A Customer Analytics Implementation: Understanding Users at Huntsman Research Informatics Shared Resource

Data analytics and user dashboards are tools that have a bigger role in the world of data than ever before. User dashboards allow users to better interact, visualize and access data, while data analytics is just beginning to expand the possibilities of how user data can be analyzed and utilized, and resources allocated. At Huntsman Cancer Institute the RISR team need for these tools has been a driving force for quality improvement and allowing product owners to further upgrade their products and workflows. To continue to do this they are proposing the design, development and implementation of Microsoft Power with BI or business intelligence with user dashboard to pull and visualize user data from their applications. This project will employ many different methods and frameworks to discover the needs and requirements of the dashboard, support its development, and future implementation. A literature and product review will be conducted, along with stakeholder interviews to discover the needs and requirements. The proposed dashboard will be built using Scrum Methodology and User Centered design principles as discovered by the needs assessment. This will allow more user input at the time of development and allow the dashboard to be built faster with a more usable finished product. An analysis of the utility and needs of Microsoft BI dashboard for RISR will be proposed for design and implementation. Product research of Microsoft BI will be presented, a table of user requirements will be documented along with completed stakeholder interviews. An example of a Microsoft BI dashboard and accompanying architecture will be available along with all relevant information for potential implementation. The research, design, and implementation of a Microsoft BI dashboard with business analytics will greatly improve the RISR teams' ability to support researchers, data, and their applications with a visualization tool and accompanying data.

1 A Customer Analytics Implementation: Understanding Users at Huntsman Research Informatics Shared Resource Stacy Wells College of Nursing: The University of Utah NURS 6881: Master’s Practicum and Project Capstone II 05/03/2023 2 Abstract Background: Data analytics and user dashboards are tools that have a bigger role in the world of data than ever before. User dashboards allow users to better interact, visualize and access data, while data analytics is just beginning to expand the possibilities of how user data can be analyzed and utilized, and resources allocated. At Huntsman Cancer Institute the RISR team need for these tools has been a driving force for quality improvement and allowing product owners to further upgrade their products and workflows. To continue to do this they are proposing the design, development and implementation of Microsoft Power with BI or business intelligence with user dashboard to pull and visualize user data from their applications. Methods: This project will employ many different methods and frameworks to discover the needs and requirements of the dashboard, support its development, and future implementation. A literature and product review will be conducted, along with stakeholder interviews to discover the needs and requirements. The proposed dashboard will be built using Scrum Methodology and User Centered design principles as discovered by the needs assessment. This will allow more user input at the time of development and allow the dashboard to be built faster with a more usable finished product. Results: An analysis of the utility and needs of Microsoft BI dashboard for RISR will be proposed for design and implementation. Product research of Microsoft BI will be presented, a table of user requirements will be documented along with completed stakeholder interviews. An example of a Microsoft BI dashboard and accompanying architecture will be available along with all relevant information for potential implementation. Conclusions: The research, design, and implementation of a Microsoft BI dashboard with business analytics will greatly improve the RISR teams' ability to support researchers, data, and their applications with a visualization tool and accompanying data. Keywords: User dashboard, Analytics, Scrum, Microsoft BI 3 A Customer Analytics Implementation: Understanding Users at Huntsman Research Informatics Shared Resource Problem Description Research informatics is a relatively new field of biomedical informatics and has recently begun to be adopted by more academic institutions. The National Center for Research Resources has stated that the future of biomedical research will involve collaborations by many scientists in diverse locations, linked through high-speed computer networks that enable submission, analysis, and sharing of data (National Institutes of Health, February 28, 2023). However, the need to collect and share data in a secure manner with numerous collaborators across academic departments or even institutions is a formidable challenge (Harris et al, 2009). This need also began to arise at Huntsman Cancer Institute and in response the Research Informatics Shared Resource or RISR Department is designed to help support genomics and research applications (Post, n.d.). RISR efforts to improve quality revealed the need to collect information regarding client data use in order to understand client behaviors and assist in making crucial business decisions. They have no basic user statistics with which they can make actionable decisions with. To remedy this problem the design and implementation of a user analytics