Next Article in Journal
Prediction of Flotation Deinking Performance: A Comparative Analysis of Machine Learning Techniques
Next Article in Special Issue
How Does Talking with a Human-like Machine in a Self-Driving Car Affect your Experience? A Mixed-Method Approach
Previous Article in Journal
Effect of Pre-Treatment, Treatment, and Extraction Technologies on the Bioactive Substances of Coriander
Previous Article in Special Issue
A Study on the Sound Therapy Interaction Design to Alleviate Homesickness: The HomeSeek App among Chinese Users
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

FRAMUX-EV: A Framework for Evaluating User Experience in Agile Software Development

Escuela de Ingeniería Informática, Pontificia Universidad Católica de Valparaíso, Valparaíso 2340025, Chile
*
Author to whom correspondence should be addressed.
Appl. Sci. 2024, 14(19), 8991; https://doi.org/10.3390/app14198991
Submission received: 18 September 2024 / Revised: 27 September 2024 / Accepted: 30 September 2024 / Published: 6 October 2024
(This article belongs to the Special Issue Advanced Technologies for User-Centered Design and User Experience)

Abstract

:
Agile software development prioritizes customer satisfaction through the continuous delivery of valuable software. However, integrating user experience (UX) evaluations into agile projects remains a significant challenge. Existing proposals address specific stages that apply UX evaluation methods but do not fully consider UX artifacts or UX events for integrating user experience into agile processes. To address this gap and support teams, we propose FRAMUX-EV, a framework for evaluating UX in agile software development using Scrum. FRAMUX-EV introduces seven UX artifacts: (1) UX evaluation methods, (2) UX design system, (3) UX personas, (4) UX responsibilities and roles, (5) UX evaluation repository, (6) UX backlog, and (7) UX sprint backlog; and four UX events: (1) pre-planning UX meeting, (2) pre-review UX meeting, (3) weekly UX meeting, and (4) weekly user meeting. The first version of the framework was developed using a seven-step methodology with a qualitative approach. A survey of 34 practitioners validated the usefulness and ease of integration of FRAMUX-EV components, yielding positive results. These findings suggest the potential of FRAMUX-EV as an interesting proposal for integrating UX into agile software development.

1. Introduction

Agile software development has become an increasingly popular practice among companies of various sizes. This approach emphasizes the frequent delivery of valuable software to ensure customer satisfaction, driven by the efforts of motivated teams [1]. In agile projects, it is important to consider users’ needs and objectives throughout the development process to design and develop a product that fulfills their expectations. However, integrating user experience (UX) into agile approaches presents significant challenges, as various issues arise when attempting to align these two disciplines [2,3,4]. This challenge is further compounded by the fact that even Scrum, the most widely used agile approach [5], lacks clear guidance on incorporating UX activities, UX evaluation methods, or UX roles. Moreover, Scrum does not address how to involve representative users effectively [6,7].
Our research on UX evaluation in agile software development revealed a lack of approaches specifically tailored for this integration. While several proposed approaches outline specific evaluation steps and methods aimed at achieving this goal, existing proposals focus on certain stages where evaluation methods are applied. However, they do not fully consider UX artifacts or UX events to effectively integrate UX into agile processes. For this reason, the development of a framework that incorporates UX events and UX artifacts to support UX evaluation in agile software development will benefit both UX and development teams. This will enable them to create high-quality products that take into account the user’s perspective, including their needs, pain points, and goals.
We propose FRAMUX-EV: a FRAMework for evaluating the User eXperience in agile software development using Scrum. FRAMUX-EV was developed using a seven-step methodology, combining different UX evaluation methods and introducing new UX artifacts and UX events. It incorporates seven UX artifacts: (1) UX evaluation methods, (2) UX design system, (3) UX personas, (4) UX responsibilities and roles, (5) UX evaluation repository, (6) UX backlog, and (7) UX sprint backlog. Additionally, it includes four UX events: (1) pre-planning UX meeting, (2) pre-review UX meeting, (3) weekly UX meeting, and (4) weekly user meeting.
A survey was conducted to collect and analyze the perceptions of practitioners. The main objective was to validate the usefulness and ease of integration of UX components (i.e., UX evaluation methods, UX artifacts, and UX events) for evaluating UX in agile software development, specifically within Scrum. In addition, practitioners were encouraged to recommend new components, suggest removing one component, or provide additional feedback. Overall, the UX components in the first version of FRAMUX-EV were positively received by practitioners. Consequently, no UX evaluation method, UX artifact, or UX event was eliminated in this first iteration. However, some refinements are necessary before proceeding with the remaining experiments and validations.
This article is organized as follows: Section 2 introduces the theoretical background; Section 3 indicates the need for a UX evaluation framework; Section 4 explains the methodology; Section 5 presents FRAMUX-EV; Section 6 details the validation; Section 7 details the discussions; and Section 8 summarizes the conclusions and outlines future work.

2. Theoretical Background

The concepts of user experience, agile software development, and Scrum are presented below. In addition, related works are analyzed.

2.1. User Experience (UX)

The ISO 9241-210 standard [8] defines UX as “a person’s perceptions and responses resulting from the use and/or anticipated use of a product, system, or service”. Similarly, Schulze and Krömker [9] characterize UX as “the degree of positive or negative emotions that can be experienced by a specific user in a specific context during and after product use and that motivates for further usage”.
In software development, UX plays a critical role in determining overall user satisfaction. A well-designed UX goes beyond ensuring ease of use; it also addresses whether the software meets user needs and expectations. Evaluating UX throughout the development process is essential, as it enables the early detection and resolution of issues before the product reaches the end user [10]. Additionally, a continuous UX evaluation ensures that the product adapts to evolving user needs and expectations.
There are numerous methods available for evaluating UX [11]. These techniques effectively identify issues related to interface design, user interaction, navigation, and usability. However, applying them in agile development environments can be challenging due to time constraints [12,13].

2.2. Agile Software Development

Agile is described as “the ability to create and respond to change, offering a means to navigate and ultimately succeed in an uncertain and turbulent environment” [14]. Sommerville [15] highlights that agile software development focuses on quickly delivering functional software through incremental releases, with each increment introducing new system features. Among various agile approaches, Scrum is the most widely adopted [5]. Other commonly used agile approaches include Kanban, ScrumBan, Extreme Programming (XP), and Lean Startup, among others [5].
Scrum is a lightweight framework designed to help individuals, teams, and organizations deliver value by developing adaptive solutions for complex challenges [16]. To use Scrum effectively and efficiently, it is essential to know and apply the Scrum values and principles, which are reflected in three pillars [16]: (1) transparency, (2) inspection, and (3) adaptation. One of the main features of Scrum is that it has events and artifacts to facilitate software development [16]. The events are (1) sprint planning, (2) sprint, (3) daily meeting, (4) sprint review, and (5) sprint retrospective, while the Scrum artifacts are (1) product backlog, (2) sprint backlog, and (3) increments.

2.3. Related Work

We identified different frameworks that integrated components in their proposal to evaluate UX in agile software development, either through UX artifacts, UX events, UX evaluation stages, or UX evaluation methods (see Table 1). We identified two types of studies: (1) proposals that integrate new elements into an existing agile approach [2,17,18,19,20]; and (2) proposals based on sequential and/or iterative agile activities [21]. These frameworks are usually composed of three to four phases, typically including Scrum events and different UX evaluation methods such as user testing, the rapid iterative testing and evaluation (RITE) method, pair design, paper prototypes, or heuristic evaluation. Although we identified one study that mentioned the presence of a UX artifact [19] and another that referenced a UX event [18], we observed a recurring issue in these studies, as they lacked detailed explanations of how these elements function or how they could be integrated into agile approaches. Therefore, no proposals focused on evaluating the UX throughout the process and phases of agile software development (incorporating all of these components) were found.

3. Need for a UX Evaluation Framework

Agile approaches are increasingly being used by companies for software development. This iterative development approach helps teams to deliver value to their customers faster and with fewer headaches [22]. Given frequent deliveries, it is important to check that what users need is actually being developed. Therefore, it is necessary to include the UX in agile software development, specifically the UX evaluation, to avoid neglecting the needs of end-users. However, some problems can be observed when integrating and evaluating UX within agile software development:
  • Difficulty in selecting the best method: Owing to the large number of UX evaluation methods, it is difficult to understand when it is convenient to apply some evaluation methods [23]. This becomes more difficult in changing and agile environments [24,25,26].
  • Users are not actively incorporated: When working in an agile way, teams often neglect to incorporate users to evaluate an idea or decide what to implement. User feedback is often ignored or considered only in specific instances such as sprint reviews [25,27,28].
  • Prioritization of design over evaluation: Practices such as software design (user-centered design) are prioritized over evaluation (evaluation-centered evaluation). Thus, the team develops software primarily by considering what users want (functionalities) rather than how they want it (needs and goals) [26].
  • Communication problems: There is a lack of communication between the design/UX team and the development team, resulting in misunderstandings in the design and loss of important information [26,27,28].
  • Lack of early UX evaluation: In agile software development, it is believed that UX can (or should) be evaluated only in the final stages. Therefore, in the early stages, decisions are made without considering users’ needs [27].
  • Difficulty in prioritizing UX work: No backlog is used to highlight user stories focused on UX. In addition, sprint goals are typically focused on development rather than UX, making it difficult to manage and check UX work [25,27].
Most of these problems arise because agile approaches present a set of recommendations or guidelines for iterative development. However, these do not explicitly indicate the tasks or actions that each team must perform to evaluate the UX. For this reason, our study proposes a framework for evaluating UX in agile software development, considering different elements to solve these problems, that is, UX events and artifacts.

