Pinnacle Award Finalist


Treasury and finance practitioners are in control of who will be crowned the Pinnacle Grand Prize Winner. The finalist with the most votes will be honored during AFP 2021.

Data and digital transformation are the story for this year’s Pinnacle finalists. All three treasury departments found ways to integrate data from both internal and external sources and then used AI, RPA and APIs to completely transform how the organization functioned leveraging that data into insights.

Learn More About Each Finalist


Bechtel’s entry focused on using a robotic process automation (RPA) tool to query Human Resource (HR) Management System database daily and compare against Treasury Management System data to identify if signatories had terminated or changed positions. When the solution ran for the first time, it generated 280 tickets that impacted 390 bank accounts where the signer had some type of HR position change or termination. The tool is now used daily to identify position changes in real-time to enable timely update of bank records, comply with long-standing audit recommendation, and prevent fraud/financial loss.

How does this solution build a base for continued growth?
The RPA piece of this process provides a solution to automate repetitive tasks and has been effective in identifying and communicating real time changes in employment status. This builds a base to potentially combine RPA with machine learning. In the current process, the robot’s job is complete once it identifies an HR change and creates a ticket. With the introduction of machine learning, it could be possible for the robot to develop some or all the required documents, resulting in fully automated end-to-end solution.

What was the hardest challenge to overcome in this process?
From the initial design through implementation and support, the learning curve for the ticketing system was the most difficult challenge. The ticketing system is customizable but there were limitations due to rules and securities, reducing the user-friendliness of the tool. Majority of users are not familiar with ticketing systems and require additional assistance. As additional tickets are created and more users operate the system, areas for improvement are identified.


HCSC’s submission centered on treasury becoming more data-driven, which meant upgrading skills, process efficiencies, automation, technology, and digital visualization techniques to better capture treasury data. In response, HCSC’s treasury team underwent a complete digital transformation. New teams were created, including a dedicated treasury systems solutions team, expanded forecast teams, and created enterprise banking service solutions function. Technology platforms were optimized, while data was centrally organized to make better decisions.

How does this solution build a base for continued growth?
Real-time data access and visibility are critical because of the speed of business. If you do not have a technology solution that can consolidate all your cash data and connect to other platforms, you will be pulling data from bank and trading websites, and then putting them into a spreadsheet. With an ever-evolving technology platform and automation mandate, many of the problems associated with manual data collection go away because the technology is always working in the background. It is meaningful to have full functionality and interconnected systems within a flexible technology solution, such as the TEAR data base we created.

What was the hardest challenge to overcome in this process?
One of the biggest challenges was integrating new technology. Real-time feedback from staff resulted in weekly patches being pushed to the platform, allowing treasury to operate more effectively as it underwent integrating with the various technology platforms. As an all-treasury systems project, the ability to access real-time information allowed treasury to be very agile, operating across multiple system spectrums and time zones.

Another challenge was co-operation, as relationship management with key banking and fintech partners proved vital. The planning meetings held with banks and fintech vendors helped all parties stay aligned on every aspect of the project. Given HCSC’s strong relationships with its banking partners, IT and project teams were set up to ensure that all partners worked to achieve a common goal.


Micron’s entry focused on optimizing fund allocation. The treasury and automation teams developed an artificial intelligence tool designed to pull data from treasury systems that were already being aggregated into a single daily file. The first week it was implemented, the tool advised a reallocation of cash investments that increased the portfolio yield by low single digits of basis points. The treasury staff now runs the model daily to search for further opportunities to increase revenues through more optimal allocation.

How does this solution build a base for continued growth?
Since development of the tool in mid-2021 the team has already began looking for opportunities to advance the end product for more efficiency. Some of those potential advancements include automating the model to run with no manual input necessary from the user. Another advancement that has been discussed is to fully implement the model into the teams daily cash positioning process. This would build off of the previous concept where every single morning the model would automatically run and solve for daily cash excess/deficits with no manual entry by the team. Based on that analysis, the model would decide where funds should be invested or where funds should be redeemed and have payment details input into the Treasury Management System for the team to approve.

What was the hardest challenge to overcome in this process?
Since members of the Treasury team were not experts in the data science behind the model and the members of the Advanced Analytics and Automation team were not experts in Treasury concepts, it was crucial for the teams to work closely over the course of the entire project to ensure what was being developed was accurate and that any recommendation in the model output was feasible (i.e. within investment policy constraints). Although it was a challenge to take the time to explain the different concepts across teams, it was a unique opportunity for each to gain the knowledge in an area that was previously less familiar.