User Experience/User Interface (UX/UI) design as the backbone of web-based climate service development
Abstract. Design practices are often undervalued in climate service development. We combine a review of 55 operational web-based climate services with User Interface (UI) and User Experience (UX) design literature, using a systematic coding framework to assess how design practices are reported. The analysis identifies recurring gaps between recommended design methods and those documented in climate service development, including the near absence of formal usability testing (1 of 55 tools) and limited reporting on web accessibility (4 of 55). Projects mostly treat design as a secondary layer to refine conventional scientific visuals, overlooking core design workflow elements such as user research, iterative prototyping, and usability testing. We outline practical steps for integrating UI/UX design throughout the development process, helping teams plan design activities, work with designers, and manage expectations across development stages. Two case studies illustrate the workflow in practice. Climate and Water supports water managers assessing climate change impacts on water resources. POSUKHA provides drought forecasts for farmers. Together, they show that while design workflows can follow shared principles, visual and interaction solutions must remain tailored to specific user contexts. We show that UI/UX design is an end-to-end process to ensure usability, trust, and informed decision-making.
General
The paper raises an important issue. Although decades of research have recommended user-centred approaches to climate service development, many services are still developed primarily around the available data. Even when projects involve users, they may lack a deliberate strategy for when and how to engage them and how to use their input to inform design decisions. Publications also often provide little detail about the methods used and what they revealed. The authors argue that design workflows should be documented as transparently as data sources and models.
Â
I found the analysis in the first part of the paper particularly valuable. The review of how UX/UI design practices are applied and documented in web-based climate services makes, to my knowledge, a novel contribution. In particular, quantifying how often user-centred design practices are documented and discussing patterns in what is reported could help funders and project managers set clearer requirements for teams to plan and resource these activities, carry them out rigorously, and document both their methods and findings.
I found the second part of the paper, beginning with Section 3.3, less valuable in its current form. The guidance would be more useful if it addressed the gaps identified in the review more directly and explained when and why particular methods are appropriate. The case studies could then show how those methods were applied and which design decisions they informed.
I would welcome a fuller discussion of the findings and of the factors that may explain why particular practices were adopted. For example, did the authors observe differences by project size or type of funder? Where publications did not mention a practice, did the authors consider contacting developers or project managers to establish whether it had been used? If they decided against doing so, what informed that decision?
The authors appear to treat unfamiliarity with UX/UI methods as the main barrier to their adoption. Other factors may be equally or more influential, including budget constraints, the priority the research and development team gives to development, and reluctance to change established ways of working. Time can also be a constraint, particularly when service development begins towards the end of a research project. I would encourage the authors to explain how they expect the proposed guidance to change current practice and to discuss the conditions that would support its use.
Sections 3.3 and 3.4 bring together useful literature on practical steps, recommendations and design principles. Given the breadth of the literature, a comprehensive overview would be beyond the scope of this paper. However, the current overview omits important techniques (e.g. task analysis, techniques in user testing). The sections also give too little practical guidance on how to design, plan  and conduct user research and usability tests, and how to analyse and interpret the results so that they provide a reliable basis for design decisions.
These sections could be reframed as a reflective overview of user-centred design, with a corresponding shift in tone. The repeated use of imperatives and phrases such as ‘you should’ makes the sections read like a set of instructions. A more explanatory style would help readers understand why particular methods are useful, when they are appropriate, and what it takes to apply them rigorously. Links to the gaps identified in Table 1 and references to more detailed methodological guidance would give readers a basis for deciding which methods to explore further.
The two case studies illustrate how UX/UI design can add value even when standard visualisation guidelines are available. Their contribution would be clearer if the authors showed what the case studies reveal about the use of those specific guidelines in practice, beyond reinforcing the importance of user research. The introduction already makes a strong case for tailoring designs to users’ needs, so the research question, ‘Can visual solutions for key components of web-based climate services be meaningfully standardized, or must they be tailored to specific contexts and audiences?’, could be made more informative by asking how far the visualisation guidelines discussed and applied provided a sound basis for initial prototypes, and where and why those prototypes needed to be adapted to meet users’ needs.
