0000-0001-5468-0724

https://vtex.com/us-en/about-us/

https://help.vtex.com/en/tracks/instore-primeiros-

Sales App operates through two fronts: a mobile

application and a web-based administrative system.

The mobile application is designed for salespeople

to carry out the entire sales and purchase process in

physical stores with customers. Salespeople can as-

sist store customers in choosing the best products and

even ﬁnalizing the payment. The web-based admin-

istrative system is aimed at store managers, allowing

them to manage salespeople and physical stores data

through an administrative dashboard. Managers can

also analyze the performance of physical stores and

salespeople in terms of sales metrics.

This article presents the efforts of the design team

that concentrated on enhancing part of the web in-

terface of Sales App. The primary objective was to

simplify some crucial tasks for physical store man-

agers. The proposed solution allows importing sales-

people or stores massive data from a spreadsheet, pro-

viding greater efﬁciency during registrations, signiﬁ-

cantly improving the user experience and the adoption

of the solution. While the act of importing ﬁles is a

common feature found in systems that involve regis-

tration, developing this functionality within the con-

text of Sales App necessitated several design activities

to create a prototype proposal. Therefore, this article

presents a whole case study using the Double Dia-

passos-e-conﬁguracoes–zav76TFEZlAjnyBVL5tRc

390

Silva, P., de Souza, A. and Monteiro, I.

Process and Challenges in Designing Data Registration via File Import in an Administrative e-Commerce System.

DOI: 10.5220/0012687900003690

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 26th International Conference on Enterprise Information Systems (ICEIS 2024) - Volume 2, pages 390-401

ISBN: 978-989-758-692-7; ISSN: 2184-4992

mond design process for this kind of bulk registration

of salespeople and stores in application.

2 BACKGROUND AND RELATED

WORK

Human-Computer Interaction (HCI) is a discipline in-

terested in the design, implementation, and evaluation

of interactive computer systems for human use, along

with the phenomena related to that use (Hewett et al.,

1992).

This ﬁeld deﬁnes certain quality criteria, such as

usability, efﬁciency, communicability, accessibility,

and user experience (Barbosa and Silva, 2010). How-

ever, capturing and including usability requirements

are an often challenge regarding HCI practice (Ogun-

yemi et al., 2019). The works mentioned in this

section seek to encompass the most common types

and challenges of research involving HCI and the e-

commerce industry.

2.1 HCI and Industry

Many researchers are dedicated to investigate prob-

lems and challenges related to the ‘industry/academia

divide’ (Goodman et al., 2011; Melo, 2012), often

exploring the factors involved, proposing solutions

and describing good practices and examples (Good-

man et al., 2011; Gray et al., 2014; Roedl and Stolter-

man, 2013; Stolterman and Pierce, 2012; Thies et al.,

2015).

Bim et al. (2019) recall that the synergy between

university and the job market is one of the objectives

of GrandIHC-BR Challenge 5: Training in HCI and

the Market. In their work, the authors present an ex-

ample of articulation between a Human-Computer In-

teraction Evaluation undergrad course with a demand

from the industry that provides services to a munic-

ipal government. The results of the evaluation done

by students were shared with the company that pro-

vides services to the city hall and the necessary cor-

rections were made. Mesquita and Darin (2021), in

turn, aims to support the training of HCI and Design

professionals through Orbis, a toolkit composed of

15 cards that guide the application of user-centered

techniques commonly used in Academia and Indus-

try, in the planning, design, and prototyping stages.

Other academia/industry integration initiatives are:

Kamaruddin (2018), Nascimento et al. (2019) and

Anirudha (2007).

Another strand of research is dedicated to inves-

tigating the different uses, applications, and effects

of HCI and UX concepts, methods, and techniques

in real industry contexts. In the literature, several

practical reports are presented, especially about the

application of different research methods to evaluate

the UX of different systems. Damian et al. (2022),

for example, present a report on the activities carried

out in evaluating the UX of mobile devices using the

Dogfooding approach, which proposes that company

employees use their own products or services in their

daily lives (Harrison, 2006). The authors therefore

discuss some practices applied by experts in UX eval-

uation, with the support of the Dogfooding approach,

at the Eldorado Research Institute, together with the

Motorola partner.

As done in this article, some authors dedicate

themselves to detail the entire HCI and UX design

process in real industry environments and systems,

describing the steps taken and the artifacts built, such

as the work of de Oliveira et al. (2022), Munim et al.

(2020), Silva and Lotthammer (2019), Lazaris and

Vrechopoulos (2014) e Cul

en and Kriger (2014).

2.2 HCI and e-commerce

The success of an e-commerce website is greatly in-

ﬂuenced by the quality of its design (Nathan and

Yeow, 2011). Therefore, since the ﬁrst e-commerce

systems, HCI and UX researchers have been inves-

tigating how this ﬁeld can contribute to good online

shopping experiences. Research involving HCI in the

context of e-commerce is the most varied, covering

topics such as virtual reality (Wu et al., 2023; Orso

et al., 2023), artiﬁcial intelligence (Liang et al., 2023;

Bawack et al., 2022), chat bots (Hsu et al., 2023), so-

cial selling (Khan and Shaw, 2019) and gender (Cao

et al., 2023).

Mu (2021) studied the non-linear relationship

between product performance and user satisfaction,

demonstrated the competitive advantage of user de-

mand research for the design of e-commerce apps,

and discussed the role and signiﬁcance of user re-

search in the design of e-commerce apps from the

whole product operation level.

Also, scholars have proposed that although usabil-

ity remains crucial for effective Web site design, it

is no longer the primary distinguishing factor (Chu

et al., 2014). A thriving e-commerce Web site should

be capable of instilling trust, captivating and involv-

ing customers, and convincing them to make pur-

chases (Patton and Jøsang, 2004; Chiu et al., 2023;

Chak, 2003; Jones, 2011; Winn and Beck, 2002).