4. Methodology

The methodology for developing the framework was divided into seven distinct steps (see Figure 1).
In Step 1, we defined the problem statement, proposed the objectives, established a methodology (including the approach, scope, and steps), and outlined the work plan.
In Step 2, we conducted a systematic literature review of frameworks, methodologies, UX evaluation methods, challenges, and recommendations related to integrating UX into agile software development [29]. We identified different studies that incorporated components of UX evaluation in agile development, focusing on roles, artifacts, events, or evaluation methods. However, no proposals were found that comprehensively evaluated UX across all phases of the agile development process, including UX events, artifacts, and evaluation methods. We identified several UX evaluation methods from these proposals, along with challenges and recommendations related to various UX stages and roles.
In Step 3, we conducted six in-depth interviews to gather and analyze the experiences of various practitioners (e.g., challenges, recommendations, relationships between roles and teams, and adopted practices) when integrating UX into agile software development [30]. Additionally, these interviews were conducted to validate the findings of the systematic literature review. As a result, we identified several methods used for UX evaluation, along with the challenges faced when integrating UX into agile environments, as well as recommendations and practices suggested by practitioners.
In Step 4, we synthesized the insights gathered from both the systematic literature review [29] (i.e., 23 UX evaluation methods, 56 problems, and 91 recommendations) and practitioner interviews [30] (i.e., 9 evaluation methods, 25 problems, and 40 recommendations). A comprehensive comparative analysis was conducted to identify common elements, the most frequently mentioned aspects, novel insights, and notable cases of success or failure. Based on this analysis, some of the most relevant findings were selected to create the initial framework proposal.
In Stage 5, based on the findings from the previous stage, we proposed different UX components to be included in the proposal: 8 UX evaluation methods, 7 UX artifacts, and 4 UX events. This proposal of components resulted in the first version of FRAMUX-EV, a framework for evaluating user experience in agile software development.
In Step 6, we conducted the first validation of FRAMUX-EV to evaluate the usefulness and ease of integration of the proposed UX components (UX artifacts, UX events, and UX evaluation methods) for evaluating UX in agile software development, specifically within Scrum. The validation involved gathering feedback from experts (developers and UX roles) regarding the elements that they suggested adding, modifying, or removing to improve the framework.
In Step 7, we defined the refinements required for the first version of FRAMUX-EV based on the information gathered in the previous step. The experts’ feedback on each component, along with future actions to refine UX events and UX artifacts, is detailed in Section 6.1, Section 6.2 and Section 7.6.

5. FRAMUX-EV: First Version

FRAMUX-EV is a framework for evaluating the user experience in agile software development using Scrum. FRAMUX-EV incorporates seven UX artifacts: (1) UX evaluation methods, (2) UX design system, (3) UX personas, (4) UX responsibilities and roles, (5) UX evaluation repository, (6) UX backlog, and (7) UX sprint backlog. In addition, FRAMUX-EV incorporates four UX events: (1) pre-planning UX meeting, (2) pre-review UX meeting, (3) weekly UX meeting, and (4) weekly user meeting. Figure 2 presents the first version of FRAMUX-EV.

5.1. Inputs

Different inputs were considered when proposing the first version of the framework for evaluating UX in agile software development:
  • Systematic literature review: The systematic literature review provided frameworks or similar (e.g., methodologies, processes, or approaches), highlighting 23 UX evaluation methods, 56 problems/challenges, and 91 recommendations/practices, which were taken as a reference to propose new UX components [29].
  • Interviews: The interviews with practitioners provided insight into the industry’s perspective, highlighting 9 UX evaluation methods, 25 problems/challenges, and 40 recommendations/practices experienced by UX roles and developers when developing software in agile projects [30].

5.2. UX Evaluation Methods

There are multiple UX evaluation methods that can be used to evaluate the UX in agile software development. We decided to select the eight evaluation methods mentioned in the interviews with practitioners and systematic literature review for the first version of FRAMUX-EV (see Table 2).

5.3. UX Artifacts

Based on the findings obtained from the practitioner interviews and systematic literature review, we proposed seven UX artifacts (see Table 3).

5.4. UX Events

Based on the findings obtained from the practitioner interviews and systematic literature review, we proposed four UX events (see Table 4).

6. Validating the Framework

We conducted a survey to gather and analyze the perceptions of 34 practitioners (P1–P34) who had worked in UX/UI roles or as developers in agile projects (see Table 5). The survey consisted of four main sections: (1) practitioners’ demographics and agile experience, (2) practitioners’ perspectives on UX evaluation methods, (3) practitioners’ perspectives on UX artifacts, and (4) practitioners’ perspectives on UX events. The main objective was to validate the proposed UX components in FRAMUX-EV using a five-level Likert scale (1—worst to 5—best) in two dimensions: (D1) usefulness and (D2) ease of integration into agile software development. In addition, practitioners were able to recommend new components, suggest the elimination of one component, or provide additional feedback.

6.1. Quantitative Results

The following section presents the perceptions of the surveyed practitioners regarding the usefulness (D1) and ease of integration (D2) of the proposed UX evaluation methods, UX artifacts, and UX events for FRAMUX-EV.

6.1.1. Quantitative Results: UX Evaluation Methods

Table 6 summarizes the survey results for dimensions D1 and D2 of the UX evaluation methods proposed in the first version of the FRAMUX-EV. A brief analysis of the descriptive statistics is provided below.
(D1) Usefulness. The UX evaluation method that obtained the best rating from the practitioners was the “usability/user test” with a score of 4.79 out of 5.00. Another UX evaluation method considered useful by the participants was “evaluation with mockups or prototypes”, with 4.26 out of 5.00. On the other hand, the UX evaluation methods with the worst usefulness perceived by the participants were “pluralistic walkthrough” and “system usability scale” with a score of 3.35 out of 5.00. However, although these methods were the worst evaluated, none were rated as “not very useful” (score 2) or “very unhelpful” (score 1).
(D2) Ease of integration. The UX evaluation method considered the easiest to integrate by the practitioners was “evaluation with mockups or prototypes”, with a score of 4.12 out of 5.00, followed by “usability/user test”, with a score of 3.65 out of 5.00. On the other hand, the UX evaluation methods with the worst ease-of-integration ratings assigned by the participants were “pluralistic walkthrough”, with a score of 2.91 out of 5.00, and “RITE method”, with a score of 2.85 out of 5.00. Surprisingly, only one UX evaluation method obtained an average score higher than 4.0 for ease of integration.
Initially, no UX evaluation method should be discarded from this initial version of FRAMUX-EV, as all received a usefulness rating of above 3. However, it is necessary to provide more detailed guidance on how and when the proposed methods should be used, with a particular emphasis on those considered more challenging to integrate into agile environments. The decision to discard UX evaluation methods or provide detailed guidance will be revisited after analyzing the qualitative feedback from the experiment.

6.1.2. Quantitative Results: UX Artifacts

Table 7 summarizes the survey results for dimensions D1 and D2 of the UX artifacts proposed in the first version of the FRAMUX-EV. A brief analysis of the descriptive statistics is provided below.
(D1) Usefulness. The UX artifact that obtained the best rating from the practitioners was “UX backlog”, with a score of 4.26 out of 5.00, followed by “UX evaluation repository”, with 4.18 out of 5.00. On the other hand, the UX artifacts with the worst usefulness perceived by the participants were “UX personas”, with a score of 3.82 out of 5.00, and “UX responsibilities and roles”, with a score of 3.29 out of 5.00. However, although these UX artifacts were the worst evaluated, none were rated as “not very useful” (score 2) or “very unhelpful” (score 1).
(D2) Ease of integration. The UX artifact considered the easiest to integrate by the practitioners was “UX responsibilities and roles” with 3.91 out of 5.00. Another UX artifact considered easy to integrate by the participants was “UX evaluation methods”, with a score of 3.85 out of 5.00. On the other hand, the UX artifacts with the worst rating assigned by the participants were “UX personas”, with a score of 3.50 out of 5.00, and “UX backlog”, with a score of 3.26 out of 5.00. Although these UX artifacts were the worst evaluated, none were rated as “difficult” (score 2) or “very difficult” (score 1). In addition, none of the UX artifacts were classified as “ease to integrate” (score 4).
Initially, no UX artifacts should be discarded from this initial version of FRAMUX-EV, as all received a usefulness rating above 3. However, it is necessary to provide more detailed explanations of how and when all the proposed UX artifacts can be used, as, surprisingly, none received a score higher than 4. The decision to discard UX artifacts or provide detailed guidance will be revisited after analyzing the qualitative feedback from the experiment.

6.1.3. Quantitative Results: UX Events

Table 8 summarizes the survey results for dimensions D1 and D2 of the UX events proposed in the first version of the FRAMUX-EV. A brief analysis of the descriptive statistics is provided below.
(D1) Usefulness. The UX event that obtained the best rating from the practitioners was “weekly UX meeting”, with a score of 4.41 out of 5.00, followed by “pre-planning UX meeting”, with a score of 4.09 out of 5.00. On the other hand, the UX events with the worst usefulness perceived by the participants were “pre-review UX meeting” and “weekly user meeting”, with a score of 3.85 out of 5.00. However, although these UX events were the worst evaluated, none were rated as “not very useful” (score 2) or “very unhelpful” (score 1).
(D2) Ease of integration. The UX event considered the easiest to integrate by the practitioners was “pre-review UX meeting”, with a score of 3.97 out of 5.00, followed by “pre-planning UX meeting”, with a score of 3.94 out of 5.00. On the other hand, the UX event with the worst rating assigned by the participants was “weekly user meeting”, with a score of 2.35 out of 5.00. Surprisingly, none of the UX events were classified as “ease to integrate” (score 4).
Initially, all UX events were perceived as useful, and none should be discarded from this initial version of FRAMUX-EV, as each received a usefulness rating above 3. However, it is important to analyze the feedback and suggestions provided by the experts to understand the reasons behind the lower score given to “weekly user meeting”. The decision to discard UX events or modify them to improve their ease of integration will be revisited after analyzing the qualitative feedback from the experiment.