The case study descriptions would also benefit from more detail about the user research methods, particularly which methods and tasks were used for the prototypes testing, and who participated in the testing. Reporting these details alongside the findings and the resulting design changes would make the case studies concrete examples of the transparent reporting the paper advocates.
Finally, the paper would benefit from distinguishing more clearly between the design of individual visualisations and the design of a climate service as a whole. A web-based service brings together visualisations, explanatory text and datasets. Its design shapes how users navigate between these elements, interpret the information and use it in their work. For instance, the title and line 63 refer broadly to web-based climate service development, while the introduction focuses primarily on visualisations. Clarifying the scope in these places would help readers see which findings and recommendations apply specifically to visualisation design and which apply to the wider service.
Title:
The title reflects the paper’s broad topic, but its scope could be more precise. ‘Web-based climate service development’ suggests that the paper examines the design of the service as a whole, while much of the introduction and practical guidance focuses on visualisations. The title should make that emphasis clearer.
Abstract:
Is clear.
Overall presentation
The overall structure is clear.
Language
The language is generally fluent, but from section 3.3 onwards a more explanatory tone would also suit the introductory scope of the guidance.
Section 1, Introduction
The introduction draws on relevant studies to make the case for user research in the development of web-based climate services.
I would welcome a clearer distinction between the design of individual visualisations and the design of web-based climate services as a whole. It is currently unclear whether the review examined practices specific to visualisation design, practices used in developing the wider service, or both.
Section 2, Methods.
The methods are well thought through, clearly described and carefully applied.
The authors make an important point that the findings reflect how transparently design methods are reported. A follow-up questionnaire to product owners could have helped verify the review findings and identify methods that teams used but did not document in their publications.
Section 3. Literature review
To my knowledge, this is the first review to quantify how user-centred design practices are reported in publications on web-based climate services. Table 1 is particularly valuable because it shows which activities are least often documented. Where teams carried out these activities, the results point to gaps in reporting. Where teams omitted them, the results point to gaps in practice and raise questions about the priority given to user-centred design.
Section 3.1. Patterns in reported practices
This paragraph contains useful insights and starting points for improving development practices. Â
Could the authors use the data they collected to explore whether reported practices vary by project type or size, or by the composition of the development team? Did the reviewed publications identify any barriers to carrying out planned user-centred design activities? In practice,there are many barriers, for instance it can be difficult to recruit users and keep them involved, align the timing and content of design activities with research and data production, and accommodate changes to established working practices within the development team.
I welcome the attention to accessibility and inclusivity, both of which are often overlooked.
Section 3.1.6 also makes an important point about the asymmetry in transparency. Reporting how services were designed and evaluated, alongside documenting data sources and models, would help readers assess the quality of the work and inform the development of future services.
Section 3.3. Design workflow
The workflow is useful as an introduction. Applying it in a project requires a more detailed understanding of the time and expertise each activity demands, how the activities fit together, and how to gain the team’s support. I suggest presenting this section as an explanatory overview and directing readers to more detailed methodological guidance.
For example, the statement that ‘typical sample sizes range from 5 to 100’ gives teams little basis for deciding how many interviews to conduct. It would be more useful to explain how that decision depends on the purpose of the interviews and the range of users whose perspectives the study needs to capture. Conducting and analysing interviews can require substantial time and resources. The workflow could also explain how to involve users and stakeholders beyond the initial interviews and who will coordinate their participation and follow up on their input throughout development.
 Section 3.4.1
The authors should clarify under what circumstances a single persona is sufficient.The number of personas should reflect meaningful differences in users’ goals, tasks and contexts, both across target groups and within groups that initially appear homogeneous.