Much of the HCI research on e-commerce is ded-

icated to the evaluation stage (Yang and Chiu, 2023;

Chiu et al., 2023; Pu et al., 2008; Belk et al., 2014;

Djamasbi et al., 2014), but it is also possible to ﬁnd

Process and Challenges in Designing Data Registration via File Import in an Administrative e-Commerce System

391

works dedicated to design activities (Hangrui, 2023;

Monteiro and Gonc¸alves, 2021; de Oliveira et al.,

2022).

Some research more aligned with Sales App’s e-

commerce context, involving interactive systems for

physical stores, are: Lazaris and Vrechopoulos (2014)

and Obermeier and Auinger (2019)

The e-commerce product described in this case

study focused primarily on usability (mainly efﬁ-

ciency) and user experience criteria to shape the entire

design solution. Therefore, the HCI quality criteria

guided us throughout the development of the solution.

Furthermore, to meet these criteria, efforts were made

to incorporate user needs into the solution develop-

ment process, aiming to create a useful functionality.

3 PROJECT CONTEXT AND

MOTIVATION

Before Sales App was made available, the conﬁgura-

tion of data for physical stores and salespeople was a

challenge for store managers because managing mul-

tiple registrations has been time-consuming and labo-

rious. Initially, store managers used a solution with a

spreadsheet-based interface to register physical stores

and salespeople. This solution was problematic be-

cause it lacked ﬁeld validation, was not easy to use,

and required technical knowledge, resulting in a time-

consuming process. These factors led to many users

(store manager) to open support tickets with the com-

pany.

Faced with these factors, the product team had the

opportunity to develop a solution to replace this rudi-

mentary registration method, aiming to provide more

ﬂexibility in the registration of salespeople and physi-

cal stores, making the process more efﬁcient with less

need for assistance and fewer errors. This opportu-

nity led to the development of Sales App, the solu-

tion designed to assist store managers in registering

salespeople and physical stores more efﬁciently with-

out requiring technical knowledge from them.

Despite the signiﬁcant improvement in the regis-

tration process, later on, the team identiﬁed a friction

point in the user experience during the salespeople

and physical stores registration process. When man-

agers needed to register multiple salespeople or phys-

ical stores, the registration process proved to be time

consuming and inefﬁcient, although much easier to

use than the previous solution. Currently, registration

is done store by store or person by person because the

system does not allow for bulk registration. Then, the

product team could further enhance the registration

process in the current solution.

To address this friction, the team embarked on de-

veloping the import functionality. This type of fea-

ture already existed in other products of the company;

however, it was quite technical and most focused on

efﬁciency and robustness, with little attention to use

quality and user experience. On the other hand, the

Sales App design team aimed to develop a function-

ality that prioritized use quality, focusing on the user

experience, while not neglecting robustness and efﬁ-

ciency in terms of internal quality of the system.

4 DESIGN PROCESS

To address the design challenge presented in the

above sections and structure the solution-thinking

process, this case study used the Double Diamond

design process in conjunction with Design Thinking

techniques.

Double Diamond process succinctly outlines the

steps of a designer’s work (Tim Browne, 2023). This

process provides a ﬂexible framework with an empha-

sis on problem understanding, continuous iteration,

multidisciplinary collaboration, and user-centered fo-

cus. The Double Diamond comprises four stages: dis-

cover, deﬁne, develop and delivery.

Design Thinking, on the other hand, represents

the way design is conceptualized and integrated into

each stage of the design process (Ambrose and Har-

ris, 2016). These stages demand designers to have

a project and user-focused mindset Kalbach (2016).

Figure 1 illustrates the techniques carried out at each

stage of our own process and their respective objec-

tives.

The discover stage, following a divergent ap-

proach, aims to fully understand the context of the

product, identify the problem being addressed, dis-

cern user needs, explore the environment, deﬁne

the scope, and pinpoint challenges and opportunities

(Tim Browne, 2023). We have applied several tech-

niques in this phase: Brieﬁng, CSD Matrix, analysis

of internal and external products, competitor analysis,

and user journey.

Brieﬁng Ambrose and Harris (2016) is a docu-

ment that aims to map out the context, objectives,

expected goals, and document the decisions made at

the beginning of the project. CSD Matrix (LiveWork,

2023) is a tool used to map the team’s doubts (D), as-

sumptions (S, from “suposic¸

oes” in Portuguese), and

certainties (C) regarding the problem. The user jour-

ney (Kalbach, 2016) is a visual map depicting the

user’s journey when facing that problem. The user

journey analysis mapped out the phases, stages, ac-

tions, and opportunities for users. The analysis of

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

392

Figure 1: Double Diamond process and the artifacts applied to this project.

similar products (Pazmino, 2015) is a technique used

to explore internal or external products that partially

address the problem, typically those offered by com-

petitors.

The deﬁne stage, following a convergent ap-

proach, aims to reﬁne and specify the problems to be

addressed by the team based on the data collected in

the previous phase. With knowledge of the problem,

the goal is to deﬁne a design proposal. This stage is

a moment of assessment, prioritization and deﬁnition

of all the ideas generated in the previous stage (Tim

Browne, 2023). In this case study, the following ac-

tivities were conducted: to deﬁne requirement, proto-

personas, use scenario and user journey map; and to

carry on an ideation workshop combining three tech-

niques: TRIZ (Alexandre Alcaraz, 2023), Brainwrit-

ing (MJV Team, 2022), and SCAMPER (MJV Team,

2022) to generate design ideas by the collaboration

between the design and engineering teams.

