From downtime to
LEAD UX DESIGNER
We set out to find good, sustainable work, and achieved a contract with Morgan Stanley to redesign the UI of the Mutual Funds Order Entry System.
I was a Sr. UX Designer at the agency Idean.
At the time, Idean NYC was facing a shortage of work, and had no Managing Director or Head of Studio.
I stepped up, reached out to Biz Dev. from the west coast who provided us with several RFP's they had received. I reviewed them and decided to focus on this one, because of the size of the opportunity, and my expertise in Financial Services.
We delivered the new Order Entry UI.
Early metrics showed a fast adoption curve, with organic acquisition of 64% of internal users in less than a month.
The biz group we worked with was asked to continue re-designing the Order Entry Interface for several other Asset Classes.
Morgan Stanley trades billions of of dollars every day through the order entry interface we designed. This was VERY mission critical.
Because of it's success, we obtained a contract to further develop a universal design language for all of their Order Entry interfaces.
Creating a new Order Entry workflow for Morgan Stanley’s internal trading ecosystem.
Our designs would initially support Mutual Funds and Options, but should be flexible enough to incorporate Equity and Futures in the near term.
The supplied business specifications charter was one of the most challenging technical documents I’ve ever had to pitch against.
The Morgan Stanley platform was a classic Organic Growth enterprise beast. You could track the fads throughout the decades it was built.
Eventhough the system architecture was ‘accidental’, and the overall user experience poor, we didn't focus our pitch on this, but rather on the missed opportunities and the monetary impact of this situation:
We proposed a resarch driven design audit, with a hard technical focus, and accionable output, but we also made sure our visual design chops showed through in the RFP answer.
With the pitch won, and the project underway, we commenced with a very intense On-site research process that involved hours of demo’s, technical documentation reviews, contextual inquiry, and candid conversations.
By the time we were ready for stakeholder interviews in the trading floors, we were well known in the halls an cubicles of One New York Plaza, and had established our credibility.
Every broker’s desk was littered with Post-it’s and ‘hacks’ which were very insightful.
Being behind in tooling sets a bad tone with traders who deal with extreme high pressure
We quickly found that the ‘modern’ systems are a UX nightmare, inducing errors and frustration.
So many dropdowns, box hunting and things to click-on really slowed down the process.
We discovered that a great deal of Options trading was still being done through paper, because of it’s ease to complete and hand off.
Many CSA’s greatly preferred the elegance of the ‘green screen’ terminal over the Win200 style trading UI. Constraints can be good.
It didn’t take long to asses the current state of technology, and all of it’s shortcomings. Brokers and CSA’s are candid people as well.
In just a few weeks of intense observation, and ethnographic research, we had enough information to understand the most common user routines, and how these mapped (or didnt) with the systems.
No one wants a lecture about everything their company does wrong. So we focused on the things that we could address, and filtered them by cost vs impact. After this, we classified our future efforts based on these four principles.
Although generally overstated, a critical design process piece is fostering collaboration with the client. Being able to work through internal politics, ego’s and agendas is always a challenge.
We hosted whiteboarding sessions with the clients in our own office, that has a much more informal feel, and everyone could safely chill and think about how to make this monster work.
Guided with the Pareto Principle, we set out to find the top 20% of tasks used by 80% of users.
We had noticed very specific routines for different sets of individuals, but then these routines had higher ‘meta-routines’ that most traders went through, and because they were unaccounted cross-silo flows, they created a lot of friction.
Product Design Process
Once we achieved consensus, and were able to elicit a design direction, we started sketching UI artifacts.
Starting with whiteboard sketches, we slowly evolved into low fidelity wireframes to test our initial hypothesis quickly.
The first, and broadest concept we tackled, was to reduce the friction when transacting from one ‘mode’ to another, for example:
Launching an order entry from the household portfolio
Launching an order entry from the household portfolio
Launching an order and seeing populate the queue seamlessly
Reverting an in-queue order back to it’s ticket form
Test and Learn
Early testing of this concept got a lot of internal traction, so we continued to iterate and polish. It’s when adding in the details and business requirements that things tend to break down.
The interface had to afford for a huge number of variables, without breaking the process, and without making it too much of a ‘wizard’. This was something that the old green screen did well, except it required a lot of training and prexisting knowledge.
We wanted to simplfy, start from the most simple case, and let the system grow in complexity as required by the specific order, without the operator having to know about all the nuances beforehand.
The three main columns that we had proposed are now fully integrated into the screen, optimized for cross funcionality and efficient operation.
The mouse is no longer the main tool to interact, as everything can be navigated with shortcuts.
There's contextually aware autofill that simplifies the selection of more obscure parameters.
The final iteration follows intuitive web patterns, with the main account selector on the top, the status bar on the side, and stacked order tickets in the center.
The tickets themselves are responsive, designed for progressive disclosure based on our improved and re-architected business logic.
As the user completes the fields on the ticket, the system expands or contracts the form based on contextual requirements.
No more mental checklists or second guessing.
A Design Systems Contract
Upon delivery of the mutual funds order entry flow, we were hired for more work to redesign all the other asset classes within the same system.
This became a valuable, multi year, design systems project for the agency.
But that's a different story...
Hundreds of millions of dollars are now being traded on a daily basis through these high performing interfaces.
- Don't be afraid of changing the process
- Think big, start small
- Cooperation over hierarchy
- Include everyone (even client) in daily's
- Dave George, PM (Remote)
- Paul Zumbrink, UX Lead
- Vanessa Fahy, Visual Design
- Alice Grantham, Design Research
To comply with my non-disclosure agreement, I have omitted and obfuscated confidential information in this case study.
All information in this case study is my own and does not necessarily reflect the views of Idean or Morgan Stanley.
© 2020 Paul Zumbrink. All rights reserved