dashboard with BI (Business Intelligence) powered by Microsoft has been proposed to bring in analytics to better support applications, users, and the department. Available Knowledge A dashboard is a data-driven clinical decision support tool capable of querying multiple databases and providing a visual representation of key performance indicators in a single report. The utility of a dashboard comes from its ability to provide a concise overview of key information (Khairat et al, 2018). Dashboards are used in many different professions and fields, including the medical field and research. Dashboards are used to monitor and share performance measures, which will enable decision-making and such demands pressure healthcare leaders to invest in projects and tools that support performance 4 improvement and enhance the quality of care provided. One of these tools is the visualization dashboard (Alhabib et al., 2020). Dashboards can be designed and displayed in many different ways and are designed based on the information that will be presented on them. Discovering the needs of the dashboard users and collaborating with them to discover the best format to visualize data was a requirement of this project. Visualization tools can be anything from charts, graphs, maps, text, and pictures. Color schemes and layout are also important concepts of dashboard design. Big data is defined as large complex data sets from new and multiple data sources. These data sets are so voluminous that traditional data processing software can’t manage them (Oracle, n.d.). Genomics and research data fit into this category and have teams dedicated to supporting them and their applications. These data sets are large and difficult to manage and store, not to mention analyze. The data that needs to be stored is often the size of a single data set from a few terabytes (TBs) to tens of trillion bytes or Petabytes (PBs) (Yang et al., 2017). The high and varying data content of these files often makes storage expensive and difficult; many on-site servers are needed just to house the data. Therefore, most of the companies have utilized the services of the cloud where they do not have to keep such huge storage and computation resources in-house (Velte et al, 2010). The value of big data relies on the ability to analyze it. The task of the RISR team is to support the researchers and applications who must use and store this data. Often the information they need to do this is complex and requires the use of analytics to meaningfully understand and troubleshoot issues across multiple applications. Analytics refers to the process of identifying, interpreting and communicating meaningful patterns of data. Business analytics refers to applying this process to answer business questions, make predictions, discover new relationships and, ultimately, make better business decisions (Anodot, n.d.). The RISR team wish to accomplish better support for their users by analyzing and better managing their users' needs. Implementing Microsoft BI with this dashboard will be able to help them improve the quality and understanding of their number of users and applications. Analytics generally have been performed by the users themselves or by statisticians and data analysts who are in short supply. This 5 meant it was extremely time-consuming and could not be scaled unless resources or time could be extended or altered. BI or business intelligence analytics based on machine learning can now improve accuracy, speed, and scale which opens more data for interpretation in real time. Following the Gartner Analytic Ascendancy model the RISR team will be able to have predictive analysis to trouble shoot future problems and optimize their workflow and products. See Figure 1 for the Garner Analytic Ascendancy model. The first version of this dashboard will be based on the first two stages of the Analytic Ascendency Model. This means the first stage of analytics is hindsight-based and asks the analyst to determine what has already happened in the data and Diagnostic analytics. The next stage is more insightdriven and asks the analyst to identify why a particular event or change in the data occurred (Anodot, n.d.). This matches with the initial information needing to be displayed on version one of this dashboard. As the RISR team improves dashboard functionality with subsequent dashboard version releases, they can explore more relationships in data analytics and even begin to have foresight in their dashboard to anticipate and predict problems before they happen. Costs associated with implementation and purchase of a product will be mentioned in general terms as the cost of purchasing the product has not been finalized. Anticipated costs include the product Microsoft BI itself; licensing, implementation, and maintenance fees can be expected. Contracts will be negotiated for these fees. No anticipated cost for Hardwares such as servers will be needed as this is a web-based application that can be used on existing servers including cloud servers. All the pricing in cloud storage is based on a flat rate of storage and the usage rate of the network. Therefore, the operations that are being performed or requested for the data stored over cloud storage play an important role in this regard (Tahir er al., 2020). Projects are prioritized based on the strategic planning of the Cancer Center leadership, including the HCI Director's Cabinet and Internal Advisory Board (Post, n.d.). In RISR, all product decisions are made by the RISR director. 