Proto-personas enable the representation of a user

group through a ﬁctional character and help the team

stay focused on user needs. Unlike personas, which

are based in ﬁeld research, proto-personas are based

in desk research and suppositions (Mara, 2020). A

use scenario (Barbosa and Silva, 2010) is a way to

describe a usage context with a speciﬁc system or in

the face of a problem involving an actor. The user

journey is a visual map that seeks to represent all user

touch points within an experience (Kalbach, 2016).

Concerning the creativity techniques applied in the

ideation workshop: TRIZ is a structured method for

creating; brainwriting is a technique in which partic-

ipants express their ideas by writing; SCAMPER is

an acronym to Substitute, Combine, Adapt, Modify,

Propose, Eliminate, Rearrange. It provides a struc-

tured way to help think outside the box and improve

the current knowledge.

In the develop stage, following a divergent ap-

proach, is the moment to address the problems deﬁned

by the team and achieve the ideal system state through

experimentation, diversity of ideas, and collaboration

with other team members (Tim Browne, 2023). In

this stage, the following activities were conducted: in-

formation architecture modeling (Garrett, 2003), in-

teraction modeling with MoLIC (Barbosa and Silva,

2010), error message development, and interface ex-

perimentation.

Regarding the techniques executed, information

architecture modeling (Garrett, 2003) assists design-

ers in planning how information will be organized in

relation to other information, considering hierarchy

and the way information will be presented to the user.

Modelling Language of Interaction as Conversation

(MoLIC) enables the modeling of system interaction

from the perspective of a conversation between the

designer and the end user (Barbosa and Silva, 2010).

In the ﬁnal stage, test and deliver, the design team

deliver the artifacts developed and test the prototypes

created (Tim Browne, 2023). In this stage, the follow-

ing activities were conducted: hand-off, usability test,

and improvements to the prototypes.

Regarding the activities in this stage, hand-off in-

volves creating delivery documentation that details

the interaction ﬂows and interface components devel-

oped. Usability test Barbosa and Silva (2010) is a

method of placing the user in a controlled environ-

Process and Challenges in Designing Data Registration via File Import in an Administrative e-Commerce System

393

ment to obtain qualitative and quantitative data re-

garding their interaction with the interface.

The design process occurred in an iterative and

ﬂexible manner, with no requirement to fully com-

plete one stage before moving on to another. There-

fore, when the team identiﬁed the need to revisit a

stage they had already passed, they did so.

5 DESIGNING REGISTRATION

VIA FILE IMPORT

5.1 Discover: the Phase to Understand

the Users and the Problem

In this phase, a CSD Matrix was created to map out

what the team knew and what still needed to be dis-

covered about the problem. In the initial version,

there were 4 certainties, 3 assumptions, and 10 doubts

identiﬁed. Subsequently, some doubts and assump-

tions were resolved and became certainties, result-

ing in a total of 11 certainties, 2 assumptions, and 8

doubts. For example, one of the doubts that turned

into a certainty was regarding the ﬁle extension to be

used for spreadsheets, such as whether the ideal ex-

tension would be csv, xls, or xlsx. The team decided

that the ﬁle extension allowed in the import feature

would be csv and xlsx. The doubts and assumptions

that arose during the project were addressed through

meetings with the engineering team, which included

a team member with a deeper understanding of the

problems and users.

Then, the user journey analysis Kalbach (2016)

was conducted. For this case study, the journey fo-

cused on users (store managers) registering stores and

salespeople. Through the journey analysis, two op-

portunities were identiﬁed: a) importing stores and

sellers via spreadsheet ﬁle and b) guided conﬁgura-

tion, which provided step-by-step instructions within

the interface.

After gaining a deeper understanding of the prob-

lem and the user journey, an analysis of external and

internal products Pazmino (2015) was performed to

understand the universe of existing solutions that ad-

dress the problem. The analysis involved searching

for products that had import functionality, both inter-

nally and externally, based on products like the Sales

App. To ﬁnd external products, we focused on com-

petitor companies, and for internal products, the Help

Center was explored using keywords such as ‘import’,

‘bulk registration’, ‘spreadsheet’, with the goal of

ﬁnding known solutions with similar functionalities.

Additionally, the import functionality within the com-

pany’s platform was investigated. With the identiﬁed

products, the aim was to understand: 1) the user inter-

action from the beginning of the import journey to the

end; 2) import resources; and 3) the design patterns

used. For the analysis of similar products, four inter-

nal VTEX products and ﬁve external competitor prod-

ucts were examined. These products involved im-

porting and/or exporting data through spreadsheets.

For each of these products, their functionalities were

mapped, along with how users interacted with them.

This analysis revealed that some products had rel-

atively simple import processes, while others offered

a wide range of features, such as product ﬁlters and

the ability to update existing records in the system.

This in-depth analysis allowed us to identify best

practices and relevant features that could be incorpo-

rated into our solution.

5.2 Deﬁne: the Phase to Deﬁne the

Problem and the Proposal Value of

Design

With the problems identiﬁed in the previous stage,

the solution requirements were modeled. These re-

quirements helped transform abstract feature ideas

into tangible entities through user stories (Vazquez

and Sim

oes, 2016). User stories were written using

the card, conversation and acceptance model. This

format was chosen for the following reasons: ease of

maintenance, understanding, modiﬁcation, and mea-

surability. The analysis of external and internal prod-

ucts helped deﬁne the essential requirements for the

import feature. Initially, 9 user stories and 20 busi-

ness rules were deﬁned.

The team realized that they could generate more

feature ideas and user stories to ensure robust func-

tionality that would provide a good user experi-

ence. Therefore, an ideation workshop was conducted

through an online meeting with 10 participants and a

moderator to encourage everyone to contribute their

ideas. The workshop steps are described in Figure 2.

