When people say "open banking," they usually mean the UK/European model: regulators forcing banks to expose APIs so licensed third parties can access customer data. It's a reasonable mental model — but it's not how India did it. India's Account Aggregator (AA) framework took a different architectural path, and understanding the differences explains a lot about why financial-data-driven lending is scaling differently in each market.
The Global Landscape in Brief
United Kingdom & Europe: regulator-mandated open banking
The UK's Open Banking initiative (since 2018) and the EU's PSD2 require banks to provide free, standardised APIs for account information and payment initiation to regulated third parties, with customer consent. Strengths: strong consumer protection, standardisation, thriving app ecosystem. Weaknesses: adoption lagged for years, API quality varied, and commercial sustainability remained murky since banks had little incentive to invest.
United States: market-led aggregation
No comprehensive mandate exists. Instead, data flows through screen-scraping and aggregator companies (Plaid being the best known) negotiating bilateral deals with banks. It works at scale but is fragile: no uniform standards, ongoing legal ambiguity about data rights, and reliance on private contracts. Recent regulatory moves (CFPB's Section 1033 rulemaking) are pushing toward mandated open banking, but implementation remains contested.
Australia: Consumer Data Right
Australia legislated a broader "Consumer Data Right" — extending beyond banking to energy and telecom — with accredited data recipients and phased rollout. Conceptually closest to India in ambition, though adoption has been slower than hoped.
India's Account Aggregator: A Different Design
India's AA framework, live since 2021 under RBI's non-banking financial company – Account Aggregator regulations, reimagines the problem. Key components:
The AA as a neutral conduit
An Account Aggregator (an NBFC-AA licensed by RBI) doesn't hold or see your data. It brokers consented transfers between a Financial Information Provider (FIP — your bank) and Financial Information User (FIU — a lender, wealth platform, etc.). Think of it as a consent-switchboard, not a data warehouse.
Consent as a signed artefact
Every data request is a structured, digitally signed consent artefact: what data, for what period, for what purpose, revocable anytime. This is stronger than the checkbox consents of legacy screen-scraping — purpose limitation and revocability are built into the protocol.
Sahamati: the ecosystem glue
Sahamati, a non-profit, acts as the ecosystem alliance driving adoption, interoperability standards, and certification across AAs, FIPs, and FIUs. Much of the framework's momentum — hundreds of banks live as FIPs, thousands of FIUs onboarded — traces to this coordinated governance layer, something no other country's model replicates.
Built for scale, not just apps
Where UK open banking primarily powers consumer apps (budgeting, payments), India's AA was designed with credit as the marquee use case: instant, verified bank statements for loan underwriting, replacing the forged-PDF economy entirely.
Head-to-Head Comparison
| Dimension | UK/EU Open Banking | US Market-Led | India AA |
|---|---|---|---|
| Driver | Regulatory mandate | Private contracts | Regulation + ecosystem |
| Data holder | Bank APIs | Screen-scraping/APIs | Bank-side, via AA |
| Consent | App-level | Varies | Signed, revocable artefact |
| Intermediary sees data? | Third party often stores it | Aggregator stores it | No — AA is blind |
| Primary use case | Payments, PFM | Lending, PFM | Lending, investments |
Why the Indian Design Matters for Credit
For underwriting specifically, the AA model has three decisive advantages:
- Veracity: data comes straight from the bank through a regulated pipe. Document fraud — editing PDF statements — simply stops working.
- Speed and cost: a consent flow takes seconds and pennies, making small-ticket MSME loans economically viable.
- Composability: AA bank data combines naturally with GSTN filings and MCA records, giving lenders a triangulated view. Digital-native lenders — including financing platforms like KredFlow — build instant approval journeys directly on this stack.
Remaining Challenges
No framework is finished. In India: consent-drop-off rates during onboarding, uneven FIP API reliability, limited coverage of some data types (and international accounts entirely), and the need for borrower education. Globally, the lesson from the UK experience — mandates alone don't create ecosystems — is precisely why India invested in Sahamati-style coordination.
The Takeaway
The world is converging on the same principle — customers own their financial data — through three different routes: mandate, market, and India's regulated-consent architecture. For B2B lending, India's version may prove the most consequential, because it was designed from day one around the exact problem credit assessment faces: getting trustworthy, current, granular financial data at near-zero marginal cost.