6 Rationale The current RISR project is being run using a project management framework called Scrum framework and the proposed dashboard will also be developed using this method. Scrum is an agile project management framework that helps teams’ structure and manage their work through a set of values, principles, and practices. These values allow teams to achieve their product goals and deadlines efficiently (Drummond, n.d.). This framework is considered iterative and incremental, allowing products to be created faster and with more stakeholder and user input during development. The RISR team uses Scrum to have product output with consistent product improvement. This pushes the product along incrementally and has helped create future directions and needs such as the user dashboard. See a diagram of Scrum framework in Figure 2. Scrum teams are defined by values and roles and have many strengths and weaknesses. The Scrum Team consists of one Scrum Master, one Product Owner, and Developers. Within a Scrum Team, there are no sub-teams or hierarchies. It is a cohesive unit of professionals focused on one objective at a time, the product goal (Schwaber and Sutherland, 2020). Scrum has advantages and disadvantages with its use. Please see Figure 3 for a pros and cons chart of Scrum methodology. Values are followed that bind the scrum team together and guide their work include courage, focus, commitment, respect, and openness (Scrum.org, n.d.). User-centered design (UCD) is a methodology used by developers and designers to ensure they’re creating products that meet users’ needs (Lowdermilk, 2013). These principles can guide the design of the mockup and help keep the focus on the needs and goals of the project. Please see figure 4 for UCD methodology. User center design principles from the Upwork website include (Upwork, 2022). • Design for user’s experience and tasks • Empathetic design • Keep it consistent 7 • Importance of requirements clarification • Use simple language • Involve adequate feedback loops • Adequate navigation mechanisms • Reduce user efforts • Iterative design process • Error free system Project management principles will also be relevant and applicable as this will be proposed to directors for consideration and implementation. Project management flow includes analyzing, evaluating, designing, and developing. This capstone project will focus on the planning stage only. Further steps will be carried out by the product owner and development team at RISR. Project management principles are relevant and must be taken into consideration when developing and implementing a new workflow and product. This will benefit the project with better control of resources, budget, time, and scope. Specific Aims The aim of this project is to analyze, design, and construct an implementation plan to integrate customer analytics in existing HCI RISR applications. This will involve building a dashboard to help with better utilization, visualization, and analysis of user data at HCI RISR The objectives will be to analyze requirements and goals, conduct a needs assessment that includes stakeholder interviews, build a BI Dashboard, and design and construct an implementation plan. Context Huntsman Cancer Institute is part of a larger academic medical center called the University of Utah Health. Huntsman Cancer Institute’s mission is to understand cancer from its beginnings, to use that knowledge in the creation and improvement of cancer treatments, to relieve the suffering of cancer patients, and to provide education about cancer risk, prevention, and care (Huntsman Cancer Institute, 2023). This goal is to improve knowledge and patient care lay the foundation for improving multiple 8 departments’ ability to better use and analyze their data through the user dashboard. Most of this project was conducted via distance but many on-site visits to design, collaborate and validate the product were also needed. Various programs will be used to assist in the build and design such as Microsoft BI, Adobe XD, and Core Browser. The RISR team develops and supports computing and information systems that allow researchers to accomplish research objectives by providing database, web, and enterprise software development (Post, n.d.). Collaborating with this team and its resources helped to direct my project, gain an understanding of their workflows, and help accomplish the goals and objectives. In turn this proposal will allow them to support their researchers, as well as themselves, with a dashboard that has BI capabilities. There are many different roles on this team such as the product owner, scrum master, developers, and designers. All roles are vital to a scrum team and will be needed as the project progresses into the production stage. The RISR team runs their current project in an iterative scrum method with weekly meetings to collaborate and develop. The main stakeholder is my preceptor and product owner of Core Browser. The current state of client data analytics at RISR is very limited. There is no analytical software in place and the team must manually go through logs to extrapolate data, and trace user activity. They must rely on records from human resources or their applications to know anything about their users. The current state of genomic storage at Huntsman Cancer Institute is the Amazon Web Services or AWS file storage system. The AWS system is reported to be cumbersome and difficult to use. Many discussions have focused on the system not meeting the product owner’s needs and data analytics is very limited. They do not feel confident in giving researchers or other clients business information or statistics. The proposal of Microsoft BI and its associated dashboard are being considered as fixes to these issues and to supplement current