6.2. Qualitative Results

The following section presents feedback and suggestions from the surveyed practitioners regarding the proposed UX evaluation methods, UX artifacts, and UX events for FRAMUX-EV.

6.2.1. Qualitative Results: UX Evaluation Methods

In addition to the methods presented in the survey, 29 additional methods were suggested for consideration. However, almost all the methods mentioned were related to UX design methods rather than UX evaluation methods (e.g., card sorting, tree tests, benchmarks, focus groups, or surveys). The only ones specifically focused on UX evaluation were heat maps (P4, P5, P16, P27), eye tracking (P4, P15), and the five-second test (P5, P21).
Half of the respondents were satisfied with the methods presented, indicating that they would not remove any of the proposed methods (17 of the 34 participants). Meanwhile, the other half were uncertain about whether to remove any method. “A/B testing” was the most frequently mentioned method for elimination, with 6 out of 34 respondents suggesting its removal (P6, P9, P12, P18, P20, P21), followed by the “system usability scale (SUS)”, with four responses (P2, P3, P16, P26). It should be noted that this does not mean that these methods are not useful, but that they may be less applicable or effective in certain situations or contexts (P4, P9, P12, P14).
Participants provided additional comments on the methods presented. For instance, P2 pointed out that “It would be necessary to clarify the methodology you are thinking about for agile software development and what kind of teams you are going to have. The methods will eventually depend on that”. P3 mentioned that “(evaluation methods) must be used at the right time, always proportional to the problem to be addressed”. P15 highlighted that “I find this list quite interesting and complete. They are essential methods with which it is possible to conduct a complete evaluation of the usability and therefore of the user experience”. Table 9 presents the comments obtained.
In summary, we noted that there are many UX evaluation methods that can be used to evaluate the UX in agile software development projects. However, each method has its own strengths and weaknesses, and it is important to choose the appropriate methods according to the specific objectives of a project (P3, P11). Taking this into consideration, a complete evaluation of usability and user experience can be conducted (P15, P25). Therefore, based on the quantitative and qualitative results, no UX evaluation method will be eliminated from this initial version of FRAMUX-EV.

6.2.2. Qualitative Results: UX Artifacts

In addition to the UX artifacts presented in the survey, 15 additional artifacts were suggested for consideration. However, almost all the suggestions were specifically related to UX design methods rather than UX artifacts (e.g., user scenarios, customer journey, or user flows). The suggestions that could be considered artifacts (e.g., documents, guidelines, and repositories) were the “repository of UX tools” (P5, P20), “KPI guidelines for stakeholders” (P17), and “content repositories” (P4).
The majority of the respondents seemed satisfied with the artifacts presented, indicating that they would not eliminate any of the artifacts proposed (24 of 34 participants). “UX personas” was the most frequently mentioned artifact for elimination, with 4 out of 34 respondents suggesting its removal (P2, P5, P19, P20), followed by the “UX design system”, with three responses (P9, P19, P26).
Although few comments were obtained, there were different opinions regarding the proposal. For instance, P15 and P20 highlighted that it is a “good initiative” and “It is an interesting proposal”, respectively. P24 and P32 highlighted the complexity of the integration and learning of new artifacts in an agile context. Table 10 presents additional comments related to each UX artifact and future actions to be considered.
In summary, there were generally no negative comments regarding the proposed UX artifacts. However, it is necessary to better specify the description of these artifacts to avoid confusion and highlight their usefulness (P2, P5, P14, P15), and to include specific considerations for some of them (P19). Considering this, these improvements will help ensure a clearer understanding and more effective use of UX artifacts in the agile development process. Therefore, based on the quantitative and qualitative results, no UX artifacts will be eliminated from the initial version of FRAMUX-EV.

6.2.3. Qualitative Results: UX Events

In addition to the UX events presented in the survey, 10 additional events were suggested for consideration. The most mentioned events were “UX feedback meeting” (P4, P19, P21) and “daily UX meeting” (P5, P7). Surprisingly, 20 practitioners considered it unnecessary to include any additional UX events beyond those mentioned in the survey.
The majority of respondents seemed satisfied with the events presented, indicating that they would not eliminate any of the artifacts proposed (25 of 34 participants). However, some practitioners suggest eliminating the “pre-review UX meeting” because it is not the right time to do it (P30) and it should be done with the whole team (P19), as well as the “weekly UX meeting”, where it is mentioned that they would not do it weekly (P32) as it is unnecessary (P4) given the time they have within the sprint (P26).
Although few comments were obtained, there were different opinions regarding the proposal. For instance, it was stressed that the “weekly UX meeting” should not have to be weekly as it takes too much time and could be unnecessary (P4, P16, P20, P32). On the other hand, it was mentioned that events and meetings should be held with the entire team (P14, P32). Table 11 presents additional comments related to each UX event and future actions to be considered.
In summary, there were no significant negative comments regarding proposed UX events. However, it is necessary to adjust the frequency of certain events to better align with agile environments (P16, P20, P26, P32). These modifications will ensure that UX activities are seamlessly integrated into agile workflows. Therefore, based on the quantitative and qualitative results, no UX events will be eliminated from the first version of FRAMUX-EV.

7. Discussions

In this section, we explain how to apply FRAMUX-EV in a sprint, in various real-world scenarios, and in other agile contexts or approaches. We include a comparative analysis between FRAMUX-EV and other existing proposals, discuss the challenges in implementing FRAMUX-EV, detail its contributions, and explain the limitations and opportunities for improving FRAMUX-EV.

7.1. How to Use FRAMUX-EV

The integration of UX practices into agile development frameworks, such as Scrum, is essential to ensure that software products meet both functional and UX requirements. FRAMUX-EV, a framework proposed for supporting UX evaluation within agile approaches, provides a flexible approach for integrating UX events and artifacts throughout a Scrum iteration. Below, we outline how FRAMUX-EV can be integrated into the traditional Scrum iteration through dedicated UX events and UX artifacts, ensuring that UX is prioritized along with development tasks:
  • Before the sprint planning, the team conducts a pre-planning UX meeting. This event aims to verify whether the prioritized UX work in the UX backlog is ready to be undertaken by the development team during the sprint.
  • During the sprint planning, the Scrum team selects items from the product backlog to be developed during the sprint. Simultaneously, the UX tasks from the UX backlog are prioritized and integrated into the UX sprint backlog to ensure that the UX requirements are covered.
  • The team meets daily during the daily meeting to track the progress of sprint tasks. During this event, both developers and UX roles ensure that any blockers or impediments in the UX work are promptly resolved. If UX team members have questions about any aspect of their tasks, they can refer to UX roles and responsibilities.
  • Once per week, a weekly UX meeting is held, where the UX team and developers discuss the progress of UX design, review updates to the UX components, and resolve issues related to the feasibility of UX designs. In this meeting, potential changes to the UX design system may be reviewed, and the team discusses the evolution of the UX work
  • Once per week, a weekly user meeting is held with users to evaluate UX work using some of the methods suggested in UX evaluation methods, discuss upcoming tasks, and obtain direct feedback on ongoing designs. The outcomes of these meetings are documented in the UX evaluation repository for subsequent analysis and refinement.
  • Just before the sprint review, a pre-review UX meeting is conducted, where the UX and development teams validate whether the UX designs and functions implemented during the sprint meet the required standards and objectives.
  • During the sprint review, the Scrum team presents the product increment, including both development and UX implementations. The UX team can demonstrate how the UX personas, UX design system, and UX evaluation repository guided the final design presented to the stakeholders.
  • After the sprint review, the team holds a sprint retrospective to discuss what went well and what could be improved, focusing on both the development and the UX integration aspects. In this event, improvements to UX artifacts, UX events, or UX evaluation methods may be proposed.
In addition, we present practical examples illustrating how FRAMUX-EV components (UX events, UX artifacts, and UX evaluation methods) can be applied across three distinct Scrum team scenarios: (1) a team with only developers and no dedicated UX roles, (2) a small team with both developers and one UX role, and (3) a larger team with multiple developers and UX roles working together. Each scenario demonstrates how UX considerations can be adapted to varying team compositions, ensuring that user-centered design remains a priority (see Table 12).
Table 12 highlights the flexibility of FRAMUX-EV, which can be adapted to fit different Scrum team structures by modifying the number of UX evaluation methods, UX events, and UX artifacts incorporated into the workflow. In teams where no dedicated UX role exists, only the core elements, such as the weekly user meeting and key UX evaluation methods, are implemented. This approach empowers developers to integrate user-centered design practices into their projects. Smaller teams with a single UX role may introduce more UX events and UX artifacts, like the weekly UX meeting and pre-review UX sessions, to support ongoing UX improvements. Larger teams with multiple UX roles can adopt the full range of FRAMUX-EV’s UX events and UX artifacts, enabling thorough UX integration across all Scrum iterations. This flexibility ensures teams of any size or level of UX expertise can maintain a strong user focus throughout the project.
While FRAMUX-EV has been proposed for integration with Scrum, its flexible structure allows it to be adapted for use in other agile approaches. For example, Kanban is an agile approach that focuses on continuous workflow rather than using fixed iterations like Scrum. In Kanban, tasks are managed visually on a board, and the goal is to improve efficiency by controlling the amount of work in progress (WIP). To integrate FRAMUX-EV with Kanban, some events would need to be adjusted. Meetings such as the “pre-planning UX meeting” or the “pre-review UX meeting” could be transformed into “milestones” to be reached within the workflow, rather than events that occur before or after each iteration. The “weekly UX meeting” would remain relevant, although it could be adjusted to a more flexible frequency depending on the pace of the work. Artifacts such as the “UX backlog” and the “UX evaluation repository” could also be integrated into Kanban, allowing for visual and continuous management of UX-related tasks.
On the other hand, FRAMUX-EV could also be integrated with the agile approach Extreme Programming (XP). XP places a strong emphasis on close collaboration between developers and clients, with short iterations and a focus on continuous improvement through rapid feedback and small software increments. FRAMUX-EV fits well with XP due to its iterative nature. Events such as the “weekly UX meeting” and the “pre-review UX meeting” can be adapted to the end of each XP iteration, enabling quick and continuous user feedback integration. The “pre-planning UX meeting” can also align with XP’s iteration planning meetings, ensuring that UX tasks are prioritized and worked on alongside functionalities. FRAMUX-EV artifacts such as the “UX backlog” and the “UX sprint backlog” can continue to be used as defined, ensuring that UX tasks align with both technical expectations and user needs.