In the ﬁrst stage, the team carried out the activi-

ties listed in Figure 2 to align the team on the prob-

lem and stimulate creative thinking, drawing on TRIZ

concepts. Deﬁning the problem ensures alignment

among all participants. Subsequently, they listed the

problems contradictions, which consisted of conﬂict-

ing or mutually exclusive requirements. Identifying

these contradictions can aid in eliciting requirements

and ideas. For the following activities, the team col-

lectively established what the ideal state of the sys-

tem would be. For instance, one statement of the

ideal state was “In terms of the interface, it should

be simple, sequential, and user-friendly.” After these

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

394

Figure 2: Stages of dynamic with team using TRIZ, SCAMPER and Brainwriting.

statements, the team was encouraged to search for vi-

sual references from images provided on other exist-

ing systems.

In the second stage, as shown in Figure 2, partic-

ipants were invited to collaborate using the SCAM-

PER technique. Each participant had 2 minutes to

contribute to each of the words in the technique. The

session moderator timed the intervals, and, when the

minutes allocated for each word ended, he guided the

participants to move on to the next word. All par-

ticipants went through all the words and contributed

ideas to enhance bulk import. The participants had

contributed 34 ideas of requirements or design guide-

lines for the interface. Following the contributions,

there was a discussion among participants about the

ideas, and they voted on the best ones, for example,

“Import and be able to view and edit spreadsheets in

the interface itself”. The third stage was dedicated to

analyzing the results and modeling the requirements,

carried out solely by the design team. By grouping

ideas by afﬁnity, we deﬁne new requirements that en-

compassed the ideas of all team members.

The ideation workshop enabled the deﬁnition of

requirements and meeting the statements made by the

team. With these ideas, the interface design had pa-

rameters for which paths to follow to meet the ideal

state deﬁned by the team during ideation, such as “To

show the user successes and errors only after the im-

port conclusion”. After the analysis, 34 user stories

were deﬁned, an example of which is “As a store man-

ager, I want to download the system’s CSV template

so that I can ﬁll it out”. However, as a decision on

what to prioritize initially, not all user stories were se-

lected for development; the user stories that were not

prioritized were reserved for a later stage.

Once the requirements were outlined, a proto-

persona and a use scenario were deﬁned for the

project; which inspired the design of the user jour-

ney for the new Sales App registration. The jour-

ney mapped out the phases, actions, thoughts, feel-

ings, ideas, and opportunities. A critical point high-

lighted in the journey was the moment when the user

received error feedback if they had made a mistake in

the spreadsheet ﬁll.

5.3 Develop: the Phase to Address the

Problems Deﬁned by Team and to

Prototype Solutions

Among the team members, the functionality was gen-

erally referred to as ‘bulk registration’, but it was

not clear whether users would understand this term.

Therefore, an analysis was conducted to see how in-

ternal products of the company with import func-

tionality referred to it. In addition to internal prod-

ucts, external products from competitors were also re-

viewed. Among the various options, the functionality

was named ‘Import Sellers’ and ‘Import Stores’.

To organize the information within this function-

ality and make it easier for users to understand, the

information architecture (Garrett, 2003) of the func-

tionality was conceptualized. Using a hierarchical ap-

proach and a top-down method, it began by designing

a major category followed by smaller parts of the in-

formation. Diagrams were used to create the informa-

tion architecture. Initially, a draft was made of how

the information, including titles, input ﬁelds, action

buttons, error messages, would be arranged within the

functionality. However, after discussions among the

design team, it was decided to switch from using the

information architecture diagram to the MoLIC inter-

action diagram (Barbosa and Silva, 2010).

With MoLIC, designers could model the interac-

tion and information of the system as a conversation

between the designer and the user. By using the tool,

designers didn’t think of the solution as interfaces but

as a dialogue between systems and users. This ap-

proach allowed for visualizing the whole user’s inter-

action with the system and organizing the information

without worrying about interface details. MoLIC re-

sulted in two diagrams, one for physical stores and

one for sellers. Each of these diagrams led to 8 scenes,

representing the main conversations between the de-

signer and the user, deﬁning the user’s interaction

Process and Challenges in Designing Data Registration via File Import in an Administrative e-Commerce System

395

Figure 3: Example of a MoLIC interaction diagram.

ﬂow with the system. An example of the MoLIC in-

teraction diagram can be seen in Figure 3 and presents

only a portion of the entire system interaction model-

ing.

With the interaction modeling completed, we de-

veloped success and error messages that provided

users with feedback on the system’s status and guid-

ance to recover from potential errors. One insightful

point was realizing that error messages were the pri-

mary form of assistance users would have during the

import journey, so they needed careful development.

The messages had to be clear, assisting users in

diagnosing possible mistakes they had made, in ac-

cordance with the company’s tone of voice. The mes-

sages were created for ﬁeld validations during import,

ﬁelds in each row of the spreadsheet, and success

messages to guide the user as much as possible re-

garding inconsistencies and the need for corrections.

In terms of error messages, the design team developed

2 success messages, 33 error messages for physical

stores, 16 error messages for sellers, and 7 ﬁeld val-

idation messages. An example of an error message

developed for sellers: “The name must be at least 3

characters long”.

With the foundations in place, it was possible to

proceed with the development of high-ﬁdelity proto-

types, using Figma and following the Company X De-

sign System. With the support of the Design System,

the project’s design team could focus more on solving

how the interface solution would address the issues

identiﬁed in the earlier phases.

During this phase, three different versions were

created by the team to arrive at the ideal version

that would provide users with a smooth and friction-

less experience when importing sellers and physical

stores. With the ﬁrst version of the functionality pro-

totype, it was observed that the layout of elements in

the prototype version might not be the most suitable.