applications. The dashboard will be built within the Microsoft BI program, which will also pull the data into the dashboard. The information technology infrastructure will play a dynamic and significant role in the project itself. Many IT decisions and designs will need to be made based on the infrastructure. The current set-up of programs like Core Browser is a full-stack system that is part of a microservices environment. Full 9 stack development is described as front and back-end design. This means client and server development can be done by the same developers. Most of the developers on the Core Browser team and future dashboard build are considered full stack developers. This will be very useful in helping design the dashboard to specification and programming the dashboard to pull the correct information. Microservices are applications that are separated into a collection of small, independently deployable services. Because microservices are designed to act independently, they are naturally consistent with agile principles that promote end-to-end team ownership (Santana et al., n.d.). Microservices are important in the project because Microsoft BI will be able to pull data from the microservices architecture to the dashboard to display relevant information, or more hopefully the applications themselves will send Power BI an ETL file that can be uploaded nightly. ETL, which stands for extract, transform, and load, is a data integration process that combines data from multiple data sources into a single, consistent data store that is loaded into a data warehouse or other target system (IBM, n.d). The applications are also in this architecture because they are easier to deploy and build in the scrum methodology. Each microservice is built and aligned around a business function to reduce the complexity of the application change-management process. Because each service is individually changed, tested, and deployed without affecting other services, time to market is accelerated (Santana et al, n.d). This also allows each individual application to be scalable and not affect other applications' performance. Please see figure 5 for traditional architecture vs microservices architecture. Interventions Through research and stakeholder interviews an understanding of the requirements of the dashboard has been established and created. A literature and product review were done to help guide understanding and questioning, and to narrow the focus and design to better suit the needs of the users. Project management methodology was also used to guide the collecting of information to be presented to the stakeholders. Information collected ranged from technical, design, implementation, requirements, needs, and product information. 10 Stakeholder interviews were conducted in a semi-structured manner. Some questions were predetermined and room for general discussion was included. My preceptor helped to engage stakeholders and arranged for interviews in-person and distance. The setting was informal, and Nancy helped to describe her vision for the project to the team members, I was able to ask more specific questions about what they wanted to see and would require. Interviews were conducted with four members of the RISR team, all holding various roles and providing multiple perspectives. I was able to interview a product owner, developer, database engineer, and director. They were chosen to be interviewed for their roles and direct impact, and by direction from my preceptor. User-Centered design was a guiding principle in the creation of our mock-ups and needs analysis. User-centered design is an iterative design process in which designers focus on the users and their needs in each phase of the design process (Interaction Design Foundation, n.d.). Due to time constraints, mockups were delayed and examples of previous healthcare dashboards are available to convey the idea of the possibilities of the visualizations. Microsoft BI was chosen to be the analytics system that would be purchased to fill this need. Microsoft Power BI is a business intelligence (BI) platform that provides nontechnical business users with tools for aggregating, analyzing, visualizing and sharing data. Power BI's user interface is intuitive for users familiar with Excel, and its deep integration with other Microsoft products makes it a versatile self-service tool that requires little upfront training (Scardina et al, n.d.). A debate between Microsoft Power BI and Tableau products was a factor but Microsoft Power BI was settled upon because of its ease of use and ability to integrate between already existing Microsoft platforms that are in use. Reviews of this product were done to understand the platform and how it functions and provides data visualization. There are prebuilt templates that can be manipulated to create the dashboard with Power BI, so developers do not have to design from scratch. The mock-up of the dashboard was done with the software program Adobe XD and within Microsoft BI itself. User-centered design principles were applied along with instruction from team members who have previous design experience with mock-ups. 11 Study of the Intervention(s) The approach used to assess the impact of the interventions will be based on the product meeting the needs and requirements set forth by the stakeholders before and during development. With Scrum methodology testing and development can happen at the same time. Requirements will be defined for each sprint and worked on until completed. The users can then test the product and give real-time feedback if they approve or want changes. The completed dashboard data will be validated