7.2. Comparison between FRAMUX-EV and Existing Proposals

As mentioned in Section 2.3, we identified six different proposals that integrate various elements for evaluating UX in agile software development [2,17,18,19,20,21]. Table 13 presents a comparison between the six analyzed proposals and FRAMUX-EV, in terms of the components they include (events, artifacts, and UX evaluation methods), the representation mode (whether figures, diagrams, or tables are used to explain the proposal), the validation performed on the proposal, strengths, and weaknesses.
Of the six proposals analyzed (see Table 13), five do not propose UX events [2,17,19,20,21], and five do not propose UX artifacts [2,17,18,20,21]. Having specific events and artifacts can facilitate the continuous and planned integration of UX evaluation in each agile iteration. In comparison with the proposals by Felker et al. [2], Maguire [17], Argumanis et al. [20], and Gardner and Aktunc [21], which do not include specific UX events or artifacts, FRAMUX-EV introduces seven UX artifacts and four UX-specific events, providing a clear evaluation structure that integrates with Scrum. The inclusion of “weekly UX meetings” and the “UX backlog” allows for iterative and continuous UX evaluation, aligned with agile principles, offering greater flexibility and responsiveness.
Regarding UX methods, most of the proposals analyzed include UX evaluation methods, except for one [18]. The number of methods included in each proposal varies between one and five. FRAMUX-EV includes eight UX evaluation methods, four of which appear in existing proposals (user testing [19,20,21], heuristic evaluation [17], evaluation with mockups or prototypes [2], and the RITE method [2]), and four additional UX evaluation methods (guerrilla testing, A/B testing, pluralistic walkthrough, and SUS), based on the results obtained from the systematic literature review conducted [29] and interviews with practitioners [30]. While Pillay and Wing [18] and Weber et al. [19] offered interesting proposals for evaluating UX within each sprint (including Lean UX principles [18] and various UX evaluation methods [19]), they did not specify when to conduct UX evaluations or how to manage UX work in each iteration. In comparison, FRAMUX-EV provides a clear structure that includes artifacts (such as the “UX evaluation repository” and the “UX sprint backlog”) and events (such as the “pre-planning UX meeting” and the “pre-review UX meeting”), facilitating UX evaluations at key moments in the agile development cycle and providing visibility into UX work.
On the other hand, three of the proposals analyzed have not been validated [2,17,18], while the other three were validated through interviews [19,21]), case studies [19], real projects [20], and questionnaires [21]. FRAMUX-EV was validated through surveys conducted with practitioners who have been working in the industry for several years developing agile projects (see Section 6). FRAMUX-EV, in addition to addressing the deficiencies of previous approaches, adds significant value by formalizing UX activities within Scrum through the introduction of specific artifacts and events. This ensures that UX evaluation becomes an integral part of the agile process, improving collaboration between UX and development teams and enabling faster and more effective feedback in each sprint. The combination of these elements provides a more comprehensive and adaptable framework for evaluating user experience in agile environments.

7.3. Challenges in Applying FRAMUX-EV

The integration of FRAMUX-EV in agile projects can present some significant challenges for different work teams. Introducing new UX events and UX artifacts requires not only adjustments to well-established workflows but also the involvement of the entire team, which can be difficult to guarantee. Teams that are familiar with their current way of working may resist the changes required to adapt to this new proposal. In addition, the time required for these UX activities, in particular the weekly UX meeting and weekly user meeting, could overload already tight sprint schedules, raising questions about their feasibility in agile projects with small iterations.
The framework’s adaptability to different team compositions is another critical factor for its success. FRAMUX-EV must work effectively across different team structures, from those without dedicated UX roles to teams that include multiple UX roles. However, this flexibility introduces a learning curve, as teams will need time to understand and properly apply the UX artifacts and UX events. Furthermore, it can be challenging for teams without UX roles to balance UX and development tasks in the backlog and sprint planning, as developers need to make sure UX tasks get enough focus but without slowing down development activities.
On the other hand, creating and maintaining up-to-date UX artifacts, such as the UX backlog, the UX evaluation repository, and the UX design system, requires the team’s continuous dedication to ensure their relevance and usefulness throughout the project lifecycle. This effort is essential to prevent UX artifacts from becoming outdated given the evolving needs of the project. Finally, user involvement is crucial to the success of FRAMUX-EV; however, obtaining and analyzing user feedback without delaying other activities could be a problem for some teams.

7.4. Contributions

This framework introduces a flexible approach to integrate and support UX evaluation in agile software development using Scrum, addressing a significant gap in the existing literature. It includes new UX components within Scrum by proposing seven UX artifacts and four UX events to ensure that UX activities are constantly integrated throughout the project lifecycle. In addition, FRAMUX-EV suggests eight different UX evaluation methods, offering a wider range of options than previous proposals and increasing flexibility to support different team compositions and sizes, with or without UX roles. Moreover, FRAMUX-EV promotes collaboration between the UX team and the development team by synchronizing efforts and facilitating early feedback, thereby decreasing communication problems. Considering all these UX components, our proposal offers a comprehensive and flexible approach compared to existing approaches for integrating and supporting UX in each iteration throughout an agile project.

7.5. Limitations

This study presents different limitations that must be acknowledged. First, the validation survey was conducted with only 34 practitioners, limiting the generalizability of the results. A larger sample size would offer more robust insights and improve the reliability of the findings. Additionally, since this is the first version of the framework, only one validation has been conducted, and it has not yet been applied in real-world agile software development projects. As a result, its effectiveness and practicality in actual settings remain to be evaluated, leaving a gap between theoretical development and real-world applicability.
Furthermore, the UX events proposed in the framework are specifically designed for use within Scrum, and their applicability to other agile methodologies, such as Kanban or Extreme Programming (XP), has not yet been addressed. However, this limitation primarily applies to UX events, as UX artifacts and UX evaluation methods could potentially be used in other agile approaches. Finally, the study lacks a detailed specification of each proposed event and artifact, including aspects such as descriptions, objectives, life cycles, frequency, duration, and key considerations. This absence may pose challenges for teams attempting to implement the framework, as it corresponds to the first version of FRAMUX-EV. However, these details will be included in the next version of the proposal after refinements and additional experiments.

7.6. Opportunities to Improve FRAMUX-EV

As presented in Section 6.2, the experiment conducted to validate the UX components of FRAMUX-EV provided valuable insights and findings on future actions that should be implemented. Suggestions for both UX events and UX artifacts highlight opportunities to improve utility and facilitate integration, so that the framework continues to evolve and can be used to integrate and support the UX evaluation into agile software development.
Thus, several changes have been identified based on the feedback gathered from experts. For UX artifacts (see Table 10), improvements will include providing more detailed explanations for all UX artifacts, with particular emphasis on the UX design system, UX personas, UX roles and responsibilities, and UX evaluation repository, to avoid confusion and highlight their utility. The feedback also suggested including a consideration for each UX artifact to indicate that; depending on the team’s capacity, these could be specified in more or less detail.
On the other hand, for UX events (see Table 11), modifications include adjusting the frequency of three events: weekly UX meeting, weekly user meeting, and pre-review UX meeting. First, the frequency of the weekly UX meeting will be adjusted to better accommodate the capabilities of different teams. Similarly, the weekly user meeting will be scheduled more flexibly, either by iterations or based on the volume of work completed. Additionally, the frequency of the pre-review UX meeting will be modified to ensure that teams have time to make changes before the sprint review. For this reason, the weekly UX meeting will be renamed to “UX meeting”, while the weekly user meeting will be renamed to “user meeting” to emphasize the flexibility of its frequency.
After implementing these changes, FRAMUX will require additional iterations to ensure it can be used to support teams in integrating and evaluating UX in agile software development. Thus, the following FRAMUX-EV iterations include the following stages and activities:
  • Iteration 2: In this phase, the changes identified during this first iteration will be implemented to develop the second version of FRAMUX-EV. This version will need to be validated with UX practitioners by conducting two experiments: (1) an experiment focused on evaluating the specification and detailed content of UX artifacts and (2) an experiment focused on evaluating the specification and feasibility of UX events. Both experiments aim to validate that the proposal meets the industry standards and project needs. Based on these expert evaluations, the necessary refinements will be identified, which will lay the groundwork for the next iteration of the framework.
  • Iteration 3: The findings from the second iteration will be applied to further refine the framework, leading to the development of the third version of FRAMUX-EV. This version will be tested in real-world projects with two case studies to validate its effectiveness and applicability. In each experiment, feedback will be collected from UX and development roles to assess the effectiveness of integrating FRAMUX-EV in agile projects. The results of these validations will be used to obtain the final necessary adjustments, preparing the framework for its final iteration.
  • Iteration 4: Finally, based on the feedback and improvements from the previous iteration, the necessary adjustments will be made to FRAMUX-EV to present its final version. This version will include all refinements identified during experiments in real-world projects and will represent a mature framework for evaluating and integrating UX into agile software development.
  • Following these iterations, FRAMUX-EV will continuously evolve, ensuring the integration of UX evaluation and agile software development processes and providing more robust and effective results.

