Trading / Risk Analysis Platform Design Sprint. UX Case Study

Thanks, disruptorleague.com/ for the photo

Black Mirror is a real-time trading platform that displays trading data to trading and support teams. This information provides a unique real-time overview of Company equities trading and is critical for the support and monitoring of the production environment.

My role: Product Designer / Design Sprint Facilitator

Design Sprint Planning

The Design Sprint is a proven methodology for solving problems through designing, prototyping, and testing ideas with users. Design Sprints quickly align teams under a shared vision with clearly defined goals and deliverables. Ultimately, it is a tool for developing a hypothesis, prototyping an idea, and testing it rapidly with as little investment as possible in as real an environment as possible: https://designsprintkit.withgoogle.com/ .

Design Sprint Goal

● Identify user pain points

● Discover new opportunities to increase user productivity

Design Process

My Design Sprint follows seven phases: Understand, Define, Sketch, Decide, Prototype, Validate, Launch.

These methods can help align a team on the right problem to solve or provide actionable solutions to a tightly scoped challenge that can be tested with users at a later date.

While it is recommended to include all the phases of the methodology. I worked on this project during a Covid-19 pandemic so I was running it 100% remotely and since I had a limited amount of time I decided to run shorter workshops utilizing these methods. Also, it was really hard to gather all the users together so I ran individual 1:1 sessions with each user.

I picked the methods that worked best for the goals we had.

Design sprint Team formation and Timeline agreement

Phase 1. Understand

HMW

HMW — “How Might We” is a starting point for most of the projects I start working on. The goal is to understand the opportunities of the project and it’s pain points. HMWs create an active framework for resolving the challenges.

As I mentioned earlier I ran this workshop remotely and I have been sending my users instructions on how to complete each exercise.

Task 1. HMW:

This exercise will help us to capture insight and pain points about using Trade Metrics within Two Sigma. This is a speed exercise, don’t spend too much time crafting and perfecting HMW statements with as little self-editing as possible. Write down as many notes as possible. It has to be short statements 1 sentence max. Aim for quantity over perfection.

HMW statement examples:

HMW: see more plots on a single screen

HMW: save plots templates that I always want to check

HMW: save popular search inquiries

HMW: make a table column resizable

HMW results:

HMW exercise typically requires a group of people collaborating in the same space. As I mentioned earlier this project happened in the time of Covid-19 so we all worked remotely. My job was to find the best collaboration platform. Since we are not able to use any not approved third party online services like Mural. I replaced it with simple Microsoft Word because it’s already approved by TS, and it is also more familiar to users.

HMW Insights From a User (sample)

Job Stories

This exercise helps everybody to understand why users use the app, and what goals they are looking to accomplish. Also, job stories are a great way to design for motivation. I personally find this exercise very handy when I jump into a new project and I need to quickly learn everything about the project.

Task 2. Job Stories:

This exercise will help us to understand your motivation to use Metrics on real world case scenarios from a user perspective. Please come up with 7–10 (the more the better) job stories following this template:

When ______ I want to______so I can

Example. When the market jumps on 3% or higher I want to check the performance of my related companies so I can find out the best time to sell them.

This example is very superficial, please be more specific with your job stories.

Job Stories; Insights from a User (sample)

• When there are big portfolio pnl swings I want to see if this is caused by trading or position pnl and comes from a particular sector/ticker so I can determine if manual intervention is required
• When there is earnings in a name i want to better monitor the trading and positional pnl of this and related names to see if we are on the wrong side of an informed move
• When we are trading i want to view how an instrument/portfolio position/gmv/outstanding compares to yesterday and historical days so i can determine if there is an anomaly
• When we are trading i want to know if we have stopped trading at the instrument/portfolio/entity level due to a hard or soft limit so i can determine if manual intervention is required
• When trading approaches the end of the day I want to monitor our risk and margin going into the close so I can make sure they are acceptable levels

User interviews

User interviews rely on stories to explore emotions, understand user goals, and assess needs. The goal of the interview is to gain empathy for the user’s experience.

Open questions

What are you working on?

How can you describe your role?

What is your day to day responsibilities?

Discovery questions

What is the future of trading?

What are your favorite metric apps you use for trading? (Within TS and outside TS)

What do you like about them?

Optimization questions

Which features are the most important for you when you check the metrics. Why?

How many traders also use these features?

What feature do you want to add to a TSS trade Metrics?

How many tactics in your arsenal?

How do you find information on which stocks belong to which subAccount?

How many subAccounts do we have in house?

How do you split the work between your peers? How do you know who is trading what?

How many tactics do we have and actively use?

Do we have some kind of documentation about each tactic that we have in house?

Interview Insights from a User (sample)

1. Display GSF report status (Number of gateway rejects) in a Metrics table + information about a reject so users don’t have to make an extra step to go to terminal to request a report
2. Display traded qty by venue. For example what amount of shares we traded on Nasdaq
3. Functionality to use the metrics aggregation for a customized portfolio, i.e. aggregating NetPnl for a custom portfolio.
4. Display validator values related to corresponded risk limits
5. Display all disconnected/aborted tactic + reason + total amount of aborted tactics

Usability Study

This method helps to identify usability issues. This involves observing the users attempting to complete a task or set of tasks while using the product. I sent a few scenarios to users and asked them to think out loud as they try to complete the task. Since we are not in the office we did it remotely by screen sharing

Usability study summary

• Limited amount of data can be displayed
• Not enough user customization
• Many of the user interactions are fiddly or buggy
• Cluttered information design that makes it difficult to know what to look at first
• Valuable features are undiscoverable

HMV Voting

After I summarized all the user insights together, my job was to figure out the priorities for the first milestone and clarify a long term plan for the project. I asked key users to help me with that. The process didn’t look like a recommended Google Sprint HMV Voting session, I just asked them verbally because it was pretty obvious for everybody that firstly we need to focus on performance and UI improvements and then on implementing new features. And frankly, for the first release, I wanted to make it as fast as possible to demonstrate to users a great potential of the UX so it will be much easier for me to recruit more users for the future Design Sprints.