In the second version of the prototype, testing was

conducted for use cases in which multiple error mes-

sages appeared, to assess how the information would

appear in the modal component and whether it would

overload the user interface. The second version, after

debate with the team, proved to be less suitable as it

required more white space to assist users in process-

ing information more easily and without overwhelm-

ing them. As can be seen in Figure 4, in the second

version, a modal element was used as a design solu-

tion, but after discussions and test results, the team

decided to use a full-screen interface (with its own

route). The third version of the prototype incorpo-

rated more white space to avoid overwhelming users

with error messages and featured a full-screen inter-

face, as shown in Figure 4.

5.4 Test and Deliver: the Phase to

Validate Prototypes and Ideas

While the development team was working on the im-

plementation, the design team planned a usability test

to assess and validate the developed solution with user

representatives. The usability test followed the steps

outlined in Figure 5.

The ﬁrst step involved presenting information

about the test and collecting consent forms. It was

made clear that participants had the autonomy to in-

terrupt the evaluation at any time. Additionally, per-

mission was sought to access data records, publish

evaluation results, and inform about the assurance of

anonymized data. A pre-test interview with partici-

pants was conducted to gain a deeper understanding

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

396

Figure 4: Some versions of the Figma prototypes.

Figure 5: Steps of Usability Test.

of their experience with the product, their role in the

company, and their proximity to users. The modera-

tor provided context for participants to imagine them-

selves in a scenario where the manager of a speciﬁc

franchise store did not have enough time to complete

necessary records in the solution. In this case, the

session participant would be responsible for record-

ing physical store and salesperson data. A post-test

interview was conducted to gather user opinions on

the prototype that had just been tested.

For the usability test, user representatives were in-

volved. Lazar et al. (2017) and Salminen et al. (2022)

argue that it is not ideal to involve user representa-

tives as a source of information for end users during

the design process. However, in cases where access

to end users is limited, involving user representatives

can be an alternative. User representatives can act as

information providers about end users, describing po-

tential needs and usage contexts. Thus, the design

team planned to conduct the usability test with com-

pany employees who were close to actual users and

had a solid understanding of the user journey within

the solution, especially regarding pain points.

The decision to involve user representatives was

made because the end users of the solution were not

accessible due to scheduling conﬂicts. Test partici-

pants, acting as user representatives, suggested im-

provements that could assist store managers in their

routines. The test was conducted with 3 representa-

tives who were frequently in contact with end users.

The participants proﬁles can be consulted in Figure 6.

P1 had a better understanding of the commercial as-

pect, and P2 and P3 had a deeper technical knowledge

of the application.

The usability test aimed to verify their understand-

ing of the import solution and collect feedback on

how the solution could help store managers. One of

the key points emphasized by the design team was

that the solution was for store managers. The role

of user representatives was to provide improvement

points based on the issues and challenges reported by

end users (store managers) to these representatives.

The usability test consisted of 5 tasks, namely:

T1: Registering the physical store; T2: Registering 8

new salespersons; T2.1: Using the function to register

multiple salespersons; T3: Downloading the available

template spreadsheet and opening the ﬁle; T3.1: Ob-

serving how the spreadsheet functions; T3.2: Filling

in the remaining data on the import screen and im-

porting; T4: Correcting the described errors; T5: Re-

submitting the ﬁle to the system to register all sales-

persons and ensuring there are no errors in the ﬁle.

On average, participants took 6 minutes and 43

seconds to complete all tasks. Regarding success rate,

tasks T2, T4, and T5 resulted in 100% success with-

out assistance; tasks T1 and T3 had some instances of

success with assistance; and only P3 had partial suc-

cess without assistance while performing T3. In terms

of other established metrics, there were only 3 errors

and 3 requests for assistance in total. 7 provides de-

tailed results of the usability test for each participant

and task.

At the end of the test, participants responded to a

post-test interview. When asked about their impres-

sions of the interfaces they had just tested, they men-

tioned that they were clean, intuitive, and had clear

error messages. Participants stated that they did not

encountered any difﬁculties in the import journey and

did not observe any negative aspects related to this

functionality. They emphasized that this new feature

would be a positive point, as it would help streamline

the manual registration process. When asked about

potential improvements in the solution, participants

highlighted the need for some help documentation for

end users to consult the requirements and precondi-

tions of each ﬁeld. Based on these improvement sug-

Process and Challenges in Designing Data Registration via File Import in an Administrative e-Commerce System

397

Figure 6: Proﬁles of representatives users.

Figure 7: Results of the usability test for each participant and task.

gestions, the design team worked on enhancements to

the prototypes, such as adding a help pop-up and doc-

umentation with prerequisites for each ﬁeld so that

end users could refer to it when in doubt about the

spreadsheet ﬁelds.

6 DISCUSSION AND FINAL

REMARKS

Designing to create a good user experience can be

a complex challenge. While the import functional-

ity may not be complex in terms of interface design,

when developed with a focus on providing a clear

and consistent system status, a signiﬁcant challenge

is added. Initially, for a system to provide clear feed-

back, it was essential to allocate time to develop error

and success messages that offer sufﬁcient assistance

to help the user recover from errors. It was also cru-

cial to align ideas between development and design

teams, as well as meticulously map the functional-

ity’s goals to meet the company’s needs. In addition

to these issues, it was important to dedicate time to

think and ﬁnd alternatives when there is no direct ac-

cess to end users.