for all user fields. After implementation usability surveys can be conducted to determine improvements or changes in workflow. Visualization of and reliability of data can be measured and validated against known information and logs. With the implementation of the software and accompanying dashboard, the ability to have more user data may develop new metrics for future use in other projects. Measures Needs and requirements for the dashboard came from stakeholder interviews. Questions were defined (see Table 1), but room for general topic discussion was made available to ensure an organic conversation. Questions were specifically aimed at allowing stakeholders to expand upon the issues and hopes of what the new dashboard could accomplish. The questions were aimed to fill about 15 minutes to respect interviewees’ time and gain concise understanding. These questions will also be asked again after implementation to ascertain if objectives were completed and define future versions of the dashboard. When choosing criteria for the usability evaluation of dashboards, the study objectives, dashboard features and capabilities, and context of use should be taken into consideration (Almasi et al, 2023). Dashboards often rely on surveys of the users to measure and validate their design; this project will also use user surveys to validate the design. Surveys such as SUS, Data, and TAM could be used. The survey questions will be aimed at measuring the end-users’ perceived satisfaction with the following attributes: usage frequency, system quality (ease of use, speed, and stability), system information quality (data relevance and accuracy), impacts on work efficiency, impacts on quality of care (effectiveness and safety), and overall user satisfaction (Alhabib et al, 2020). This initial version of this dashboard will be available only to limited users, survey numbers are expected to be small in the 5–10-person range. The 12 limited number of people using this dashboard can allow for in-person follow-up and emailing of surveys. Given the role of dashboards in the decision-making process and the multiplicity of questionnaires, it can be challenging to select a proper questionnaire for the usability evaluation framework of dashboards (Almasi et al, 2023). Other measures will be needed to validate the dashboard including validating the metrics and fields such as making sure that active users are defined and pulling correctly into the charts and visualization tools. This will be done internally with the development team during development and implementation. User testing can be done during Scrum sprints and will need to continue periodically as part of the maintenance of the application. Implementation measures and benchmarks will also need to be defined and addressed to ensure a smooth transition and implementation of the dashboard. Analysis This dashboard will be used by a small number of product owners. Interviews and survey data will need to be reviewed and interpreted from this small sample size. Five stakeholders were interviewed and consisted of product owners, directors, engineers, and developers. It can be acknowledged that both quantitative and qualitative methods play a significant role in technology development and progress. While quantitative methods have some advantages, such as cost-effectiveness and higher suitability for studies with a large sample size, qualitative methods (e.g., think-aloud) are beneficial for providing details about problems to which quantitative methods do not commonly apply (Almasi et al, 2023). While qualitative data will be more usability and feasibility-based quantitative measures can be more concrete and data-driven such as statistical analysis on survey results and actual data results such as the number of users can be gathered. Ethical Considerations Even with improved accuracy and better visualization of the data, there are still ethical concerns to consider. Key ethical concerns raised by Big Data research include respecting patients’ autonomy via the provision of adequate consent, ensuring equity, and respecting participants’ privacy (Howe et al, 13 2020). As mentioned before multiple collaborators across multiple sites may be accessing and using data, how do we make sure and maintain appropriate permissions in such a vast network. Making sure correct inferences are made from data and data is validated can prevent errors and inferior business outcomes. Confirmation bias, availability heuristics, Simpson’s paradox, and nonnormality biases must all be addressed and understood when working with data. Making sure there is an appropriate understanding and use of data to improve workflows and outcomes will further data validation and confidence in the dashboard. Results The questions asked of stakeholders are shown in Table 1. Main stakeholders and interviewees include product owners (2), RISR director, developer, and a designer. A review of their answers provided the needs requirements reported in Table 2. Many stakeholders stated the same information which culminated in basic user statistics that they could give interested business clients, and analytics of how users are interacting with their applications. Requirements included application utilization such as the number of logins, peak login time, as well as the number of new and active users. A review of the proposed product Microsoft Power BI was conducted to ensure that requirements from a technical and user aspect could be met and understood. Power BI will fit into the current microservices architecture and will be able to pull data from numerous applications in the architecture. Please see Figure 6 for