Account Aggregator Statement Pipeline for Indian Lenders
Under the RBI Account Aggregator framework, the lender Financial Information User holds the consent and the regulatory obligation. TransactIQ supports that role — payload normalisation across OneMoney, Setu, Finvu, and CAMS-AA; consent-boundary enforcement including revocation handling; AA-vs-PDF accuracy delta in the audit record; and the same 40+ engineered signals across both AA and PDF inputs.
What TransactIQ does for an AA-enabled lender pipeline
Nine capabilities that map to the operational shape of a working AA pipeline at an Indian lender FIU.
Multi-AA payload normalisation
Production AA pipelines rarely sit on a single AA. TransactIQ normalises payloads across OneMoney, Setu, Finvu, and CAMS-AA into a single internal model — masking the format variations so the downstream credit policy does not depend on which AA the consent flowed through.
AA + PDF mix as the realistic shape
Most lender pipelines are AA-where-available, PDF-where-not. Some applicant accounts are not yet AA-enabled, some consents fail, some are revoked mid-flow. TransactIQ handles both inputs through the same signal model and surfaces an AA-vs-PDF accuracy delta in the audit trail so credit teams can see where each input contributed.
FIP/FIU role separation
The AA framework draws a clear architectural line between Financial Information Provider (the bank), Account Aggregator (the consent broker), and Financial Information User (the lender). TransactIQ operates squarely in the FIU role — consuming the AA-delivered payload under the consent the FIU obtained. No FIP impersonation, no consent boundary blurring.
Consent revocation handling
Under the AA framework, the borrower can revoke consent at any point. TransactIQ supports the operational pattern: existing decisioned outputs are retained per the consent-at-time, in-flight pulls cancel on revocation, downstream re-pull attempts are blocked, and the revocation event is recorded in the per-statement audit record.
AA data-quality remediation
AA payloads have known quality issues: missing narration tokens, mis-classified transaction categories, period-boundary gaps, and FIP-side parser bugs. TransactIQ surfaces these explicitly rather than silently absorbing them — quality flags per statement and per signal, so the credit policy can apply appropriate caution.
AA-vs-PDF accuracy delta
For the same applicant account, the AA payload and the PDF statement do not always agree — period coverage may differ, transaction-category fields may be sparse on the AA side, and counterparty narration may be richer on PDF. TransactIQ surfaces the delta as part of the audit record so credit policy can choose its evidence source.
NBFC-AA list awareness
The current operational AA list is small (OneMoney, Setu, Finvu, CAMS-AA, and a handful of others); the lender FIU integration set typically follows the same shape. TransactIQ supports the production list and tracks new entrants — adding payload-normalisation support as new AAs reach scale.
40+ engineered signals on AA data
Salary regularity, EMI obligation density, bounce predictor, recurring vendor outflows, GST-payment outflows, top counterparties, average daily balance bands — the same signal set produced from PDF runs natively on AA payloads, with the AA-vs-PDF delta surfaced where it matters.
AA consent artefact in audit trail
Per-statement audit record includes the AA consent handle, FIP source, consent period, and revocation status. The lender FIU can produce the full consent-and-pull lineage for any borrower or regulator review without reconstructing it from separate systems.
Why TransactIQ for Account Aggregator
Four dimensions where a lender FIU technology team typically compares vendors on AA.
| Dimension | Typical incumbent posture | TransactIQ |
|---|---|---|
| Single-AA vs multi-AA support | Many BSA vendors integrate one AA deeply and treat others as secondary. | Normalisation across OneMoney, Setu, Finvu, and CAMS-AA from day one. New AAs added as they reach production scale on the FIU integration set. |
| AA-vs-PDF parity | AA and PDF often run through separate pipelines, with different signal coverage and different output shapes. | AA and PDF feed the same signal model; AA-vs-PDF delta is surfaced as part of the audit record. Credit policy chooses its evidence source explicitly. |
| Consent boundary enforcement | Consent state is typically tracked in the lender FIU app layer; downstream BSA usage may not respect it. | Consent revocation cancels in-flight pulls, blocks re-pull attempts, and is recorded in the per-statement audit record. Decisioned outputs retain the consent-at-time stamp. |
| AA data-quality remediation | Quality issues in AA payloads (missing narration, mis-classified categories) are often silently absorbed. | Quality flags per statement and per signal surface AA data-quality issues so credit policy can apply caution rather than treat AA as automatically clean. |
See TransactIQ on your AA + PDF mix
Send a representative sample across your AA-and-PDF mix. Receive an AA-vs-PDF accuracy delta benchmark and an FIU integration-fit scoping.
Request AA benchmark