Scenarios, use cases and task descriptions are essential components of the design workflow. They should be grounded in user research and developed with input from users. They help teams design prototypes around what users need to accomplish and define realistic tasks for usability testing.
Section 3.4.2
The guidance on interviews is too brief to help readers formulate questions that elicit useful evidence. For instance, ‘How can the solution become useful and easy to use?’ combines two broad questions. Observation could also be included as a way to understand users’ tasks and working contexts.
Section 3.5. User interface
Section 3.5 summarises considerations for interface design, and Appendix B introduces the underlying principles. Readers who are new to these principles would need more explanation, worked examples or references to detailed guidance to apply them effectively.
The workflow covers identifying the target audience, conducting interviews and designing the interface. It should also address prototyping and methods for testing and evaluation. This is especially important given that the literature review found that almost none of the sampled publications reported user testing. The paper could explain which testing methods suit different questions, what expertise is needed to conduct them rigorously, and how teams can analyse and interpret the findings so developers can act on them.
Sections 3.6 and 3.7
These sections focus on visualisations, so the heading of 3.6 should reflect that scope.
As with the earlier design guidance, the visualisation principles and guidelines in Sections 3.6 and 3.7 are basic, incomplete (for instance of uncertainty visualisation) and often too general to guide design decisions. Readers new to visualisation may find the topics worth exploring, but they would need more explanation and worked examples to understand when and how to apply the principles effectively. Experienced designers are likely to find little here beyond what they already know. The cited publications offer more detailed guidance. I suggest shortening the overview, avoiding the imperatives, and explaining what each key reference can help readers with.
If the aim is to explain how the guidelines informed the case studies, I suggest discussing them alongside the visualisations they helped shape. Which guidelines provided a sound basis for the initial prototypes? Which design choices required the most revision after user testing, and why? Showing where the findings supported or prompted adaptations to the guidelines would clarify what the case studies contribute.
Section 4 Results: case studies
4.1 Climate & Water – web tool for water managers
4.1.1 User research. The main text describes a survey, while Section 3.3 presents interviews as a principal method and Appendix K indicates that interviews were also conducted. Please clarify which methods were used, how many interviews were conducted and how they were carried out. Surveys and interviews can answer different questions, so it would be helpful to explain why each method was chosen and what it contributed to the requirements.
Lines 444 and 452 refer to a single persona. How was this profile derived from the user research, and why was one persona considered sufficient? Even within a group such as water managers, differences in tasks and requirements may warrant two or three personas.
The authors state that the persona helped connect decisions about default parameters, navigation and visualisation formats to the planning needs of water managers. Please give concrete examples of decisions that the persona informed.
Section 4.1.2 describes how existing services and visualisation practices informed the designs and how these were subsequently tested with users. Clarifying the persona’s role would show how these sources of evidence shaped the design. Were scenarios, use cases or task descriptions developed to guide prototyping and testing?
4.1.2 User testing and conclusions. Lines 489–490 mention user testing but give little detail about the method. Were participants asked to explore the tool freely, complete predefined tasks or think aloud? How long did the sessions last, and how closely did the participants reflect the intended users? Appendix K suggests that testing took place online and in a group. Please confirm the format and explain whether it was chosen for methodological or practical reasons.
The conclusions beginning at line 520 would be clearer in a separate paragraph. The preceding description of how uncertainty was displayed leaves it unclear whether water managers helped choose the format. Please explain their role and, if their feedback influenced the choice, indicate what changed as a result.
 4.2 POSUKHA
Background and scope. This case study needs a clearer account of why POSUKHA was initiated, who provided the data, who developed the service and how the target audiences were identified. Was the service intended for all farmers in Ukraine or for particular groups? The POSUKHA website also lists audiences such as civil society organisations, ecologists and managers. Please clarify how their different needs were addressed.
4.2.1 User research and design. Please explain why a survey was chosen to elicit requirements and whether interviews were considered. Report how many people responded, how they were recruited and which intended user groups they represented. The paper advocates transparency in reporting design methods, and this case study could demonstrate what that looks like in practice.