8. Conclusions and Future Work

In this article, we proposed FRAMUX-EV, a framework for evaluating user experience in agile software development using Scrum. The proposal presents seven UX artifacts: (1) UX evaluation methods, (2) UX design system, (3) UX personas, (4) UX responsibilities and roles, (5) UX evaluation repository, (6) UX backlog, and (7) UX sprint backlog and four UX events: (1) pre-planning UX meeting, (2) pre-review UX meeting, (3) weekly UX meeting, and (4) weekly user meeting.
A survey of 34 practitioners validated the usefulness and ease of integration of the proposed UX components. Overall, the UX evaluation methods, UX artifacts, and UX events were perceived as useful, with most receiving ratings above 3 out of 5. However, some components were seen as more challenging to integrate into agile workflows. Based on quantitative and qualitative feedback, no UX components were eliminated from the initial version of FRAMUX-EV. However, some refinements are required: (1) provide more detailed guidance on how and when to use the proposed UX evaluation methods, especially those considered more difficult to integrate; (2) better specify the descriptions of UX artifacts to avoid confusion and include specific considerations for their use; and (3) adjust the frequency of certain UX events to better align with agile timelines and team structures.
Future work will focus on implementing these refinements and conducting further validation of the framework through case studies of real agile software development projects. In addition, a detailed specification of each event and artifact proposed in FRAMUX-EV will be provided. This will help evaluate FRAMUX-EV’s effectiveness in practice and identify any additional improvements required. By providing a structured approach to UX evaluation within agile processes, FRAMUX-EV aims to benefit both UX and development teams by creating high-quality products that truly consider users’ needs and goals. Further research and practical applications will help to evolve the framework to better serve agile and UX communities.

Author Contributions

Conceptualization, L.F.R. and D.Q.; methodology, L.F.R. and D.Q.; software, L.F.R. and D.Q.; validation, L.F.R.; formal analysis, L.F.R. and D.Q.; investigation, L.F.R. and D.Q.; resources, L.F.R. and D.Q.; data curation, L.F.R.; writing—original draft preparation, L.F.R.; writing—review and editing, L.F.R. and D.Q.; visualization, L.F.R. and D.Q.; supervision, D.Q and C.C.; project administration, L.F.R., D.Q. and C.C.; funding acquisition, L.F.R. All authors have read and agreed to the published version of the manuscript.

Funding

Luis Felipe Rojas is supported by Grant ANID BECAS/DOCTORADO NACIONAL, Chile, No. 21211272.

Institutional Review Board Statement

The study was conducted in accordance with the ethical standards defined in the regulations of the Pontificia Universidad Católica de Valparaíso, Chile (protocol code BIOEPUCV-H 779-2024, date of approval: 4 June 2024), the Declaration of Bioethics and Human Rights of 2005 by UNESCO, and the ANID regulations for studies involving humans.

Informed Consent Statement

Informed consent was obtained from all subjects involved in this study.

Data Availability Statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.

