FIN∫EGRAL
Use cases Verify Understand Protect Govern Get access
Understand — step 03 of the pipeline

Every field can tell you where it came from.

Twenty-six institutions in nine countries describe the same salary payment twenty-six different ways. We resolve that into one canonical record — and keep the receipts, so you can click any figure and see which sources agreed, when each was read, and exactly how it was derived.

Inspect a field When sources disagree
{{ p.name }} {{ p.body }}

Provenance per field

Click any figure. See exactly how we know it.

One applicant, three institutions, two countries. This is the canonical record your API returns — and every field in it carries its own sources, timestamps and derivation.

ent_9K4M2Q · A. Okonkwo
Canonical record {{ fieldCount }} fields
Provenance {{ activeName }} {{ activeKey }} {{ activeBadge }}
{{ activeValue }}

{{ activeExplain }}

Sources consulted
{{ s.name }} {{ s.raw }} {{ s.state }} {{ s.readAt }}
How it was derived {{ activeDerivation }}
{{ m.label }} {{ m.value }}

Normalisation

The same salary, described four ways.

This is the actual shape of the problem: no shared reference, no consistent casing, no agreed date format, no currency minor-unit convention. The canonical column is what your code sees.

Institution What the bank returns Canonical Rule applied
{{ n.bank }} {{ n.country }} {{ n.raw }} {{ n.canonical }} {{ n.rule }}
The raw payload is kept alongside the canonical value, not discarded. If our mapping is ever wrong, you can see what the institution actually said and we can correct it without re-reading the account.

Conflict policy

When two sources disagree, we do not split the difference.

Averaging a conflict produces a number that no institution would confirm. Five rules decide instead, in this order.

{{ r.no }} {{ r.name }} {{ r.body }}

The pipeline

Understand is where evidence becomes an answer.

{{ s.no }} {{ s.name }} {{ s.body }}

One schema. Every market. Receipts included.

Provenance is on by default — there is no premium tier for knowing where your data came from.