The design process and the HCI/UX adopted tech-

niques were fundamental to designing for a good user

experience. The discovery phase helped understand

the state of the problem and identify the ideal points

to address. The deﬁning stage helped engage the

team in ideation workshop and to understand different

viewpoints on the problem, allowing the collection

of important insights that inﬂuenced solution model-

ing and requirement deﬁnition. In the development

stage, MoLIC allowed the team to design the inter-

action. Not focusing on the elements of the interface

was initially challenging, but dedicating time to de-

velop error messages was crucial, as these messages

would be the primary way to help users during the im-

port journey in recovering from possible errors. De-

veloping high-ﬁdelity prototypes using the company’s

design system was important to validate ideas and al-

lowed for changes without signiﬁcant costs. In the

ﬁnal stage, usability testing allowed designers to un-

derstand certain aspects related to the distribution of

elements in the interface and user interaction with the

solution. This, in turn, inﬂuenced interface improve-

ments.

The lessons from this case study address the chal-

lenges of the design process, particularly during the

discovery and testing stages and the involvement of

solution users in these stages. Design processes, es-

pecially the Double Diamond used in this case, em-

phasize the importance of directly involving end users

to understand their pains and provide a variety of de-

sign techniques to be used during the stages, always

keeping solution users in focus and designing for a

good user-centered experience. However, this case

study found limitations in accessing end users at that

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

398

time, meaning that even with design team efforts, it

was not possible to directly involve end users in de-

sign research activities, as emphasized by design pro-

cesses. Thus, one of the design team’s questions was:

Who should we involve during the design process to

discover the needs of solution users? In addition to

this question, the team reﬂected on: What alternative

design techniques could be used due to limitations in

accessing users?

To address this limitation and still develop a prod-

uct that took into consideration end users, user repre-

sentatives were involved during the research and eval-

uation stages. They collaborated by providing infor-

mation about end users, the context of solution use,

routine activities, and points of need. Involving user

representatives was a challenge because some tech-

niques had to be adapted. For example, during usabil-

ity testing, it was necessary to consistently emphasize

that the solution was designed for store managers. We

always asked user representatives to share what they

thought would help end users.

This case study is far from questioning the de-

sign process and its effectiveness but aims to reﬂect

on alternatives when it is not possible to follow a lin-

ear design process and use data collection techniques

with end users, as commonly presented. Thus, we

emphasize a ﬁrst lesson: the design process may not

be linear, and some design techniques may not be the

most suitable for the project, depending on the con-

text. It is necessary to always evaluate the reasons

behind those techniques and collaborate with people

outside the design team to obtain different views of

the artifacts, to avoid wasting unnecessary energy on

something that will not work properly and can be set

aside when perceived that they will not bring much

value to the project.

Another lesson learned by this case study is the

possibility of learning about interface patterns, inter-

actions, and mental models of end users through the

analysis of internal and external products related to

the solution being developed. Since it was not possi-

ble to talk to users about their needs, this case study

carefully observed the interaction patterns of compet-

ing and similar platforms and raised a range of re-

quirements. With the requirements, the team could

observe technical feasibility and how much each re-

quirement would add value to the product and develop

it. It is worth noting that this analysis does not replace

moments of validation and research with real users.

Finally, another lesson learned is considering the

kind of involvement the representatives have with real

users. During this case study, it was crucial to recruit

user representatives who had some level of knowl-

edge about end users to capture valuable information

about end users. Another aspect that this work em-

phasized during the discovery and evaluation stages

with user representatives was to present usage scenar-

ios in which user representatives imagined themselves

taking on the role of real users so that they always had

in mind who the product was intended for. Thus, dur-

ing the evaluation stages, user representatives always

brought information about end users, such as ”Store

managers always have difﬁculty registering multiple

stores” and ”Maybe presenting ﬁeld validation rules

can help store managers.” Involving people close to

end users does not replace the participation of end

users in some phase of the design process.

Based on the lessons learned, some reﬂections

arise after this case study. Are there ideal moments

for user representatives to collaborate during the de-

sign process? Can a product developed addressing

user representatives to talk about the needs, pains, or

activities of end users of the solution be considered

a product that actually addressed the pains and needs

of users? What other care in the execution of design

techniques and the design process would be possible

to overcome the challenges of involving this type of

user when there are no end users of the solution avail-

able?

This case study presented, in a detailed and in-

depth manner, how the systematic use of HCI and UX

tools and artifacts allows the design of solutions fo-

cused on quality of use. Therefore, the main contri-

bution of this article is the view of the use, advan-

tages, disadvantages of the techniques applied in a

real industry case and an alternative scenario of so-

lution research and evaluation, which involved user

representatives during the design process. It is worth

noting that involving user representatives is not ideal,

but when there are limitations, it can be an alterna-

tive. The points addressed in this work can inspire

students and professionals in the ﬁeld and contribute

to the professionalization of this type of approach.

For future work, with regard to Sales App, there is

the opportunity to explore individual registration (not

involving importing ﬁles), making it sequential and

allowing greater ﬂexibility to users during registra-

tion, as established as the ideal state of the application

during the ideation process. There is also the intention

to carry out a new usability test with end users (store

managers) instead of user representatives. In addition

to this research, carry out a comparative analysis of

case studies that use the same tools discussed in this

article, to discover differences and moments in which

the tools are used.

Process and Challenges in Designing Data Registration via File Import in an Administrative e-Commerce System

399

ACKNOWLEDGEMENTS

The authors would like to thank the support of VTEX

Lab, the VTEX company’s partnership program with

universities in Brazil that seeks to leverage the pro-

fessional training of students through challenges on a

global scale. The authors also thank the participants

of the usability test.

REFERENCES

Alexandre Alcaraz (2023). Triz – a ferramenta

para soluc¸

ao criativa de problemas e inovac¸

ao.

https://www.vilage.com.br/blog/triz-a-ferramenta-

para-solucao-criativa-de-problemas-e-inovacao/.

Ambrose, G. and Harris, P. (2016). Design Thinking:

Colec¸

ao Design B

asico. Design B

asico. Bookman Ed-

itora.

