UPP execs describe how co-created digital tools are shaping custodial data while cutting operational friction across the growing plan
The University Pension Plan (UPP) is reinventing the data infrastructure behind its investment operations, replacing the standardized reporting that most Canadian pension plans still rely on, with a platform built around curated, decision-ready datasets.
UPP’s approach to building digital capabilities alongside its service providers, rather than simply procuring them, reflects what Ken Lui, director of investment finance at UPP in Toronto, described as the next stage at the CIBC Mellon tech and innovation conference, hosted at Oasis Immersion in Montreal on Wednesday.
According to Lui, UPP’s formation in 2021 gave the plan a rare opening to design its data infrastructure without inheriting legacy reporting systems. The standard model of pulling canned reports no longer fits what his team needs, and the priority is control over how information enters the plan’s own environment, particularly as the plan serves over 46,000 members across six universities and 21 affiliated organizations.
Data that drives portfolio decisions
“That means we can build our vision of the robust modern asset owner reporting platform,” he said. “We want to be intentional about how we bring data into our platform. We're looking for ways to bring in data that’s easier, faster and more reliable. That data is important to us because we make actual investment decisions based on those data points, from FX hedging to portfolio rebalancing. It’s less about making reporting convenient, and more about making the datasets more effective for us,” said Lui.
That distinction runs through UPP’s broader technology push. Rather than accepting the reports a custodian generates on a fixed schedule, Lui’s team worked with CIBC Mellon and BNY to shape a data delivery model that fits UPP’s internal workflows. The older method of running multiple reports and manually merging files is what the Plan wanted to eliminate.
What emerged is a pilot platform, called Project Fuel internally, that organizes information into curated data models grouped by business context rather than by report type. Lui said the ability to merge datasets on a single platform and define the shape of his own queries was the core gain.
“The pilot phase was really incredible as well because we had instantaneous feedback and it allowed me to make sure that the tool hasn't deviated from what we intended it to be in the first place,” Lui said. “CIBC Mellon holds a tremendous amount of data for us so we really want to be able to bring that data into UPP, into our environment, to make meaningful connections and decisions. I think that's what we achieved here.”
The most immediate payoff, Lui said, is time. Because of the platform, it frees UPP’s staff from assembling reports, which means they can focus on analyzing the data itself, which he expects will make their day-to-day work more substantive.
Beyond that, he pointed to Application Programming Interface (API) connectivity as the next priority – pulling curated custodial datasets into UPP’s own systems and layering internal data on top to build richer analytics and support sharper investment decisions.
Operational efficiency from a different angle
Heather St. James, senior director of cash operations at UPP in Toronto, echoed the focus on operational efficiency but from a different angle in a separate panel. She said much of UPP’s growth depends on bringing new universities into the plan, a process that runs across multiple teams and external partners with tightly sequenced dependencies.
Moreover, her asset transfer team alone needs to coordinate a web of activities between UPP, CIBC Mellon and each incoming employer, where one delayed task can hold up the next. While she acknowledged the existing process works, as UPP and CIBC Mellon have onboarded several employers over the last five years, the manual coordination through emails, meetings and spreadsheet project plans is labour-intensive.
“We’re not looking to fix something that’s broken. We’re looking for ways to make an established process more efficient and scalable,” she said, noting that led UPP to explore the Appian platform with CIBC Mellon, which allows the team to build tailored onboarding plans for each university while keeping the underlying framework consistent.
“The reality is no two university onboardings are going to be exactly alike,” she said. “We also use a timeline tool to create different onboarding scenarios with proposed target dates. This allows us to look ahead to understand how activities and dependencies align and potentially identify scheduling challenges before onboarding even begins.”
The people doing the work should build the solution
What set this project apart, St. James said, was UPP’s direct involvement in building the solution rather than simply submitting requirements and waiting for delivery, adding that having the people who perform the work shape the tool matters because they understand where friction tends to show.
“It's important that the people who actually perform the task work on shaping the solution. We understand the nuances, the process, where challenges tend to arise, and what would make a more meaningful difference in how we work together,” said St. James. “While technology is an important enabler, it's really the collaboration that makes the difference.”


