Shipped
Led the design and API build for the Financial Goals feature at Assiniboine Credit Union, replacing free-text review notes with structured goals and review history so any advisor can pick up where the last one left off.
- Product
- Advisor platform at Assiniboine Credit Union
- Skills
- Backend API Development, System Design, Stakeholder Management
- My role
- Software Developer
- Timeline
- May – Aug 2026
Technologies Used
Tech Stack
In brief
The short version
- Problem: Advisors tracked members' financial goals in free-text notes, often just "review done", and leadership couldn't see how many reviews were happening.
- What I did: Designed the feature and built its .NET API: structured goals plus review history. I recommended extending an existing service rather than adding a new one, for a team of five already running close to 30.
- Result: In the first month, 948 goals were created by 66 of 320 advisors, covering 560 members.
Problem
Notes were doing a job they weren't built for
A financial review is a structured conversation between an advisor and a member. Together they pick one to three key financial goals the member commits to acting on, and the advisor follows up on each one afterward. There was no dedicated web tool to capture those goals, track their details, or manage the follow-up. Advisors used notes.
Notes are searchable, but advisors write them for all kinds of reasons, and they're often thin: sometimes just "review done". That tells the next advisor nothing when they're sitting in front of the member.
What a bad day looked like: scrolling through every note to find the last financial review (if one even exists), then starting the conversation from scratch because there's no context on the member's goals or plan from last time.
The executive advice team and the service delivery team raised the problem. They wanted two things: a more detailed record of what was discussed in a review, and insight into how many advisors were completing financial reviews with members.
Objective
Give any advisor a member's goals and last review in one place, and give leadership a way to see reviews happening.
Who it's for
320 financial advisors, plus the executive and service delivery teams who wanted visibility.
My role
Owning the feature end to end
My manager raised this alongside other work. I took it because it sits squarely on why I build software: tools that help advisors give members confidence and clarity in their financial plans, with business value for leadership on top.
I was assigned the whole feature and asked to propose the solution. I wrote the Azure DevOps feature and user stories, then built the .NET REST API service while another software developer built the UI screens. I wrote unit tests for the API's service logic and validation, and releases ran through Azure DevOps pipelines that block deployment when tests fail. Our Product Owner approved every piece of UI wording, a senior software developer signed off on each design decision I made, and I demoed incrementally to advisors and stakeholders.
Who was involved
A product manager, my manager, the senior software developer who signed off on design, a second software developer on the UI, a Product Owner, and QA.
The decision
Extend an existing service, or build a new one?
The situation. At sprint planning, some teammates recommended a new service for this feature, which is how the team had built every feature so far.
My pushback. Five of us were already maintaining close to 30 services, and the burden was showing. I raised the nanoservices anti-pattern and wrote a decision document with two options and a recommendation. A financial goal alone wasn't big enough to justify its own service, and it's easier to start well-scoped and split later than to justify a new one up front (Sam Newman's premature-decomposition warning in Building Microservices).
My recommendation. Add the goals to the existing net worth statement service, and rename it the Financial Planning Service (planned, not done yet), since both belong to the same business domain. It also gave us deployment, and a review starts from the member's net worth statement.
The risk I named. This breaks a pattern the team had agreed on, so I asked the team to decide together. The document also covered the other outcome: if the team kept the pattern, we'd open a story on how service-per-entity scales. The team agreed.
Why not a note box. A tick-box note on the services tab would have cost almost nothing, but advisors could still write "review done". Free text was the original problem.
The trade-off I accepted
That service was built only to manipulate and persist net worth items. Extending it breaks the team's service-per-entity pattern, and the plan is to rename it so the name fits its wider job.
The cheaper option I passed on
A tick-box note on the services tab would have been almost free, but it wouldn't have made review notes structured.
Diagrams
How the decision and the flow worked
Redrawn as generic diagrams, with nothing from the internal system.
Where should Financial Goals live?
Architecture decision
The three options on the table, with what each cost. I recommended extending the net worth statement service, and the senior software developer signed off on the design.
Financial review lifecycle
MVP flow
What an advisor does from opening a member's profile to completing a review, and what the next advisor sees. The executive view was deferred.
MVP
What shipped
Financial goals
Endpoints to create, read, update, and delete a member's goals. Each goal has a name, follow-up date, status (not started, in progress, complete, or cancelled), and notes. Advisors see them when they open the member's profile.
Financial reviews
An advisor starts a review for a specific member, the review date is recorded, and they can view and edit its goals after saving. Completing a review closes it out with a summary.
A cap of three goals
A review holds one to three goals, and the system prevents adding a fourth. It keeps each review focused on what the member is committing to act on.
Review history
Past reviews are kept, so the next advisor can study what was discussed before the conversation starts.
Access auditing, reused
The app already audits which profiles advisors open, so I built on that instead of adding something new.
Out of scope for v1: reporting on goals and a Power BI dashboard for executives. The Product Owner chose to get the feature into advisors' hands and collect feedback first.
Outcome
First month in production
Delivered in August 2026, unit tested, tested end to end by QA, and with no new service added.
948
Financial goals created by advisors in the first month.
66
Advisors created goals, out of 320 financial advisors, about one in five in the first month.
560
Members with an active goal, each one a conversation the next advisor can pick up.
From a usage query covering the first month after release. In that time 82 goals were completed and 3 were cancelled; the rest were still open or in progress.
Learnings
Learnings
The cheap fix can leave the real problem in place
A note box on the services tab would have cost almost nothing, but advisors could still write anything, including "review done". The problem was unstructured data, so the fix had to be structured data.
Reuse isn't free
Extending the net worth service meant no new service to maintain, but it also meant stretching a service built for one narrow job and breaking a team pattern. Its name now needs to catch up. Reuse moves the cost somewhere else.
A pattern break is a team decision
I recommended an option that broke an agreed pattern, so I wrote down both options, the risk, and a fallback, and asked the team to decide together. They agreed. A deviation nobody understands looks like an accident to the next developer.
Ship to users, then build the reporting
The Product Owner deferred the executive dashboard so advisors could use the feature and give feedback before we built reporting on top of it.
Model how the experts work
Next time I'd run an event storming session with the advisors whose reviews work best, then turn how they run a review into steps the software can guide others through. I used the same technique on Graceland and would bring it in earlier here.