An AI-native personal finance platform. Track income and expenses, hold a budget that adapts, and get insight that reads like advice instead of a pie chart.
Budgeting apps categorise. They don't explain.
Every finance app can tell you that you spent $847 on restaurants. None of them tell you that it happened because you moved further from the office, or that at this rate the trip you're saving for slips by two months.
- Charts that describe the past and predict nothing.
- Categories you spend more time correcting than using.
- Budgets that are fixed while your life is not.
- Insight that stops at 'you spent more than last month'.
A model that reasons over your ledger.
Spanzo builds a structured picture of your finances — recurring flows, one-offs, seasonality, drift — and reasons over it. It surfaces causes and consequences, not just totals, and the numbers are computed deterministically before the model ever sees them.
- Transactions categorised on ingest, learning your corrections.
- Budgets that adapt to income that varies month to month.
- Forecasts that say when a goal actually lands.
- Every insight traceable to the transactions behind it.
What Spanzo does.
Expense tracking
Automatic categorisation on ingest. Correct it once and the model remembers — for that merchant and the ones like it.
Income tracking
Built for income that isn't a salary. Detects recurring flows, invoices, and irregular deposits without being told.
Budget management
Envelopes that rebalance against actual income rather than the number you optimistically typed in January.
AI insights
Plain-language explanation of what changed and why, grounded in your data. Every claim links to the transactions behind it.
Financial analytics
Cash-flow projection, burn rate, and runway. The metrics a CFO would use, aimed at one person's balance sheet.
Private by construction
Bank data is encrypted per-user. Nothing you own is used to train a shared model, and you can export or delete all of it.
How it's put together.
Every layer exists because a guarantee demanded it.
Clients
Flutter on iOS and Android, Next.js on the web. One API contract, so a feature ships everywhere at once.
FlutterNext.jsIngest
Bank aggregation normalised into a canonical transaction shape, then deduplicated against what's already stored.
GoKafkaAnalysis
Deterministic aggregation first — totals, trends, and forecasts are computed in code, never guessed by a model.
PythonPostgreSQLReasoning
An LLM narrates the computed numbers and answers questions. It reads results; it never invents them.
LLMRAG
A look inside.
Interface walkthrough coming with the next release.
Priced to be predictable.
Indicative pricing while we finalise plans. Nothing here is a surprise later.
Free
Enough to see where the money actually goes.
- 2 connected accounts
- Expense & income tracking
- Monthly budgets
- 3 months of history
Plus
Most popularThe full picture, with the model on top of it.
- Unlimited accounts
- AI insights & Q&A
- Adaptive budgets
- Cash-flow forecasting
- Unlimited history
Household
Shared finances, separate privacy.
- Everything in Plus
- Up to 5 members
- Shared goals & envelopes
- Per-member privacy controls