Vitech's Fund Analytics Manager enables fund managers and investors to build relevant data models to track their investments over time.
Type and Timeline
Product Design, UX, and Data Visualization.
September - November 2021.
Product Managers: Sharon Ironside, Christa Punturieri, and Shannon Carroll. UX Team: Emm Pakdee (Director of UX), Gautam Krishnan (Sr. UX Designer). Engineering (Framework) Team.
Figma Design, FigJam, InVision, Balsamiq Mockups, and React.js.
Ideation, Research, Prototyping, Visual Design, and User Testing.
Vitech broadly serves three industrial verticals - Insurance, Investments, and Retirement. Fund Analytics Manager is built for the investments vertical giving fund managers the power to manage the full life cycle of the fund, from fund creation to portfolio management, to investor allocations and communications. Fund capital transactions, fees and expenses, as well as investment gain/loss can be allocated through to build out each investor’s capital account, which allows for immediate access to an investor’s position.
Introduction to the Platform
Vitech's next-generation React platform is a robust framework that serves as a base for all our new self-service apps. We use React Material under the hood that is modified to serve Vitech's Design System components.
- To identify and prioritize use cases to validate the opportunities for the team to focus on.
- To learn about business requirements, user needs, and technology capabilities/constraints.
- Conducting competitive research, gathering inspiration, and exploring a visual design that sets up our customers for success.
A rudimentary version of what we set out to build existed in our Core Admin platform for many years. We spoke to a few customers and SMEs regarding this and found a lot of scope for improvement while building a new application. They have, at times, shown to be leading indicators for market needs. To note specific examples:
- Classifying and displaying funds has always been a challenging task. A given portfolio might have a mix of different kinds of funds, and a portfolio would often contain hundreds of funds.
- While running calculations on funds, we have had users reporting difficulties with the UX as the fields required to perform a calculation would change from one type of calculation to another.
- We had a demo request for the new platform from Bridgewater Associates, one of our largest clients.
- As we have been rolling out a lot of self-service applications based on the next-gen React platform, it seemed appropriate to build one for investors and fund managers.
- Fund managers often share portfolio data with their clients, partners, employees, and other stakeholders. We wanted to build a tool that would make this sharing of data simple and robust.
Research Phase: Competitor Research
We looked at various SaaS portfolio managers and were inspired by Quicken, Personal Capital, Yahoo Finance, and more.
Research Phase: Defining Success
Measures of Success
We will consider the product to be a success when:
- Fund managers are able to create their portfolios and perform calculations like Internal Rate of Return (IRR), Fund IRR, Time-Weighted Rate of Return (TWR), Return Over Time (ROT), etc. We will only focus on the four types of aforementioned calc types.
- Our existing customers switching from the Core Admin tool to the new Fund Analytics Manager. This would imply a base-level success as our users would be using the same product to perform the tasks that they have done using our older product.
- Some more metrics need to be further refined for early adopters and for when the product is generally available for all customers. This application will be offered as a standalone solution to new customers who aren't familiar with our current Core Admin tool.
- We've gotten positive feedback from user interviews confirming that our solution is at par or better than other tools they have used for the same function. This will be conducted with Bridgewater Associates.
Research Phase: User Flows
After interviewing our SMEs and client fund managers, we were able to further define the requirements and create epics to work on. We broke it down as follows:
Research Phase: Storyboarding
After user interviews, sketching user flows, and storyboarding, we were able to further refine the scope of the product.
Workflow (High Level)
- Ability to view and save calculated data as interactive grids and/or charts.
- Reduce clicks and multiple tabs. A user should be able to get to their desired results in as few clicks as possible.
- Allow switching to the Core Admin application while the Analytics screen is enabled.
- Ability to save and print a model.
- Save calculated data.
- Create/build a new analytics model to use the model whenever needed for running calculations from time to time.
- Create/add a new analytics model folder to the library of analytics models.
- Ability to edit/update my existing model.
- Review, edit and update existing models.
- Let users choose their calculation type (only IRR, Fund IRR, TWR, and RoI for now).
- Present multiple fund criteria for the user to select from.
- Manually run a calculation and save when necessary.
- Let users design their dashboards with charts, with the ability to print.
- Format dashboards and models to optimize for print.
- Let the user configure the content of the analytics dashboard screen for each of the saved analytics model(s).
Usability Testing - High Level Findings
"Very often design is the most immediate way of defining what products become in people's minds."
- Jony Ive
- Remote (in-situation) usability testing was conducted on a convenience sample of 4 test participants (2 internal and 2 external) from 09/10/2021 - 10/14/2021.
- Participants were shown the clickable prototype and instructed to perform three tasks and provide feedback on their experience interacting with the UX design concept/clickable prototype and potential features via the think-aloud method.
- The Internal sessions lasted approximately 10 -15 minutes and were recorded via MS Teams with both video and audio. The External session lasted 30 minutes.
- Following data collection, we performed content analysis by watching recorded test sessions to extract user feedback in aggregate and identify overarching themes.
Usability Testing - Findings Summary
We were able to validate a lot of our hypotheses and understood where we fell short.
- All 4/4 participants were able to build a new Investment IRR model.
- All 4/4 participants were able to navigate between models, home page, and the application settings.
- All 4/4 participants were able to edit a pre-existing model. They were able to change the investments and perform a different type of calculation when asked to.
- All 4/4 participants were able to identify the model they created most recently and navigate to it.
Our test participants used the following adjectives to describe their experience:
Usability Testing - Insights and Recommendations
Traffic Light Report Summary of Tasks Performed by Test Participants via UX Design Concept.
Model Creation Learnings
- Step 1: All test participants were immediately clear on how to select the IRR.
- Step 2: All participants understood what was required to select the desired entity and desired funds.
- Step 3: All participants understood what was required to select the desired investment filters.
- Step 4: All participants understood what was required to select the desired investments.
- Step 5: 2/4 participants easily understood what was required to select the Calculation parameters. 2/4 participants found the parameter screen confusing.
- Step 6: Participants valued the importance of reviewing what information is being displayed on the review & confirm screen for reassurance. All participants were immediately clear that they can edit all the steps.
Model Creation Recommendations
- Combine a couple of steps to reduce clicks and reduce the whitespace to reduce scrolling to select the continue button.
- All investments within a fund should be selected by default to reduce clicks.
- Move the chart type to the model view page. The user would want to slice and dice the data differently based on the chart type.
View Model Learnings
- 4/4 participants were able to easily view ad access the Saved Model Results.
View Model Recommendations
4/4 participants were able to easily access the dashboard, navigate to and between the models they created.
The drawer dropdown menu that allows switching between models should also be present on the home page.
Detailed High Fidelity Design