TD Bank · Visionary of the Year 2025 · Finalist · Sole Inventor
A patent submission that turns personal financial management into a living system. Adjust a savings goal and the budget rebalances, the timeline shifts, and recommendations appear, all in real time, all on one screen. Invented, designed, and prototyped by one person.
Drag any slider, including the date you want. Watch the projection, the timeline, and the budget respond the way the invention does, and let it fix the plan for you when you drift.
A parent sits at the kitchen table after bedtime, moving numbers around to make next month work. Soccer fees, a surprise repair, a summer trip they hope they can afford. Every change raises the same question: if I do this, what happens to that?
Today's tools cannot answer it. Budgets and savings goals live on separate screens, connected in reality but disconnected in the interface. TD's own concept research found people misreading static dashboards, unsure what a monthly total meant or whether a red number was a charge or a balance. The math is left to the customer, and that is exactly where clarity breaks down.
The insight behind this invention was a single question: why don't we show people how their savings affect their budget, and how their budget affects their savings?
Change a savings goal and the budget recalculates. Change a budget category and the goal's end date moves. Every dependent element updates live, in the same view, while the user is still deciding. No save button, no second screen, no mental math.
The system knows when a goal has become unreachable under the current budget. Instead of letting the user fail quietly, it interrupts with specifics: what changed, what it costs in time, and which categories have room to give. Recommendations are generated from the user's own spending patterns.
Scaling, color, and contextual highlights direct attention to what a change actually affects. A timeline that visibly stretches communicates a delayed goal faster than any table. The animation is not decoration; it is the explanation.
The same adaptive logic that rebalances a family's vacation fund can rebalance a small business's cash flow or a wealth client's retirement plan. It is designed as an overlay service on core banking systems, deployable across segments without replacing them.
These are the actual screens submitted with the patent filing, designed at production fidelity in TD's design language. Each flow shows the same engine reacting to a different human moment. Select a step and the phone follows.
Screens tell you what the system decides. Motion shows you how it feels to be answered in real time. This recording walks the two flows above where the engine does its most visible work: the positive path, where trimming a goal earns a green early-finish confirmation, and the date-first path, where picking a new month recalculates the contribution and frees up cash on the spot.
Captured directly from the interactive prototype, at real interaction speed, with no cuts.
The pitch to the committee was not "nicer charts." It was that clarity is a growth engine: customers who can see consequences make more decisions, keep more products, and call support less. The projections below were modeled for the review using industry benchmarks, JD Power digital-banking studies, and TD pilot data.
Personalized savings journeys encourage consistent deposits and reinvestment, modeled from comparable digital savings launches at North American banks.
Predictive insights surface lending, wealth, and protection offers at the moment they are relevant, not in a generic campaign.
Combined value across deposit growth, conversion, retention, and reduced cost-to-serve, against an estimated $10–20M build.
Figures are projections prepared for the TD Visionary of the Year review, not audited results. Methodology combined peer-institution benchmarks, Layer6 AI simulations, and conversion ratios from comparable banking transformations.
This did not come from a roadmap. It came from watching people struggle and asking a question the product had never asked. Then it was written up rigorously enough to enter a patent pipeline and out-place entire teams.
Production-fidelity prototypes in the bank's own design language, three complete flows, every state accounted for: warnings, recoveries, positive paths, and edge cases like unreachable dates.
Feasibility detection, behavioral recommendations, and an agentic assistant concept all treat AI as a design material: something that anticipates, explains, and acts, while the human stays in control.
The submission modeled deposits, conversion, retention, and cost-to-serve against build cost, because an idea only ships when the people funding it can see its consequences too.