a proposed diagram of data flow in and out of Power BI. There are numerous ways to pull data from each application or source, but the goal is to have data pulled from all services in the architecture. The anticipated goal is for an ETL file to be pulled into Power BI nightly to update the dashboard and its visuals. For technical aspects of the product itself, Power BI is a cloud-based application that requires minimal server space. See Table 3 for specific technical specifications. Power BI integrates with all other Microsoft products such as excel, teams, etc., which made it a desirable option in their system. The functionality of this software allows for ease of use to create the visualizations on the dashboard once the information has been pulled into the software, all that is required is a basic understanding of excel according to Microsoft. Due to time constraints, two examples of potential 14 prototypes have been included for reference of what the anticipated dashboard could resemble see Figure 7 and Figure 8. The implementation of the dashboard will be timed in conjunction with many milestones. First, it will need the approval of the proposal and then the purchase of Microsoft BI. At the time of this project, the last completed discussion was that Microsoft BI would be approved. There has been continuous change and no official decision between Tableau and Microsoft BI. This affects the implementation timeline and how the mock-ups would look and even some functionality of what they are trying to accomplish. Next purchase of the product and contracts would need to be approved per Huntsman procedures. The decision for which product is chosen is outside the project scope and will be needed to accurately portray the abilities and visualizations of the dashboard. Core Browser implementation will be the priority and needs to be completed before this project can be initiated by the team. The proposed implementation is estimated to be in late summer 2023. This may cause delays or barriers to the project if Core Browser is delayed or needs more support than anticipated after its implementation. The anticipated estimate of the actual implementation of the dashboard is roughly 1 year in total with completion anticipated Summer of 2024. With this just being a proposal only broad goals will be referred to most design elements, decisions, timelines, and changes will be done during development in the scrum process. Discussion Summary The RISR team at Huntsman Cancer Institute wants to restructure how data used for research is analyzed and stored. They want to implement a dashboard to improve visibility into patterns of data use. User requirements were investigated through semi-structured interviews. Vended solution capabilities were researched and the RISR team is proposing to use Microsoft BI to help implement analytics and the proposed dashboard. An implementation and evaluation plan has been developed. Interpretation This dashboard with its BI capabilities will be expected to improve workflows and give a better understanding of user needs in the research applications to their product owners. The team should gain an 15 understanding of the number of users they have and the traffic their applications get. This will help them better allocate their resources. With the BI capabilities and later versions, they will begin expanding and be able to see trends to make predictions of how their services and users are functioning. The possibility of expansion really is unlimited and as they design, more potential can be unlocked. With the acceptance and implementation of the Microsoft BI platform, more usability between these applications will allow more meaning fully understanding and data. The Microsoft system has potential to expand and help all users gain insight and then be able to easily share that information across multiple applications. The ease with which they can pull data from multiple sources should be greatly increased and make the users more productive and data easier to visualize. Continual validation will be a maintenance activity that will need to be performed often because there are so many different applications there are being continually updated. Definition of authorized users and user activity will also need to be reviewed often in order to validate who users are and the permissions they can have. I do foresee that some type of education will need to be developed in using the application itself, not just the dashboard. Many insights have been gleaned from this project. Working and collaborating with an agile Scrum team is about using a problem-solving mindset and open communication. Since collaboration and deadlines are constant, you must be able to problem solve and perform at a constant high level. In these teams, you must also have an engaged leader and product owner, without these the work will not progress in a timely manner and a scrum team can fall apart quickly and the project can get behind or be put in jeopardy. Scrum projects must keep the end goal in mind and have very clear instructions for each individual piece of the process, without clear instructions you can’t lead your developers or team to satisfactory product completion. The feasibility of this project is rather appealing and manageable, and overall sustainability is rather high. What I found most difficult to complete were the user needs and requirements. I had to have multiple talks and interviews to narrow down what could be viable for the initial version of the dashboard. The scope and vision of this project were rather large and being able to narrow it down took a lot more time than anticipated. Being