UBS Pipeline Database Project
As an Associate at UBS, I noticed a lot of frustration across the Public Finance Investment Banking department with all of our manual processes. I saw an opportunity for us to introduce automation and improve data management, and decided to pitch a Strategic Design role to the head of our department. I became responsible for building new systems and processes to improve efficiency. I conducted all strategy, research, design, & coding myself.
Finding the real solution, not just fixing a broken document
The first problem I sought to address was our disorganized and inconsistent Weekly Pipeline report. This single PDF guided our Weekly Pipeline Call, which was meant to give 90 bankers across the country shared visibility into everything that was in the pipeline across the department, from client RFPs to client meetings & pitches. It failed miserably.
The document itself wasn’t the biggest problem - the data beneath it was. It was manually compiled in Excel by a new junior banker each week who individually emailed every (busy!) team for updates on Monday mornings. As a result, the data was rarely current, full of duplicate and incorrect items, and the PDF file had to be manually reformatted every single week. The entire process to create this document that wasn't even legible in its final form took 3 hours of time that could've been better spent on client-facing work.
Why it mattered:
Unreliable pipeline data and reports meant leaders were steering a 90-person department on information nobody trusted. Not to mention that because the department spanned every region and sector, reusable client pitches and ideas went ignored and unknown to the teams who could have leveraged them. Reformatting the PDF file would have treated the symptom. The real solution was building a single reliable source of truth for client touchpoint data, with the manual portions of the work automated away.
Approaching the build thoughtfully
I interviewed across every level - junior and senior bankers, regional heads, compliance, and the department head. Each stakeholder group used the pipeline differently, and a tool that ignored any one of them would be abandoned. I turned those conversations into process diagrams and a requirements list to make sure I didn't leave out anything that could be addressed through the build process.
With effectively no budget, I audited what we already had licensed and chose Microsoft Access. It was certainly not the most high-tech option, but the right one on cost, customization, deployability, and ease of maintenance. I scoped it into a four-phase roadmap to de-risk the build and show value early, then presented the plan to the department head for buy-in and approval.
I designed and coded both the front and back ends of the database system, teaching myself SQL and VBA through forums as I went. The system was tastefully designed with automated email generation, one-click report generation, full search capabilities, and automated compliance checks. I tested with a small group of bankers before the department-wide launch, then built training materials, ran live demos, and stayed on as steward of data integrity.
*Please note, the Pipeline Database and associated work is property of UBS. I have recreated aspects from memory for illustration purposes.
Leveraged Microsoft Access instead of coding from scratch
In a perfect world, we would have been able to hire a developer to help us build a fully custom enterprise software. The reality was that not only did we not have the budget, but our organization was very strict on security. Getting the approvals would have taken months that we just didn't have. In leveraging the custom capabilities within Microsoft Access, I knew it wouldn’t necessarily scale forever, but it was free, ready to use right away, and something the team could keep running without an engineer until we got the budget and resources to build something more substantial.
Rolled it out in phases instead of all at once
My colleagues waited longer for the full tool, but the phased approach help me build trust early and let me fix and iterate features based on how people actually used it.
Realizing better outcomes
I set out to automate key repetitive tasks, reduce time spent manually formatting reports, increase knowledge share throughout the department, and create a single system that allowed us to manage our client interactions more effectively. These objectives led to real outcomes, which included:
-
Cut manual work from hours to seconds. The weekly report and required compliance approvals/emails were automated end to end. These processes that would have taken 1-3 hours were now reduced to 1-2 seconds.
-
Made knowledge actually findable. I designed our reports to include keywords and clickable links to files on the shared drive. Bankers loved this feature because they no longer had to hunt down decks or proposals discussed on the Pipeline Call that they found interesting. I received a lot of positive feedback of how helpful this feature was.
-
Improved strategic planning. With reliable, current data, group heads could monitor client coverage and reallocate resources deliberately.
-
Earned a promotion to Director. The impact led to my promotion to Director and a new mandate to build an automated revenue-tracking system for the firm’s municipal deal activity. I started the revenue project, but didn't get to finish it before the firm acquired Credit Suisse and subsequently laid off the entire Public Finance Investment Banking Department.
Reduced
Task Time
Tasks that took 1-3 hours now took 1-2 seconds
Added Search Capabilities
Keywords and clickable file links made content findable
Improved Strategic Planning
Increased visibility helped monitor client coverage
Promoted to
Director
Earned a promotion thanks to the work & outcomes realized