Use case
Utility bills and statements mapped to address schema
Name, address, issuer and date extracted from any bill — the four fields referencing checks actually need.
The problem
Proof of address is the slowest step in onboarding: every issuer formats bills differently, and photos are rarely clean.
- 01Issuer variety. Utilities, banks, councils and telecoms each lay out bills their own way — templates never cover them all.
- 02Photo quality. Folded paper, shadows and cropped edges break rigid parsers and bounce applicants back.
- 03Four-field hunt. Reviewers only need name, address, issuer and date — but they dig through whole pages to find them.
- 04Stale documents. A bill from fourteen months ago passes a tired reviewer. Recency checks need the issue date, parsed.
The solution
Fieldry pulls the four PoA fields from any bill or statement, dated and evidenced.
- 01Issuer-agnostic. No per-utility templates. Any bill layout maps to the same four-field schema.
- 02Date extracted. Issue date parsed and validated, so recency rules run on data instead of eyeballing.
- 03Mismatch surfacing. Bill name or address disagreeing with the ID surfaces side by side with evidence.
- 04Photo-tolerant. Skew and glare handled by OCR; low-confidence fields flagged rather than guessed.
Extraction checks
Search the PoA checks.
See what each check returns.
Every check Fieldry runs on a proof-of-address document.
Issuer-agnostic
No per-utility templates. Any bill layout maps to the same four-field schema.
Date extracted
Issue date parsed and validated, so recency rules run on data instead of eyeballing.
Mismatch surfacing
Bill name or address disagreeing with the ID surfaces side by side with evidence.
Command preview
Referencing-ready values
Exactly the fields referencing checks need, each cited to page and bbox.
Anatomy of an address check
An applicant uploads a crumpled electricity bill photo. A reviewer needs four facts from it — and proof they are current.
Fieldry extracts name, address, issuer and issue date, checks recency and format, and compares against the ID on file.
- InputBill photo, any issuer
- OutputFour cited fields + recency verdict
- ReviewOnly mismatches need eyes
From bill photo to verified address
Reviewer confirms or rejects from evidence — address record closed.
AnalystAddress format validated; name and address compared with the ID record.
FieldryFour fields extracted with confidence; issue date parsed for recency.
FieldryBill photo arrives — folded, shadowed, issuer-specific layout.
Document
4 milestones, newest first.
Key benefits
Address verification without the ping-pong.
Any issuer
Utility, bank, council or telecom — one schema covers every bill format.
Recency enforced
Stale bills flagged by rule, not by reviewer memory.
Mismatch-first review
Clean matches pass untouched; analysts see only disagreements.
File-grade proof
Address records link to bill page and bbox for later audits.
Verify an address from any bill
Run a real bill photo on free credits and check the four fields against the source.
Get startedRelated products
Related use cases
Related industries
Utility bills, bank statements, council tax letters and tenancy agreements. The schema needs name, address, issuer and date — whatever carries all four passes.
The issue date is parsed and compared against your window (e.g. 3 months). Out-of-window bills are flagged automatically.
Skew, glare and folds are handled by OCR; fields below your confidence threshold route to review with the source snippet attached.
Your policy decides — Fieldry surfaces bill-vs-ID disagreements with both evidences; auto-reject, manual review or step-up verification is your call.
Parse your first document today
One API for extraction, evidence and automation.