01Core step
Drop in one recent PDF
Upload a bank, brokerage, or retirement statement. WealthLens stores the file only long enough to extract text and generate the review.
PDF upload + AI analysis
WealthLens MVP
Upload one statement and get a fast read on fees, allocation, and risk.
This build is intentionally narrow: capture an email, accept one PDF, extract the text, run a structured analysis, and return a result page that feels immediately valuable.
Allocation sleeves
4
Cash, equity, bond, alternative
Fee estimate
0.35%-0.95%
MVP directional range
Result speed
< 1 min
One upload to one saved result page
PDF upload only. No account linking, no aggregation layer, no OAuth wall.
Lead capture is built into the form, then the analysis is stored against that email.
The backend extracts statement text, runs an AI analysis, and persists the result in Postgres.
Results focus on allocation, estimated fees, risk posture, and 2-3 immediate optimization moves.
How it works
One upload path, one analysis pipeline, one saved result page.
The MVP avoids edge-case sprawl and focuses on the core action users care about: drop in a statement and get a credible set of insights back.
02Core step
Let the backend read the statement
The parser pulls readable text from the PDF, then the analysis engine estimates allocation mix, fee drag, and overall risk posture.
03Core step
Land on a visual result page
Each upload gets a reference-backed results page with the saved email, the generated analysis, and a concrete set of next actions.
Result page
A compact visual story instead of a generic text dump.
The analysis page is designed to answer four practical questions immediately: what is the portfolio shape, how expensive is it, how much risk is implied, and what should change first.
Allocation readout
A quick estimate of how much sits in cash, equities, fixed income, and alternatives.
Fee pressure
An estimated annual fee drag with the clearest reasons the statement looks expensive.
Risk score
A simple 1-100 posture score that helps frame concentration, stability, and downside sensitivity.
Optimization moves
Two or three concise actions designed to feel specific enough to be immediately useful.
Stored output
Backed by Postgres, keyed by reference.
Public upload reference for each result page
Captured email stored with the generated analysis
Structured JSON persisted for later reuse
PDF text extraction happens server-side before analysis