Acknowledgments

The authors would like to thank all practitioners who were involved in the experiment for this study.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Beck, K.; Beedle, M.; Bennekum, A.V.; Cockburn, A.; Cunningham, W.; Fowler, M.; Grenning, J.; Highsmith, J.; Hunt, A.; Jeffries, R.; et al. Agile Manifesto. 2001. Available online: https://agilemanifesto.org/ (accessed on 1 September 2024).
  2. Felker, C.; Slamova, R.; Davis, J. Integrating UX with scrum in an undergraduate software development project. In Proceedings of the 43rd ACM Technical Symposium on Computer Science Education, Raleigh, NC, USA, 29 February–3 March 2012; pp. 301–306. [Google Scholar]
  3. de Oliveira Sousa, A.; Valentim, N.M.C. Prototyping Usability and User Experience: A Simple Technique to Agile Teams. In Proceedings of the XVIII Brazilian Symposium on Software Quality, SBQS’19, Fortaleza, Brazil, 28 October–1 November 2019; Association for Computing Machinery: New York, NY, USA, 2019; pp. 222–227. [Google Scholar] [CrossRef]
  4. Kuusinen, K.; Väänänen-Vainio-Mattila, K. How to Make Agile UX Work More Efficient: Management and Sales Perspectives. In Proceedings of the 7th Nordic Conference on Human-Computer Interaction: Making Sense Through Design, NordiCHI ’12, Copenhagen, Denmark, 14–17 October 2012; Association for Computing Machinery: New York, NY, USA, 2012; pp. 139–148. [Google Scholar] [CrossRef]
  5. Digital.ai. 16th State of Agile Report. 2022. Available online: https://info.digital.ai/rs/981-LQX-968/images/AR-SA-2022-16th-Annual-State-Of-Agile-Report.pdf (accessed on 1 September 2024).
  6. Lárusdóttir, M.K.; Cajander, Å.; Gulliksen, J. The Big Picture of UX is Missing in Scrum Projects. In Proceedings of the International Workshop on the Interplay between User Experience (UX) Evaluation and System Development (I-UxSED), Copenhagen, Denmark, 14 October 2012; pp. 49–54. [Google Scholar]
  7. Kikitamara, S.; Noviyanti, A.A. A Conceptual Model of User Experience in Scrum Practice. In Proceedings of the 2018 10th International Conference on Information Technology and Electrical Engineering (ICITEE), Xiamen, China, 7–8 December 2018; pp. 581–586. [Google Scholar] [CrossRef]
  8. ISO 9241-210; 2010 Ergonomics of Human-System Interaction—Part 210: Human-Centred Design for Interactive Systems. ISO: Geneva, Switzerland, 2019. Available online: https://www.iso.org/standard/77520.html (accessed on 1 September 2024).
  9. Schulze, K.; Krömker, H. A framework to measure User eXperience of interactive online products. In Proceedings of the ACM International Conference on Internet Computing and Information Services, Washington, DC, USA, 17–18 September 2011. [Google Scholar] [CrossRef]
  10. Roto, V.; Väänänen-Vainio-Mattila, K.; Law, E.; Vermeeren, A. User experience evaluation methods in product development (UXEM’09). In Human-Computer Interaction–INTERACT 2009: 12th IFIP TC 13 International Conference, Uppsala, Sweden, 24–28 August 2009; Proceedings, Part II 12; Springer: Berlin/Heidelberg, Germany, 2009; pp. 981–982. [Google Scholar]
  11. Experience Research Society. User Experience. Available online: https://experienceresearchsociety.org/ux/ (accessed on 9 December 2023).
  12. Krause, R. Accounting for User Research in Agile. 2021. Available online: https://www.nngroup.com/articles/user-research-agile/ (accessed on 9 December 2023).
  13. Persson, J.S.; Bruun, A.; Lárusdóttir, M.K.; Nielsen, P.A. Agile software development and UX design: A case study of integration by mutual adjustment. Inf. Softw. Technol. 2022, 152, 107059. [Google Scholar] [CrossRef]
  14. Agile Alliance. What is Agile? | Agile 101 | Agile Alliance. Available online: https://www.agilealliance.org/agile101/ (accessed on 1 June 2022).
  15. Sommerville, I. Software Engineering, 9th ed.; Pearson Education India: Chennai, India, 2011. [Google Scholar]
  16. Schwaber, K.; Sutherland, J. The 2020 Scrum Guide. 2020. Available online: https://scrumguides.org/scrum-guide.html (accessed on 9 December 2023).
  17. Maguire, M. Using human factors standards to support user experience and agile design. In International Conference on Universal Access in Human-Computer Interaction; Springer: Berlin/Heidelberg, Germany, 2013; pp. 185–194. [Google Scholar]
  18. Pillay, N.; Wing, J. Agile UX: Integrating good UX development practices in Agile. In Proceedings of the 2019 Conference on Information Communications Technology and Society (ICTAS), Durban, South Africa, 6–8 March 2019; pp. 1–6. [Google Scholar]
  19. Weber, B.; Müller, A.; Miclau, C. Methodical Framework and Case Study for Εmpowering Customer-Centricity in an E-Commerce Agency–The Experience Logic as Key Component of User Experience Practices Within Agile IT Project Teams. Lect. Notes Comput. Sci. Incl. Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinform. 2021, 12783, 156–177. [Google Scholar] [CrossRef]
  20. Argumanis, D.; Moquillaza, A.; Paz, F. A Framework Based on UCD and Scrum for the Software Development Process. Lect. Notes Comput. Sci. Incl. Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinform. 2021, 12779, 15–33. [Google Scholar] [CrossRef]
  21. Gardner, T.N.; Aktunc, O. Integrating Usability into the Agile Software Development Life Cycle Using User Experience Practices. In ASEE Gulf Southwest Annual Conference, ASEE Conferences. 2022. Available online: https://peer.asee.org/39189 (accessed on 1 June 2022).
  22. Atlassian. What is Agile? | Atlassian. Available online: https://www.atlassian.com/agile (accessed on 1 June 2022).
  23. Vermeeren, A.P.O.S.; Law, E.L.C.; Roto, V.; Obrist, M.; Hoonhout, J.; Väänänen-Vainio-Mattila, K. User experience evaluation methods: Current state and development needs. In Proceedings of the Nordic 2010 Extending Boundaries—Proceedings of the 6th Nordic Conference on Human-Computer Interaction, Reykjavik, Iceland, 16–20 October 2010; pp. 521–530. [Google Scholar] [CrossRef]
  24. Blomkvist, S. Towards a Model for Bridging Agile Development and User-Centered Design; Springer: Berlin/Heidelberg, Germany, 2005; pp. 219–244. [Google Scholar] [CrossRef]
  25. Salah, D.; Paige, R.F.; Cairns, P. A systematic literature review for Agile development processes and user centred design integration. In Proceedings of the 18th International Conference on Evaluation and Assessment in Software Engineering, New York, NY, USA, 13–14 May 2014; pp. 1–10. [Google Scholar] [CrossRef]
  26. Isomursu, M.; Sirotkin, A.; Voltti, P.; Halonen, M. User experience design goes agile in lean transformation—A case study. In Proceedings of the 2012 Agile Conference, Dallas, TX USA, 13–17 August 2012; pp. 1–10. [Google Scholar] [CrossRef]
  27. Sy, D.; Miller, L. Optimizing Agile user-centred design. In Proceedings of the Conference on Human Factors in Computing Systems, Florence, Italy, 5–10 April 2008; pp. 3897–3900. [Google Scholar] [CrossRef]
  28. Chamberlain, S.; Sharp, H.; Maiden, N. Towards a framework for integrating agile development and user-centred design. Lect. Notes Comput. Sci. Incl. Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinform. 2006, 4044, 143–153. [Google Scholar] [CrossRef]
  29. Rojas, L.F.; Quiñones, D. How to Evaluate the User Experience in Agile Software Development: A Systematic Literature Review. Submitt. J. Under Rev. 2024. [Google Scholar]
  30. Rojas, L.F.; Quiñones, D.; Cubillos, C. Exploring practitioners’ perspective on user experience and agile software development. In Proceedings of the International Conference on Industry Science and Computer Sciences Innovation, Porto, Portugal, 29–31 October 2024. [Google Scholar]
  31. Jordan, P. Designing Pleasurable Products: An Introduction to the New Human Factors, 1st ed.; CRC Press: London, UK, 2000. [Google Scholar]
  32. Nielsen, J.; Molich, R. Heuristic evaluation of user interfaces. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Washington, DC, USA, 1–5 April 1990; pp. 249–256. [Google Scholar]
  33. Grace, E. Guerrilla Usability Testing: How to Introduce It in Your Next UX Project—Usability Geek. Available online: https://usabilitygeek.com/guerrilla-usability-testing-how-to/ (accessed on 1 July 2022).
  34. Medlock, M.C.; Wixon, D.; Terrano, M.; Romero, R.; Fulton, B. Using the RITE method to improve products: A definition and a case study. Usability Prof. Assoc. 2002, 51, 1562338474–1963813932. [Google Scholar]
  35. Nielsen, J. Putting A/B Testing in Its Place. August 2005. Available online: https://www.nngroup.com/articles/putting-ab-testing-in-its-place/ (accessed on 1 July 2022).
  36. Nielsen, J. Summary of Usability Inspection Methods. 1994. Available online: https://www.nngroup.com/articles/summary-of-usability-inspection-methods/ (accessed on 1 July 2022).
  37. Brooke, J. SUS-A quick and dirty usability scale. Usability Eval. Ind. 1996, 189, 4–7. [Google Scholar]
Figure 1. Process of developing the framework.
Figure 1. Process of developing the framework.
Applsci 14 08991 g001
Figure 2. First version of FRAMUX-EV.
Figure 2. First version of FRAMUX-EV.
Applsci 14 08991 g002
Table 1. Characteristics and limitations of the related works reviewed.
Table 1. Characteristics and limitations of the related works reviewed.
AuthorsType of StudyStagesUX Evaluation MethodsUX Artifacts or UX Events
Felker et al. [2]Integration of elements to Scrum1. Scrum events1. Paper prototypes
2. Formative UX evaluation
3. RITE method
4. Brief-regular tests
Not mentioned
Maguire [17]Integration of elements to Human-Centered Design1. Context of use
2. User and organizational requirements
3. Design & prototype
4. Evaluation of designs
1. Heuristic evaluation
2. User walkthroughs
3. Labs tests/field trials
Not mentioned
Pillay and Wing [18]Integration of elements to Scrum1. Sprint 0
2. Scrum events
3. A UX cycle embedded in the sprint
Not mentionedUX event:
1. Lean UX cycle
Weber et al. [19]Integration of elements to Scrum1. Phase 1
2. Phase 2
3. Phase 3
4. Phase 4
1. 5-s test
2. (Un)moderated usability test
3. Cognitive walkthrough
4. Card sorting
5. Eye tracking
UX artifacts:
1. Documented ‘lessons learned’
2. User story map
Argumanis et al. [20]Integration of elements to Scrum1. Initiation Phase
2. Planning Phase
3. Implementation phase
1. Design Evaluation
2. User Test (Thinking aloud)
3. Pair Design
Not mentioned
Gardner and Aktunc [21]Sequential and/or iterative activities for agile in general1. Understand
2. Ideate
3. Decide
4. Prototype
5. Test
1. User testingNot mentioned
Table 2. UX evaluation methods selected for the first version of FRAMUX-EV.
Table 2. UX evaluation methods selected for the first version of FRAMUX-EV.
UX Evaluation MethodDefinition
Usability/user testingParticipants are asked to verbalize the thoughts that they have when experiencing a product or product concept. Where the item under evaluation is a finished product or an interactive prototype, participants may be set some specific tasks to do or may be given the chance for some free exploration of the product [31].
Heuristic evaluationInspection method that allows to find potential usability/UX problems in the design of a user interface. A small group of usability experts judge a user interface to check if it complies with the principles of usability design regarding certain heuristics [32].
Evaluation with mockups or prototypesIt consists of evaluating a mock-up or prototype by discussing it with a user and asking them questions about it.
Guerrilla testing“Guerrilla usability testing is a way to evaluate how effective an interface is by testing out its visual design, functionality and general message on its intended audience and capturing their responses. What makes guerrilla usability testing unique is that participants are not recruited in advance. Instead, members of the public are approached by those conducting the study during live intercepts in cafés, libraries, and malls, or in any other natural environment” [33].
RITE method“A RITE test is very similar to a ‘traditional’ usability test. The usability engineer and team must define a target population for testing, schedule participants to come into the lab, decide on how the users behaviors will be measured, construct a test script and have participants engage in a verbal protocol (e.g., think aloud). RITE differs from a ‘traditional’ usability test by emphasizing extremely rapid changes and verification of the effectiveness of these changes” [34].
A/B testing“Unleash two different versions of a design on the world and see which performs the best” [35].
Pluralistic walkthrough“Uses group meetings where users, developers, and human factors people step through a scenario, discussing each dialogue element” [36].
System usability scale (SUS)An instrument used to measure usability perception. It is composed of 10 Likert-scale questions and produces a score from 0 to 100 where individual item scores are not significant on their own [37].