As in the Climate & Water case, please explain why a single persona was considered sufficient and show how it informed design and development. Which of the challenges identified in Oleksandr’s profile led to specific design decisions? Were scenarios, use cases or task descriptions developed to translate user requirements into features or to guide prototype testing?
Prototype testing and evaluation. The account describes how user research informed the requirements, but leaves it unclear whether the prototypes were tested and evaluated. Testing is an essential part of the workflow presented earlier in the paper. If testing was conducted, please report who took part, what they were asked to do, how the findings were analysed and what changes followed. If it was omitted, please explain why and discuss what this means for the conclusions that can be drawn from the case study.
 Figure 13. The colours used to indicate drought and excess water may be difficult to distinguish for users with red–green colour vision deficiencies, including deuteranopia and protanopia; so would not meet the requirements for web accessibility.
Concluding statement. In the sentence beginning ‘Together with Climate & Water, this supports the conclusion…’, I suggest referring to 'visualisation principles' The sentence could read, ‘Together, the two case studies suggest that standard visualisation principles can inform initial designs, while the resulting visual and interaction solutions need to be adapted to the intended users and context.’
Discussion
5.1
The point made in lines 577–582 is important and should open the paragraph.
The statement that ‘user scenarios drive design decisions’ highlights a gap in Section 3.3. User scenarios describe what users are trying to accomplish and the context in which they work. The methodology should explain how task analysis and scenarios inform requirements, guide design decisions and provide a basis for realistic prototype testing.
At line 584, hiring a designer appears to have worked well in this project. The discussion should clarify what made that arrangement effective and under what conditions it could work elsewhere. As line 600 suggests, implementing user-centred design requires the support and active involvement of the project manager and the wider team. This is particularly important when a service draws on complex research data. The designer needs to work closely with users, researchers and developers to understand users’ needs, assess what the data can support and develop appropriate solutions.
Line 593 notes that one study used survey findings, while interviews with target users in one of the case studies led to different visualisation choices. This raises a methodological question worth addressing in Sections 3.3 and 4. What did each method reveal, and how did those findings inform the design? Please clarify whether the differences in design can be attributed to the methods used or may also reflect differences in users, tasks or contexts.
The finding that there is ‘no one-size-fits-all solution for visual design in web-based climate services’ is unsurprising. Farmers and climate researchers, for example, use climate information for different tasks, and those differences have implications for visualisation design. The difficulty in practice is that researchers and developers may assume they already know what users need, that a single service can meet the needs of several audiences, or that users will invest time in learning to use it and ultimately find it useful. Requirements analysis and user testing give teams a way to examine these assumptions and revise their design decisions in light of the findings.
The observation in line 613 that user research also influenced which content was included deserves more attention. In data-driven projects, the available data often determine what a service presents.
.5.4 Limitations
Lines 655–657 do not identify a specific limitation.
The conclusion states that ‘the needs and tasks of the target audience must drive every stage of the climate service design process’. The main text should give task analysis, user scenarios and use cases the same prominence. Please explain how these are developed with user input, how they are shared with developers, and how they inform requirements, design and prototype testing.
The conclusions place considerable emphasis on unfamiliarity with user-centred design methods. They should also acknowledge the practical and organisational barriers that can prevent teams from applying these methods, even when they know how to use them. These include demands by project funders, limited time and budgets, difficulty recruiting and retaining users, project schedules that leave little room for iteration, and reluctance to change established ways of working. Funders and project managers can help address these barriers by allocating time and resources for user research and testing throughout development and requiring teams to report how the findings informed design decisions.
Note: I appreciated the opportunity to read this paper. I found the analysis of reported design practices particularly valuable. By showing which activities are least often documented, the review gives researchers, developers, project managers and funders a concrete basis for improving how user-centred design is planned and reported. I look forward to seeing how the paper develops.