able to speak with someone who had good insight really helped me focus on 16 the immediate needs. I also found it difficult being able to put this information into visuals that made sense and could be easily interpreted. Being able to actively observe RISR Scrum meetings helped me understand how this application would be beneficial and developed. Working with the team gave me great insight as to how this product will be created within the scrum team and it's pleasing to know that user input will be a main developer of the product itself. Tackling this project in sprint increments will be valuable in creating a final product that will meet and hopefully surpass all needs and expectations. After development and implementation, the real capabilities of the BI dashboard may begin to improve understanding of users and workflow and continue to grow in capabilities and quality of the products and services under the RISR team. Limitations This project lays the foundation for the future proposal to build and implement this dashboard. Limitations do include the rejection of this proposal outright for various reasons from cost to other priority projects. This dashboard does rely on the implementation and purchase of Microsoft BI platform. As previously mentioned, there were multiple products considered that could be used instead of Microsoft BI and there is still some uncertainty regarding the final product that will be selected. Getting good usability feedback from the dashboard could be difficult and many iterations may need to be built to have good usability. Usability surveys often rely on people taking and giving good input and feedback. Getting good visualizations that are meaningful and able to make improved business recommendations can be difficult as this would be an entirely new program and workflow for users. The project would be under strict time limitations and must be designed with deadline parameters. There are many data limitations that could be possible in this dashboard such as biased or misleading data that is not interpretable. Can patient care recommendations come from the data alone or must it be viewed in context with other research and evidence? Big data limitations such as viability and usability and privacy concerns are often brought up. The question often asked is big data able to be managed and not generalized in negative ways? There will be some limitations for the information being pulled into the dashboard from some programs due to universal logins on applications. This will limit the ability to see 17 user logins per application. Some applications do not keep logs of user information so there may be limited information that may be even pulled from certain applications. Another limitation is the definition of a user. There is no technical or business definition of a user and it may be different per application or institution. Being able to define a user within the system may be a difficult limitation. Conclusions Throughout this process, I’ve learned to be patient with what I don’t know but to continue to ask questions and reframe the questions until a satisfactory acquisition of learning has happened. During the project the importance of how data can be used has been continually impressed upon me, and that we are just scratching the surface of what is possible. Supporting future data endeavors and workflows will continue to be a high priority. Dashboards really have the ability to grow and support future data visualization and have a high impact on multiple professions. RISR product owners will now have the possibility of adding data analytics into their workflow and be able to make more concise business decisions. 18 Acknowledgements I would like to thank the RISR team at Huntsman Cancer Institute for all their direction and advice, as well as for allowing me to shadow their department and attend their meetings. I would especially like to acknowledge and thank Nancy Ho RN, MSN for her guidance and support as my preceptor and Chris Macintosh RN, PHD for the confidence boost and support. 19 References Almasi, S., Bahaadinbeigy, K., Ahmadi, H., Sohrabei, S., & Rabiei, R. (2023). Usability evaluation of dashboards: A systematic literature review of tools. BioMed research international, 2023, 9990933. https://doi.org/10.1155/2023/9990933 Alhabib, D., Alumarn, A., & Al-Rayes, S.A. (2020). Emergency room visualization dashboard user satisfaction in Saudi Arabia. Informatics in Medicine Unlocked, 21, 100493. Anodot. (n.d.). AI analytics. Retrieved March 15, 2023, from https://www.anodot.com/learning-center/aianalytics/ California Department of Technology. (2017). 3.8.1 Agile tools and techniques. Retrieved March 21, 2023 from https://projectresources.cdt.ca.gov/agile/agile-tools-and-techniques/ Drummond, Claire. (n.d.). Scrum a guide to scrum: How it works, what is it, and how to start. Atlassian. Retrieved on December 1, 2022, from https://www.atlassian.com/agile/scrum Fachat, Andre. (January 30, 2019). Challenges and benefits of the microservice architectural style, part 1. IBM. Retrieved on January 20th, 2023 from https://developer.ibm.com/articles/challenges-andbenefits-of-the-microservice-architectural-style-part-1/ Harris, Taylor, R., Thielke, R., Payne, J., Gonzalez, N., & Conde, J. G. (2009). Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics, 42(2), 377–381. https://doi.org/10.1016/j.jbi.2008.08.010 Howe Iii, E. G., & Elenberg, F. (2020). Ethical challenges posed by big data. Innovations in clinical neuroscience, 17(10-12), 24–30. Huntsman Cancer Institute. (2023) About us. https://healthcare.utah.edu/huntsmancancerinstitute/about-us IBM. (n.d.). ETL (extract, transform, load). Retrieved April 6th, 2023, from What is ETL (Extract, Transform, Load)? | IBM Interaction Design Foundation. (n.d.). User centered design. Retrieved February 15, 2023, from https://www.interaction-design.org/literature/topics/user-centered-design 20 Khairat, S. S., Dukkipati, A., Lauria, H. A., Bice, T., Travers, D., & Carson, S. S. (2018). The impact of visualization dashboards on quality of care and clinician satisfaction: Integrative literature review. JMIR Human Factors, 5(2), e22. https://doi.org/10.2196/humanfactors.9328 Lowdermilk, & EbscoHost. (2013). User-centered design (1st edition.). O'Reilly. National Institutes of Health. (February 28,2023) The future of biomedicine. National Institutes of Health. Retrieved on April 10, 2023 from https://www.nih.gov/about-nih/what-we-do/nih-turningdiscovery-into-health/future-biomedicine Oracle. (n.d.) What is big data? Retrieved on December 2, 2022, from https://www.oracle.com/bigdata/what-is-big-data/ Pradhan, Ritika. (March 27, 2020). Product management methodologies: Scrum vs agile vs kanban. Udacity. Retrieved on March 3, 2023, from https://www.udacity.com/blog/2020/03/productmanagement-methodologies.html Post, Andrew. (n.d.) Research informatics. University of Utah Huntsman Cancer Institute. Retrieved on March 21, 2023, from https://risr.hci.utah.edu/aboutus.html Santana, C, & Roland, B. (n.d.). Improve time to market with microservices. IBM. Retrieved on November 17th, 2022, from https://www.ibm.com/cloud/architecture/architectures/microservices Scardina, J., Horwitz, L. (n.d). Microsoft power BI. Tech Target. Retrieved on March 29th, 2023, from https://www.techtarget.com/searchcontentmanagement/definition/Microsoft-Power-BI Schwaber, K. & Sutherland, J. (2020) The 2020 scrum guide. Scrum Guides. Retrieved on 3/15/2023, from https://scrumguides.org/scrum-guide.html Scrum.org. (n.d.). What is scrum? Scrum.org. Retrieved on December 1, 2022, from https://www.scrum.org/learning-series/what-is-scrum Sisense. (n.d). Retrieved on March 29th, 2023 from Hospital Performance Dashboard to Improve Operations I Sisense 21 Tahir, A., Chen, F., Khan, H. U., Ming, Z., Ahmad, A., Nazir, S., & Shafiq, M. (2020). A systematic review on cloud storage mechanisms concerning e-healthcare systems. Sensors (Basel, Switzerland), 20(18), 5392. https://doi.org/10.3390/s20185392 Upwork. (February 8, 2022). The key principles of user center design (UCD). Upwork. Retrieved on January 25, 2023, from https://www.upwork.com/resources/user-centered-design-principles Velte, A. T., Velte, T. J., Elsenpeter, R. C., & Elsenpeter, R. C. (2010). Cloud computing: A practical approach. Yang, C.-T., Liu, J.-C., Chen, S.-T., & Lu, H.-W. (2017). Implementation of a big data accessing and processing platform for medical records in cloud. Journal of Medical Systems, 41(10), 1–28. https://doi-org.ezproxy.lib.utah.edu/10.1007/s10916-017-0777-5 22 Tables and Figures Table 1 Survey Questions Developed Questions In an ideal world what would you like to see from this dashboard? What specific data are you looking for? How will you use this data? What barriers do you anticipate? How do you currently gather the data you need and what does that process look like? 23 Table 2 RISR Dashboard Requirements Data Category Time Periods Version 1 Requirements # of Active Users Daily, Weekly, Monthly, Annually, Per Defined Period # of New Users Daily, Weekly, Monthly, Annually, Per Defined Period Application Utilization Daily, Weekly, Monthly, Annually, Per Defined Period Total Amount of Users in Applications Daily, Weekly, Monthly, Annually, Per Defined Period Future Version Requirements Total # of users Daily, Weekly, Monthly, Annually, Per Defined Period Total # of active users Daily, Weekly, Monthly, Annually, Per Defined Period Total # of inactive users Daily, Weekly, Monthly, Annually, Per Defined Period Total # of new users Daily, Weekly, Monthly, Annually, Per Defined Period Total # of user in each application Daily, Weekly, Monthly, Annually, Per Defined Period 24 Table 3 Microsoft BI Requirements Component Requirement 4.8 If the server doesn't have internet access, you can manually install the .NET Framework from Microsoft .NET Framework 4.8 (Offline Installer) for Windows. .NET Framework For more information, recommendations, and guidance about the .NET Framework 4.8 see .NET Framework Deployment Guide for Developers. Windows 8.1, and Windows Server 2012 R2 require KB2919355 before installing .NET Framework 4.8. Power BI Report Server requires a minimum of 1 GB of available hard-disk space. Hard Disk Addition space will be required on the database server that is hosting the report server database. Minimum: 1 GB Memory Recommended: At least 4 GB Minimum: x64 Processor: 1.4 GHz Processor speed Recommended: 2.0 GHz or faster x64 Processor: AMD Opteron, AMD Athlon 64, Intel Processor type Xeon with Intel EM64T support, Intel Pentium IV with EM64T support 25 Windows Server 2022 Windows Server 2019 Datacenter Windows Server 2019 Standard Windows Server 2016 Datacenter Operating system Windows Server 2016 Standard Windows 10 Home Windows 10 Professional Windows 10 Enterprise Windows 11 26 Figure 1 Gartner Analytic Ascendancy Model (Anodot, n.d.) 27 Figure 2 Scrum Framework (California Department of Technology, 2017) 28 Figure 3 Scrum Pros & Cons (Pradhan, 2020) 29 Figure 4 Phases of User Centered Design (Interaction Design Foundation, n.d.) 30 Figure 5 Microservices Architecture (Fachat, 2019) 31 Figure 6 Data flow from Microsoft BI 32 Figure 7 Microsoft BI Healthcare Dashboard Example (Sisense, n.d.) 33 Figure 8 Power BI Example of Healthcare Dashboard 2 (Microsoft, n.d.)