Table 3. UX artifacts proposed for the first version of FRAMUX-EV.
Table 3. UX artifacts proposed for the first version of FRAMUX-EV.
UX ArtifactDescriptionHow to Use It
Applsci 14 08991 i001UX guidelines that include a list of all recommended methods for evaluating UX in agile projects. It includes a definition of each method, a detailed explanation of their application, steps to follow, expected results, and recommendations on when to use them.1. Choose UX evaluation methods to use in your project.
2. Select the most appropriate UX evaluation method for each experiment.
3. Follow the step-by-step process defined for each UX evaluation method.
4. Use the results to inform UX decisions.
Applsci 14 08991 i002UX guidelines that contain principles, reusable components, interactions, color palettes, and typography defined by the UX team to incorporate designs.1. Create a comprehensive guide of reusable components and design standards.
2. Share with the development team to ensure the feasibility of components.
3. Update regularly to incorporate new patterns or design components.
4. Use as a reference during design and development to ensure alignment and consistency.
Applsci 14 08991 i003UX guidelines that include the personas created by the UX team. It includes a template for creating personas, recommendations, and best practices, and a database with contact information of representative users.1. Collect user data through UX research.
2. Create personas that capture relevant information of representative users.
3. Use personas during design discussions to ensure the user’s perspective is considered.
4. Update personas when project goals evolve or new user segments are identified.
Applsci 14 08991 i004UX guidelines that contain all the UX work that should be done during the development of a project. It includes a detailed description of the activities to be performed, the estimated time required, the UX role that has the responsibility to perform them, and recommendations on the activities.1. Select the UX task to be carried out during the project.
2. Assign UX tasks to different UX roles.
3. Ensure that each UX role understands the specific responsibilities for each task.
4. Consult for additional information on the assigned UX tasks
Applsci 14 08991 i005UX document that includes the results and conclusions of UX evaluations. It includes details about the type of evaluation, its objective, questions/activities to be performed, number of participants, results obtained (anonymous), and conclusions.1. Document the results and findings after each UX evaluation.
2. Store detailed reports of each UX evaluation, including key metrics and outcomes.
3. Review to guide future design decisions.
4. Use as a reference for UX improvements and trend analysis over time.
Applsci 14 08991 i006UX artifact that contains a prioritized list of UX work required to complete an agile project (i.e., research, design, and evaluation), including UX acceptance criteria, definition of ready, and definition of done. If the project does not have a UX team, UX acceptance criteria, definition of ready, and definition of done should be included for each user story.1. List all UX tasks required for the project (research, design, evaluation).
2. Assign priorities based on user needs and project objectives.
3. Review and update the UX backlog at each sprint to consider new priorities and tasks.
4. Ensure UX tasks are aligned with the development backlog to maintain focus on UX.
Applsci 14 08991 i007UX artifact that includes the UX sprint goal, the UX backlog items selected for the sprint, and the plan for delivering them. If the project does not have a UX team, the UX work should be included in the sprint backlog.1. Select priority UX tasks from the UX backlog for the sprint.
2. Assign deadlines and responsible members to each of the selected tasks.
3. Update the status of UX tasks during daily meetings to track progress.
4. Review the completion of UX tasks during the sprint retrospective and adjust future planning accordingly.
Table 4. UX events proposed for the first version of FRAMUX-EV.
Table 4. UX events proposed for the first version of FRAMUX-EV.
UX EventDescriptionHow to Conduct It
Applsci 14 08991 i008Meetings between the development team and the UX team to validate and/or refine designs prior to sprint planning.1. Prepare detailed designs.
2. Present UX work for validation and feedback.
3. Address any feasibility concerns or technical blockers with the developers.
4. Confirm that UX acceptance criteria, definition of ready, and definition of done for each user story are clearly defined and agreed upon.
Applsci 14 08991 i009Meetings between the development team and the UX team to validate and/or refine designs prior to sprint review.1. Review all UX work and ensure it meets UX acceptance criteria.
2. Identify any remaining issues.
3. Collaborate with developers to make final adjustments to the designs based on the review.
4. Confirm all designs developed are ready for the sprint review.
Applsci 14 08991 i010Weekly meetings between the development team and the UX team to discuss UX issues. For example, the feasibility of designs, changes to the UX component guide, design inquiries, or developer solutions for minor bugs.1. Document ongoing design issues or questions.
2. Analyze the feasibility of upcoming UX tasks with developers.
3. Discuss updates to the UX component guide.
4. Collaborate with the team to resolve minor design or technical problems.
Applsci 14 08991 i011Weekly meetings between users and the UX team (development team, product owner, stakeholders, and managers are optional) to (1) ask questions about designs, user stories, priorities, etc. and (2) evaluate the proposed designs.1. Select the UX evaluation methods to use.
2. Schedule a meeting with users.
3. Evaluate the UX work.
4. Ask questions about future UX work.
Table 5. Practitioners’ profile and experience.
Table 5. Practitioners’ profile and experience.
IDEducational LevelYears of ExperienceRoles Undertaken
P1Master12UX Writer
P2Master15Consulting and Design Director, UX Researcher, UX Consultant, UX Lead
P3College education2UX Researcher, Conversational UX
P4College education12UX Writer
P5College education1UX Researcher, UX Designer
P6Master12Project Leader, Scrum Master
P7Master17Project Leader, Lead UX
P8College education1UX Researcher, UX Designer
P9Doctorate10UX Researcher, Project Leader, Academic Professor
P10College education3Full-Stack Developer
P11Master17UX Researcher, UX Writer, Back-End Developer, Full-Stack Developer, Product Owner, Scrum Master, UI Designer, Product Manager
P12College education5Project Engineer, Manager, Developer
P13College education8UX/UI Designer
P14Master16Development Team Leader, Scrum Master
P15Post Doctorate15Full-Stack Developer, UX Designer, Scrum Master, Academic Professor
P16Doctorate20Full-Stack Developer, Front-End Developer, Creative Technologist, Software Engineer 2
P17Master16Scrum Master, Agile Coach
P18College education5UX Researcher
P19College education16Lead UX, UX Designer
P20College education2Project Manager, Product Owner, Scrum Master, QA Producer
P21College education2UX Researcher, UX/UI Designer
P22Master14Agile Delivery Lead, Developer, Scrum Master, Agile Team Facilitator
P23College education1Product Owner
P24Post Doctorate22Product Owner, Academic Professor
P25Master10UX Researcher
P26College education1UX Designer, IT Auditor
P27College education1Front-End Developer
P28College education15Senior Software Engineer, Lead Front-End Engineer, Full-Stack Developer, UX Designer, Product Manager
P29College education1Android Front-End Developer
P30Master4IOS Developer
P31College education8Product Owner
P32Master5UX Researcher
P33College education8Senior Product Designer, UX/UI Designer, Product Designer
P34College education16Developer, Product Owner, UI Designer
Table 6. Survey results for dimensions D1 and D2 of UX evaluation methods.
Table 6. Survey results for dimensions D1 and D2 of UX evaluation methods.
UX Evaluation MethodsD1—UsefulnessD2—Ease of
Integration
MeanSDMeanSD
Usability/user test4.790.413.650.98
Heuristic evaluation3.820.833.530.99
Evaluation with mockups or prototypes4.260.964.120.73
Guerrilla testing3.971.033.350.92
RITE method3.500.792.850.99
A/B testing3.971.093.381.10
Pluralistic walkthrough3.351.042.911.00
System usability scale3.350.883.380.65
Mean3.88 3.40
Table 7. Survey results for dimensions D1 and D2 of UX artifacts.
Table 7. Survey results for dimensions D1 and D2 of UX artifacts.
UX ArtifactsD1—UsefulnessD2—Ease of
Integration
MeanSDMeanSD
UX evaluation methods4.150.743.850.86
UX design system4.121.073.740.79
UX persona3.820.973.500.93
UX responsibilities and roles3.291.143.910.79
UX evaluation repository4.180.803.590.78
UX backlog4.260.903.260.90
UX sprint backlog4.121.013.590.86
Mean3.99 3.63
Table 8. Survey results for dimensions D1 and D2 of UX events.
Table 8. Survey results for dimensions D1 and D2 of UX events.
UX EventsD1—UsefulnessD2—Ease of
Integration
MeanSDMeanSD
Pre-planning UX meeting4.091.003.940.95
Pre-review UX meeting3.851.103.970.87
Weekly UX meeting4.410.743.851.08
Weekly user meeting3.851.022.351.07
Mean4.05 3.53
Table 9. Additional comments on UX evaluation methods mentioned by practitioners.
Table 9. Additional comments on UX evaluation methods mentioned by practitioners.
ParticipantAdditional Comments
P2“It would be necessary to clarify the methodology you are thinking about for agile software development and what kind of teams you are going to have. The methods will eventually depend on that”.
P3“(Evaluation methods) must be used at the right time, always proportional to the problem to be addressed”.
P4“RITE is complex because you are improving from one opinion/experience, without considering more user observations to make a total improvement”.
P11“Each method is related to a stage and what you want to evaluate, not all of them work with everything”.
P12“A method based on guiding and supporting proposals is more effective because it helps to mitigate disagreements among the users”.
P15“I find this list quite interesting and complete. They are essential methods with which it is possible to conduct a complete evaluation of the usability and therefore of the user experience”.
P17“Evaluating mockups does not let you know what users do, but what they could do. They are useful for providing input to the development team”.
P19“Complicated question, as it depends on the type of system/flow/design you need to evaluate the method to use, but the user testing, evaluation of mockups or prototypes, or user interviews are effective and easy to estimate in time as a task”.
P23“I believe that the methods that work best are those that make the user interact directly with the implemented design”.
P24“(The evaluation methods mentioned) provides an excellent roadmap”.
P25“I believe that an appropriate combination of some of these can provide valuable information for the evaluation”.
P32“I consider that the most complex thing to do is the heuristic evaluations because of the time involved and that they do not go with any agile methodology”.
P34“Of all that exists, we use very little, which is bad for the final product”.
Table 10. Additional comments and future actions for UX artifacts.
Table 10. Additional comments and future actions for UX artifacts.
UX ArtifactAdditional CommentsFuture Actions
UX evaluation methods- “I recommend performing evaluations that involve UX roles and developers together to somehow relate and involve them so they can understand the results together and better understand the design constraints and possibilities”. (P15)The expert’s comment will not be considered. A method has already been proposed that focuses on evaluation by involving both UX roles and developers (pluralistic walkthrough). We also include methods that can be used independently by either UX roles or developers.
UX design system- “There was a lack of consideration of UI content inputs”. (P4)
- “The design system is a necessary tool in every project, but it will depend on its size if it should be taken as a project within an agile project”. (P19)
The comments of the experts will be considered. The UI components will be specified in greater detail in the final version. Additionally, we will include a consideration for each artifact to indicate that; depending on the team’s capacity, these could be specified in more or less detail.
UX personas- “I believe that the artifact related to UX personas is unnecessary, as user personas are developed based on interviews with actual users. Predefined archetypes may not accurately reflect the specific users we are working with”. (P5)The expert’s comment will not be considered. At no point did we indicate that archetypes or personas would be predefined. We agree that these should be created using UX research methods.