Anirudha, J. (2007). HCI and SE – The Cultures of the

Professions, volume 4559, pages 108–112.

Assunc¸

ao, Wagner S.; Fagundes, P. F. R. A. S. P. C. (2019).

Com

ercio Eletr

onico. Grupo A.

Barbosa, S. and Silva, B. (2010). Interac¸

ao Humano-

Computador. Elsevier Brasil.

Bawack, R. E., Wamba, S. F., Carillo, K. D. A., and Akter,

S. (2022). Artiﬁcial intelligence in e-commerce: a

bibliometric study and literature review. Electronic

markets, 32(1):297–338.

Belk, M., Germanakos, P., Asimakopoulos, S., Andreou, P.,

Mourlas, C., Spanoudis, G., and Samaras, G. (2014).

An Individual Differences Approach in Adaptive Wav-

ing of User Checkout Process in Retail eCommerce,

volume 8527, pages 451–460.

Bim, S. A., Oliveira, O., and Junior, G. S. (2019). Constru-

indo pontes entre academia, ind

ustria e governo: uma

experi

encia de sucesso no ensino de avaliac¸

ao em ihc.

pages 76–81. Sociedade Brasileira de Computac¸

ao -

SBC.

Cao, Y., Zhai, W., and Zhai, W. (2023). The impact of

gender and visual presentation of advertising on user

experience in mobile shopping apps. In Nah, F. and

Siau, K., editors, HCI in Business, Government and

Organizations, pages 353–364, Cham. Springer Na-

ture Switzerland.

Chak, A. (2003). Submit now: Designing persuasive Web

sites. New Riders.

Chiu, Y.-J., Chen, H.-H., and Chen, C.-Y. (2023). Key suc-

cessful factors of e-commerce platform operations. In

Nah, F. and Siau, K., editors, HCI in Business, Gov-

ernment and Organizations, pages 150–169, Cham.

Springer Nature Switzerland.

Chu, H.-L., Deng, Y.-S., and Chuang, M.-C. (2014). Per-

suasive Web Design in e-Commerce, volume 8527,

pages 482–491.

Cul

en, A. L. and Kriger, M. (2014). Creating Competitive

Advantage in IT-Intensive Organizations: A Design

Thinking Perspective, volume 8527, pages 492–503.

Damian, A. L., Pereira, M., Luz, B., and Leite, J. (2022).

Avaliac¸

ao da experi

encia de usu

arios por meio da

abordagem dogfooding. pages 17–23. Sociedade

Brasileira de Computac¸

ao.

de Oliveira, G. M., Carvalho, A. N., Lamego, B. S., Mon-

teiro, I. T., Gonc¸alves, E. J. T., and Basilio, A. F.

(2022). Reporting the application of user experience

tools and proxy users in an industrial process based

on double diamond. In Soares, M. M., Rosenzweig,

E., and Marcus, A., editors, Design, User Experience,

and Usability: UX Research, Design, and Assessment,

pages 57–74, Cham. Springer International Publish-

ing.

Djamasbi, S., McAuliffe, D., Gomez, W., Kardzhaliyski,

G., Liu, W., and Oglesby, F. (2014). Designing for

Success: Creating Business Value with Mobile User

Experience (UX), volume 8527, pages 299–306.

Garrett, J. (2003). The Elements of User Experience: User-

centered Design for the Web. Voices (New Riders).

New Riders.

Goodman, E., Stolterman, E., and Wakkary, R. (2011). Un-

derstanding interaction design practices. pages 1061–

1070. ACM.

Gray, C. M., Stolterman, E., and Siegel, M. A. (2014).

Reprioritizing the relationship between hci research

and practice. pages 725–734. ACM.

Hangrui, C. (2023). Research on the optimization design

of mobile vending cart service process. In Nah, F.

and Siau, K., editors, HCI in Business, Government

and Organizations, pages 202–218, Cham. Springer

Nature Switzerland.

Harrison, W. (2006). Eating your own dog food. IEEE

Software, 23(3):5–7.

Hewett, T. T., Baecker, R., Card, S., Carey, T., Gasen, J.,

Mantei, M., Perlman, G., Strong, G., and Verplank,

W. (1992). Acm sigchi curricula for human-computer

interaction. Technical report, New York, NY, USA.

Hsu, P.-F., Nguyen, T. K., Wang, C.-Y., and Huang, P.-

J. (2023). Chatbot commerce—how contextual fac-

tors affect chatbot effectiveness. Electronic Markets,

33(1):1–22.

Jones, C. (2011). Clout: The role of content in per-

suasive experience. In Marcus, A., editor, Design,

User Experience, and Usability. Theory, Methods,

Tools and Practice, pages 582–587, Berlin, Heidel-

berg. Springer Berlin Heidelberg.

Kalbach, J. (2016). Mapping Experiences: A Com-

plete Guide to Creating Value Through Journeys,

Blueprints, and Diagrams. O’Reilly Media.

Kamaruddin, A. (2018). An industry involvement to en-

hance the hci small-scale project learning experience.

integrac¸

ao da industria em projetos dos alunos.

Khan, Z. M. H. and Shaw, N. (2019). Adding ‘social’ to

commerce to inﬂuence purchasing behaviour. In Nah,

F. F.-H. and Siau, K., editors, HCI in Business, Gov-

ernment and Organizations. eCommerce and Con-

sumer Behavior, pages 257–273, Cham. Springer In-

ternational Publishing.

Lazar, J., Feng, J., and Hochheiser, H. (2017). Research

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

400

Methods in Human-Computer Interaction. Elsevier

Science.

Lazaris, C. and Vrechopoulos, A. (2014). Human-

Computer vs. Consumer-Store Interaction in a Multi-

channel Retail Environment: Some Multidisciplinary

Research Directions, volume 8527, pages 339–349.

