← Selected work

Personal build · Personal finance product

Budgets were not the problem. Knowing whether we were following one was.

My family had reliable income and could make a monthly budget, but we still found ourselves with a pile of money at the beginning of the month and too little halfway through. I built Balanced to turn bank activity into simple, readable evidence of whether we were actually keeping to the plan.

Role
Creator and product engineer
Context
Household budgeting and cash-flow visibility
Disclosure
Personal project

The problem

The problem

Income was not the problem, and creating a budget was not the problem. The missing piece was an easy way to compare everyday spending with that budget before the month was already lost. Existing spreadsheets could record the plan, but they did not naturally bring in the transactions or make the remaining room in each category obvious to everyone in the household.

What I built

The system, not just the screen.

A transaction pipeline tied to an actual plan

Connected bank accounts through Plaid, queued transaction imports, stored the account and category context, and attached spending to a budget item. Users could add or reassign transactions manually too, so the household plan did not have to wait for a bank feed to be perfect.

A monthly control panel instead of a retrospective

Built monthly budgets with category-level progress states, planned and actual remaining amounts, “safe to spend,” uncategorized-transaction alerts, upcoming bills, and overspend callouts. Repeating bills and prior budgets could roll into a new month rather than starting every plan from scratch.

A small product for more than one household

Added goals, daily recaps, web push notifications, shared household accounts, self-serve plans, and account-management controls. Around 10–15 people used the product, but its most important job was giving our own family the visibility to stay on budget.

System sketch

How the system works.

A simplified view of the system boundary and the work it moves through.

  1. Bank accounts
  2. Plaid connection
  3. Queued transaction import
  4. Categorized spending
  5. Budget progress + safe to spend

Key decisions

Key decisions

  1. Treat transaction import as a useful input, not the source of truth: manual creation, reassignment, and categorization keep the budget usable when a bank connection lags.
  2. Make each budget item legible at a glance with current amount, remaining amount, and a progress state instead of requiring a financial-reporting mindset.
  3. Use scheduled imports, manual refresh, reauthentication prompts, and notifications to reduce the operational burden of a third-party financial-data connection without pretending that dependence disappeared.
  4. Keep the outcome focused on household behavior—can we safely spend this?—rather than creating a more elaborate investment or accounting product.

What it enables

What it enables

Balanced was a godsend while we used it: it gave our family a real-time-enough view of the budget and helped a small group of other users do the same. We eventually moved to banks whose built-in budgeting experience fit our needs better. That transition was not a product failure; it was evidence that the visibility problem Balanced solved had become a standard expectation—while Plaid delays and periodic reconnections remained the tradeoff of building on bank aggregation.

Return to selected work

Similar problem?

Facing a similar kind of product or systems problem?

Tell me where the complexity is piling up. If there’s a strong fit, I’ll reply personally.

A few sentences is plenty.

Your note goes straight to me—no mailing list.