Phase 2. Define

Success Metrics & Signals

• Deliver a rebuilt Metrics front end as defined in collaboration with existing Metrics users
• Acceptance of the project by the TSS trade desk; achieve feature parity and performance targets
• Retirement of the existing Metrics UI
• Bring Metrics up to date with the current web technology stack in TS

Design Principles

• Meet the users’ needs
• Consistency
• Accessibility
• Usability

Phase 3. Sketch / Wireframes

I skipped all the preliminary exercises of this stage and jumped directly to the last stage called Solution Sketch. I collected all the most important ideas from previous phases and came up with a polished, fully fleshed out idea for the solutions that are thought to be the best. I created a detailed Solution Sketch. A Solution Sketch included all states of the concept to help illustrate how it works, as well as a clear title to help people remember it.

Phase 4. Select

Because Metrics is a key part of TSS’s support story, we need to make sure that it meets traders’ needs. We’ll partner closely with users, soliciting feedback as development proceeds by use of mockups and demos.

After presenting our sketch to the users I still had some hesitations about some parts of the new UI so I decided to run a survey to better understand the users.

Phase 5. Prototype

Logo

Based on the survey, most of our users don’t really care about the logo and project name, but for me, Product Name is a big part of success. I wanted to set my product apart from other similar products. Also, I wanted to make people remember this Product easier so I came up with a creative and catchy name: “Black Mirror”.

Header Navigation menu

First of all the black color is a very strong color, I wear a marketing hat here. I wanted to deliver a message that this product is very reliable and has useful functionality. In my opinion, black can best deliver this kind of message to the user.

Also, I added the icons next to each menu item which makes our navigation more intuitive. Plus some additional “Slack” and “Help” icons for simpler excess

Before

After

Dashboard

Current UI doesn’t provide concise and fast information about blockers and other critical signals that require immediate actions. Dashboard tends to solve this problem.

Metric selection

Study showed that users very often switch between the Metrics, and it’s one of the most important instruments of the product, but current UI provides an inconvenient drop-down menu which makes users make extra clicks. Tabs tend to solve this problem to make the user experience more intuitive and UI cleaner.

Before

After

Search

Search is also a very important component of the product, users use it all the time so I decided to make it more prominent and move it to the right place. Also, the search bar needs a placeholder so user know which columns will be affected by search

Before

After

Filters

There are two ways users can apply filters. First is through a search bar and second, manually using a drop-down menu, the problem is that users didn’t even know they could make it through a drop-down menu so I decided to move each filter next to its column.

Before

After

Column preferences

Users will be able to set up column display preferences with an option to see which columns are hidden.

Row density

Research showed us that most of the users prefer compact row density so I decided to stick with this for now and potentially come up with two display options for future releases.

Action buttons

“Hide Table” and “Export” buttons are not of use too much so I decided to consolidate them into the icons.

After

Column width

Research showed that 57% of the users prefer column title width but it’s handy to have an option to resize some columns.

Plots

Cells selected to display on Plots must be highlighted so users can clearly understand which cells are displaying on the plots at the moment.

Pagination

Pagination and number of rows I moved at the bottom of the table which is more intuitive for users.

Workspaces

The idea behind workspaces is to improve user experience. For example, right now Jim

keeps all his search queries and filters in Excel spreadsheets so it’s very impractical to always keep Excel open and copy-paste search queries from Excel to the app. The concept of workspaces is to let users save their everyday operations within an app, it will save them tons of time. Also, they don’t have to manually enter all the search queries which are sometimes a pretty long process.

Black theme

Research showed that 43% of users prefer a black theme so I think it makes sense to give users options to switch between black and white themes

Phase 6. Test/Validate

Stakeholder Review

Unfortunately, our key stakeholders are very busy so they didn’t participate in the entire Design Sprint, but they did a final review to approve all the changes and to give us green light for further development. I didn’t participate in this stage of the design sprint, our project manager handled this activity.

Technical Review

Similar to the Stakeholder Review, the Technical Review is a Design Sprint method to secure buy-in from the engineer or technical lead for our product. Our front end developer participated through the entire design sprint so she was giving feasibility feedback the entire time so it was pretty fast and stressless.

Phase 7. Launch

The product is under development at the moment and as soon as it’s ready we have to track our success metrics and discover new opportunities.

Success Metrics & Signals

• Deliver a rebuilt Metrics front end as defined in collaboration with existing Metrics users
• Acceptance of the project by the TSS trade desk; achieve feature parity and performance targets
• Retirement of the existing Metrics UI
• Bring Metrics up to date with the current web technology stack in TS

Takeaways

It was my first experience applying Google Design Sprint in the organization and I am very glad I was able to do so. I feel like this method is much more practical in terms of collaboration. Also, it really helps everybody to stay on the same page during the process, this is crucial for me as for a Product Designer especially in the financial field where I have to learn a lot of new things on the fly. Google design sprint was a new experience for my users as well, and I am glad they loved it. I felt they were so enthusiastic about it.