Liang, S., Alimu, N., Si, H., Li, H., and Mi, C. (2023).

Inﬂuence of artiﬁcial intelligence recommendation on

consumers’ purchase intention under the information

cocoon effect. In Nah, F. and Siau, K., editors, HCI in

Business, Government and Organizations, pages 249–

259, Cham. Springer Nature Switzerland.

LiveWork (2023). Livework studio - design

thinking, innovation, andc service design.

https://liveworkstudio.com.br/.

Loh, S. (2014). 31 tipos de sistemas de informac¸

ao - 31

maneiras de a tecnologia da informac¸

ao ajudar as

organizac¸

oes.

Mara, A. (2020). UX on the Go: A Flexible Guide to User

Experience Design. Taylor & Francis.

Melo, P. (2012). Interac¸

ao academia-ind

ustria: onde esta-

mos e para onde queremos ir.

Mesquita, V. and Darin, T. (2021). Orbis: Um toolkit para

apoio no ensino e pr

atica de t

ecnicas do design cen-

trado no usu

ario. pages 73–78. Sociedade Brasileira

de Computac¸

ao.

MJV Team (2022). Brainwriting: conhec¸a

a t

ecnica para gerar novas ideias.

https://www.mjvinnovation.com/pt-

br/blog/brainwriting/.

Monteiro, I. and Gonc¸alves, E. (2021). Experi

encia com a

molic na ind

ustria: ensino e aplicac¸

ao em um projeto

para o com

ercio eletr

onico. pages 64–72. Sociedade

Brasileira de Computac¸

ao.

Mu, C. (2021). Application of user research in e-commerce

app design. In Nah, F. F.-H. and Siau, K., editors, HCI

in Business, Government and Organizations, pages

120–130, Cham. Springer International Publishing.

Munim, K. M., Islam, I., Rahman, M. M., and Islam, M. N.

(2020). Adopting hci and usability for developing

industry 4.0 applications: A case study. pages 1–6.

IEEE.

Nascimento, N. M. D., Vivacqua, A. S., and Silva, M. F. D.

(2019). Uma proposta de integrac¸

ao de er

Agil com

ihc.

Nathan, R. J. and Yeow, P. H. (2011). Crucial web us-

ability factors of 36 industries for students: a large-

scale empirical study. Electronic Commerce Research,

11:151–180.

Obermeier, G. and Auinger, A. (2019). Human-Computer

Interaction in Physical Retail Environments and the

Impact on Customer Experience: Systematic Liter-

ature Review and Research Agenda, volume 11588,

pages 51–66. Springer International Publishing.

Ogunyemi, A. A., Lamas, D., L

arusd

ottir, M. K., and

Loizides, F. (2019). A systematic mapping study

of hci practice research. International Journal of

Human-Computer Interaction, 35:1461–1486.

Orso, V., Campanini, M. L., Pierobon, L., Portello, G.,

Monaro, M., Bettelli, A., and Gamberini, L. (2023).

Exploiting 3d web to enhance online shopping: To-

ward an update of usability heuristics. In Nah, F. and

Siau, K., editors, HCI in Business, Government and

Organizations, pages 450–460, Cham. Springer Na-

ture Switzerland.

Patton, M. A. and Jøsang, A. (2004). Technologies for

trust in electronic commerce. Electronic Commerce

Research, 4:9–21.

Pazmino, A. (2015). Como se cria: 40 m

etodos para design

de produtos. Blucher.

Pu, P., Chen, L., and Kumar, P. (2008). Evaluating product

search and recommender systems for e-commerce en-

vironments. Electronic Commerce Research, 8:1–27.

Roedl, D. J. and Stolterman, E. (2013). Design research

at chi and its applicability to design practice. pages

1951–1954. ACM.

Salminen, J., Jung, S.-g., Kamel, A., Froneman, W., and

Jansen, B. J. (2022). Who is in the sample? an analy-

sis of real and surrogate users as participants in user

study research in the information technology ﬁelds.

PeerJ Computer Science, 8:e1136.

Silva, J. B. D. and Lotthammer, K. (2019). O uso de uma in-

terface conversacional como ferramenta de interac¸

em um erp para ind

ustria da construc¸

ao. pages 47–48.

Sociedade Brasileira de Computac¸

ao - SBC.

Stolterman, E. and Pierce, J. (2012). Design tools in prac-

tice. pages 25–28. ACM.

Thies, A., Ljungblad, S., and Claesson, I. S. (2015). Beyond

ict: How industrial design could contribute to hci re-

search. Swedish Design Research Journal, 13:22–29.

Tim Browne (2023). The double diamond: A univer-

sally accepted depiction of the design process.

https://www.designcouncil.org.uk/our-resources/the-

double-diamond/.

Vazquez, C. and Sim

oes, G. (2016). Engenharia de Requi-

sitos: software orientado ao neg

ocios. Brasport.

Winn, W. and Beck, K. (2002). The persuasive power of de-

sign elements on an e-commerce web site. Technical

Communication, 49(1):17–35.

Wu, M., Xu, M., and Ge, J. (2023). Analyzing customer ex-

perience and willingness to use towards virtual human

products: Real person generated vs. computer pro-

gram generated. In Nah, F. and Siau, K., editors, HCI

in Business, Government and Organizations, pages

476–486, Cham. Springer Nature Switzerland.

Yang, Y.-C. and Chiu, T.-P. (2023). The study of differ-

ent types of menus layout design on the e-commerce

platform via eye-tracking. In Nah, F. and Siau,

K., editors, HCI in Business, Government and Or-

ganizations, pages 314–328, Cham. Springer Nature

Switzerland.

Process and Challenges in Designing Data Registration via File Import in an Administrative e-Commerce System

401