UX roles and responsibilities- “The roles and responsibilities artifact may not be very useful”. (P14)The expert’s comment will not be considered since the comment was made only once.
UX evaluation repository- “Evaluation repositories are useless, but because people do not have time in the work process to review and contemplate them”. (P2)The expert’s comment will not be considered since the comment was made only once.
UX backlog- “We initially maintained a separate UX backlog but encountered two issues: (1) the UX work wasn’t adequately visualized, and (2) the development team often diverged and handled the UX tasks independently. Therefore, I recommend against keeping separate backlogs. Now, all work is consolidated into a single backlog”. (P3)
- “Both the UX backlog and the UX sprint backlog could be included in the general backlog and sprint backlog of the project, they just need to be well specified to avoid confusion”. (P23)
The recommendations and comments of the experts will not be considered. We have already included a consideration indicating that if the project does not have a UX team, UX acceptance criteria, definition of ready, and definition of done for each user story should be included.
UX sprint backlog- “Both the UX backlog and the UX sprint backlog could be included in the general backlog and sprint backlog of the project, they just need to be well specified to avoid confusion”. (P23)The expert’s comment will not be considered since the comment was made only once. We have already included a consideration indicating that if the project does not have a UX team, UX acceptance criteria, definition of ready, and definition of done for each user story should be included.
Table 11. Additional comments and future actions for UX events.
Table 11. Additional comments and future actions for UX events.
UX ArtifactAdditional CommentsFuture Actions
Weekly user meeting- “The weekly user meeting seems unnecessary to me”. (P4)
- “The meeting with users would not be weekly”. (P20)
- “Testing with users takes time, so I do not consider that doing it weekly would bring many benefits or differences because if you consider the time in which users test, analyze the information and make changes could take more than a week. So, I think that testing once a month or after a certain number of changes in the interface would be more enriching”. (P26)
- “Given the time we have for the sprint we don’t get to have weekly meetings with users”. (P32)
The recommendations and comments of the experts will be considered. We will change the name of the event to User Meeting. In addition, we will define that the ideal frequency of the event would be one or two sprints. Finally, we will include a consideration that, instead of using weeks, the event could be held after a certain volume of work has been completed.
Pre-review UX meeting- “I don’t see the sense of a meeting with UX before the sprint review”. (P17)
- “The UX meeting before the review can be useful if there was no communication during the development to explain how it works before showing it in the review. But if there is constant communication, it may be unnecessary”. (P20)
- “While it is always useful to hold meetings, holding them just before the review may not be the best time in my personal opinion”. (P30)
Some recommendations and comments from experts will be considered. We will keep the event since only two comments refer to the fact that it could be unnecessary. In addition, we will define the ideal frequency of the event to be one or two days before the sprint review.
Pre-planning UX meeting- “If the backlog is defined from the beginning and is kept in a static way, this event is very unnecessary”. (P20)The expert’s comment will not be considered. In agile environments, the backlog is an artifact that must always be refined and updated, never remaining static.
Weekly UX meeting- “Perhaps a biweekly UX meeting would be appropriate to keep pace with the development of the sprint and reduce costs”. (P16)The expert’s comment will be considered. We will change the name of the event to “UX meeting”. In addition, we will define the ideal frequency of the event to be one or two weeks.
Table 12. Application of FRAMUX-EV components across different Scrum team scenarios.
Table 12. Application of FRAMUX-EV components across different Scrum team scenarios.
Scenario123
Description of rolesNo dedicated UX roles. Developers handle basic UX evaluation alongside development tasks.One UX role is part of the development team (three developers + one UX). The UX role collaborates closely with the developers on UX work.Three UX roles are part of the development team (six developers + three UX). The UX team works in close collaboration with developers, sharing UX responsibilities across multiple areas.
Which artifacts to use and how to use them1. UX evaluation methods: Developers use it to learn about UX evaluation methods that do not require the involvement of a dedicated UX role.
2. UX backlog: Developers manage UX tasks within the same backlog as development work.
3. UX sprint backlog: Developers plan UX tasks within the same sprint backlog as development work.
1. UX evaluation methods: The UX role uses it to identify appropriate UX evaluation methods based on available time and resources.
2. UX design system: The UX role uses it to define a basic design system with reusable components and interactions.
3. UX backlog: The UX role is responsible for updating the UX-related tasks, which are integrated into the general backlog.
4. UX sprint backlog: The UX role is responsible for tracking the UX tasks for the sprint, which are integrated into the general sprint backlog.
1. UX evaluation methods: The UX team uses it to identify appropriate UX evaluation methods based on available time and resources.
2. UX roles and responsibilities: The UX team uses it to clearly define the responsibilities of each member to ensure task completion and alignment with project objectives.
3. UX persona: The UX team uses it to create the personas based on research. Contact information for representative users is maintained and regularly updated.
4. UX design system: The UX team uses it to define a detailed system with reusable components, interaction patterns, and style guides.
5. UX evaluation repository: The UX team uses it to store and reference a wide range of results and conclusions from UX evaluations.
6. UX backlog: The UX team is responsible for detailing, prioritizing, and updating UX tasks, which are maintained in a dedicated UX backlog.
7. UX sprint backlog: The UX team is responsible for tracking and managing the UX tasks and goals for the sprint, which are included in a dedicated UX sprint backlog.
Which events to conduct and how to conduct them1. Weekly user meeting: Developers meet some users to conduct basic UX evaluation.1. Weekly user meeting: The UX role conducts user meetings to collect feedback.
2. Weekly UX meeting: The UX role meets with developers to discuss and resolve design and usability issues.
1. Pre-planning UX meeting: The UX team meets with developers to ensure that all design components are ready before the sprint planning.
2. Pre-review UX meeting: The UX team meets with developers to ensure that everything developed meets UX standards before the sprint review.
3. Weekly user meeting: The UX team conducts multiple user meetings to collect detailed feedback from representative users.
4. Weekly UX meeting: The UX team meets and collaborates with developers to discuss and resolve design and usability issues.
Which UX evaluation methods to useThere are UX evaluation methods that can be applied by developers as designing experiments or interpreting the results does not require advanced knowledge in UX design and/or evaluation. For instance, developers can conduct an “evaluation with mockups or prototypes” or apply a questionnaire like the “system usability scale (SUS)”.Most of the proposed methods can be used by some roles with UX knowledge. In addition to the UX evaluation methods mentioned in the previous scenario, the presence of a UX role allows the use of “heuristic evaluation”, “usability/user testing”, and “A/B testing”. Furthermore, since both a UX role and developers are involved, the “pluralistic walkthrough” can be considered.Any of the proposed UX evaluation methods can be applied regularly by the dedicated UX team, ensuring thorough feedback and evaluation of designs. Given the increased number of UX roles, “guerrilla testing” and “RITE method” have become viable options.
Table 13. Comparison between existing proposals and FRAMUX-EV.
Table 13. Comparison between existing proposals and FRAMUX-EV.
Felker et al. [2]Maguire [17]Pillay and Wing [18]Weber et al. [19]Argumanis et al. [20]Gardner and Aktunc [21]FRAMUX-EV
Components0 UX events, 0 UX artifacts, and 4 UX evaluation methods0 UX event, 0 UX artifacts, and 3 UX evaluation methods1 UX event, 0 UX artifacts, and 0 UX evaluation methods0 UX events, 2 UX artifacts, and 5 UX evaluation methods0 UX events, 0 UX artifacts, and 3 UX evaluation methods0 UX events, 0 UX artifacts, and 1 UX evaluation method4 UX events, 7 UX artifacts, and 8 UX evaluation methods
Representation modeNot includedNot includedDiagram (sprint 0 and sprint 1)Tables to explain each phaseBPMN diagramsFigure (questions and activities per stage)Diagram (artifacts and events) and tables to explain each component
Validation performedNot mentionedNot mentionedNot mentionedInterviews with 5 experts, case study (user testing)Software development project (four 1-week sprints)Questionnaires and interviews (not many details mentioned)Survey to 34 practitioners
Strengths(1) Integrates Scrum events with UX evaluation methods
(2) Uses iterative methods like RITE for rapid and frequent testing
(1) Focus on usability and user-centered design
(2) Use of UX methods such as user testing and heuristic evaluations
(1) Introduces the Lean UX cycle within the sprint
(2) Facilitates rapid and continuous iteration of design elements and UX evaluation
(1) Includes useful elements (user story mapping and lessons learned) for work planning
(2) Includes several UX methods (such as user testing)
(1) Includes evaluation, design, and implementation phases within the agile cycle
(2) Validation through a real case study (software development project)
(1) Proposes an iterative approach, facilitating recurrent UX evaluation
(2) Flexible approach that can be adapted to different agile environments
(1) Provides a clear structure with seven UX artifacts and four UX events
(2) Facilitates continuous UX evaluation throughout each sprint
Weakness(1) It does not include specific UX artifacts, limiting the formalization of UX evaluation
(2) It does not present a structure for integrating UX evaluation throughout the development cycle
(1) It is not adapted to agile frameworks (it is presented in a general way), which limits its applicability
(2) It does not include specific UX artifacts or events to manage work continuously in an agile manner
(1) It does not clearly define UX artifacts
(2) Lacks detail on how and when to conduct UX evaluations within agile events
(1) It does not provide a clear structure on how to manage UX work throughout iterations
(2) It does not include specific UX events, making it difficult to integrate UX into agile software development
(1) It does not include specific UX events or artifacts to integrate within agile development
(2) UX evaluation is performed in specific phases, which may delay user feedback
(1) Not specifically tailored for Scrum or other popular agile frameworks
(2) It does not include specific UX artifacts or events, limiting clarity and structure for integrating UX into the agile process
(1) Lacks validation in large-scale real-world projects
(2) The complexity of implementing some UX artifacts and events may pose challenges for teams without defined UX roles
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Rojas, L.F.; Quiñones, D.; Cubillos, C. FRAMUX-EV: A Framework for Evaluating User Experience in Agile Software Development. Appl. Sci. 2024, 14, 8991. https://doi.org/10.3390/app14198991

AMA Style

Rojas LF, Quiñones D, Cubillos C. FRAMUX-EV: A Framework for Evaluating User Experience in Agile Software Development. Applied Sciences. 2024; 14(19):8991. https://doi.org/10.3390/app14198991

Chicago/Turabian Style

Rojas, Luis Felipe, Daniela Quiñones, and Claudio Cubillos. 2024. "FRAMUX-EV: A Framework for Evaluating User Experience in Agile Software Development" Applied Sciences 14, no. 19: 8991. https://doi.org/10.3390/app14198991

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop