Stacking is concurrent borrowing that no single lender sees in full
Loan stacking is one borrower taking several loans from different lenders in a short window, before the credit bureau shows the earlier ones. The US Consumer Financial Protection Bureau (CFPB), writing about buy now, pay later (BNPL) in 2022, described it as concurrent loans at different lenders that the borrower may be unable to repay. It added that BNPL lenders' guardrail of low starting limits for first-time borrowers can erode when an applicant can borrow from several lenders at once.
Concurrent borrowing often has ordinary reasons. The Digital Credit Market Inquiry into Kenya, run by the Competition Authority of Kenya and Innovations for Poverty Action (IPA) and summarised by IPA in 2021, surveyed 793 digital financial service users. It found that 33% of digital credit users had more than one active digital loan before March 2020, from one provider or several. The main reasons given were emergencies (52%) and low loan limits (35%). Among users with loans from several providers, 96% usually took them within 30 days of each other, and these users faced a higher default risk than users with several loans from one provider. Kenya is not one of our markets, and the gap there was wider than a reporting lag: non-bank digital lenders were not allowed to report to credit reference bureaus at the time of the 2021 summary.
For a lender, the cause matters less than the effect. A stacked applicant looks like any other at the moment of decision, and the loan is approved on an incomplete picture of existing debt. Whatever the borrower's reason, the application alone does not tell you which loans are already open.
Fast digital decisions leave bureau data little time to catch up
When the Reserve Bank of India (RBI) shortened bureau reporting intervals in 2024, it cited the faster turnaround in credit underwriting through digital processes and the need for credit reports to reflect more current information. The CFPB said the low financial and operational burden of BNPL, and its availability from several lenders at virtually every digital retailer, may contribute to stacking.
The reporting lag is set by rules and calendars
A bureau holds what lenders have reported, on a schedule. In India, an RBI instruction of January 2015 asked for monthly updates. An August 2024 circular moved credit institutions, NBFCs included, to a fortnightly basis, as on the 15th and the last day of the month, with submission within seven calendar days. It took effect on 1 January 2025 and gave bureaus five calendar days from receipt to ingest the data.
From 1 July 2026, December 2025 amendments replace that with four reference dates a month: the 9th, the 16th, the 23rd and the last day. For NBFCs the schedule is in paragraph 10(2) of the RBI's consolidated credit information reporting directions. Lenders submit incremental accounts, which include newly opened ones, within four calendar days of the first three dates, and a full file for the month end by the 5th of the next month. A companion amendment gives credit information companies three calendar days from receipt to ingest the files and send back any rejection reports, and points to an illustration of the timetable in its Annex-IX. The figure shows a fictional identity applying to five lenders within about 48 hours against that lag.
- Lender AApproved 10th at 09:30
- Lender BApproved 10th at 17:55
- Lender CApproved 11th at 08:40
- Lender DApproved 11th at 19:30
- Lender EApproved 12th at 09:20
- 9th
- Reference date. The first loan opens the next day and misses this file.
- 16th
- Next reference date, the first file that can include the loans.
- 20th
- Latest lender submission, four calendar days after the 16th.
- 23rd
- Latest bureau ingestion, three calendar days after receipt.
Apply those deadlines to a loan opened the day after a reference date. It misses that date's file, waits for the next one, then needs the submission window and the bureau's ingestion window before it can appear in a report. On the stated deadlines that is roughly 13 to 15 days until the bureau has ingested it (our arithmetic from the dates, not a figure the RBI publishes). The cited texts do not say when ingested data appears in a report. Lenders and bureaus may agree shorter intervals, and a lender that reports late adds to the delay, so read the range as the worst case for a compliant lender, not a typical delay. In the figure, all five decisions are made before the first loan reaches the bureau.
Digital lending is covered. The RBI's 2025 digital lending directions require regulated entities to report lending done through their own digital lending apps, and those of their lending service providers, to credit information companies irrespective of nature or tenor, so short-tenor products are reportable too, under the schedule that applies to the lender. This section uses India because the RBI publishes fixed deadlines; for Kazakhstan, Nigeria and Peru, check the schedule that applies to your own bureau and lender type. In 2021 the Kenyan inquiry's authors judged monthly bank reporting too slow to detect multiple borrowing and said a daily reporting credit information system would help.
Stacking shows up in enquiries, shared devices and linked identities
No single signal proves stacking. This list is practitioner judgement, not a regulatory list; the linked guides give the sources for the device, identity and velocity signals. Read them together.
- Several bureau enquiries from different lenders in a short window, with no matching new accounts on the report yet.
- One device, or a cluster of devices, starting applications for different products or different lenders' flows. Emulators and device farms are the extreme case.
- One phone number, national ID, bank account or email address appearing in several applications with small changes to the name or date of birth. See recycled and synthetic identities.
- Several applications from one identity or device inside the same hour or day, which is what a velocity check counts. See velocity and behaviour signals.
Visibility is the constraint. A lender sees its own applications and whatever the bureau holds. A partner that screens traffic before delivery sees only the applications that pass through it, and not what happens at other lenders. Cross-lender stacking is visible only where data is shared.
No single control closes the stacking blind spot
| Control | What it catches | Where it falls short |
|---|---|---|
| Bureau enquiries | Applications at other lenders that are not yet loans on the report, where the bureau records enquiries. | Depends on how enquiries are recorded. The CFPB noted that soft enquiries left BNPL lenders with no view of borrowing elsewhere. |
| Faster reporting | The window itself: a loan reported sooner is visible sooner. | Set by regulation and bureau agreements, not by one lender. The RBI text allows shorter intervals by mutual agreement between lender and bureau. |
| Own-data lookback | The same phone number, ID, device or bank account across all your earlier applicants, declined ones included. | Sees only your own applications, not other lenders'. |
| Velocity on identity | Repeated use of one phone number, ID or device within a short period. | Counts only what the lender or partner sees, and needs thresholds that real borrowers rarely cross. |
| First-loan limits | Caps the exposure from any one lender when limits start low and rise only after on-time payments. | The CFPB noted that these guardrails can erode when borrowers can take concurrent BNPL loans from several lenders. |
| Partner-side checks | Repeat applications from one device or identity, counted by the velocity check before the applicant reaches the lender. | Cannot see loans at other lenders, which stay with your own underwriting and bureau checks. Rules are agreed per market and product, and the lender keeps the credit decision. |
The CFPB concluded, for BNPL, that addressing stacking needs near-real-time data on an applicant's BNPL usage with competitors, and a coordinated effort across lenders to provide it. Where such sharing exists, use it. Where it does not, the gains come from reading the bureau at decision time, reporting your own loans on schedule or faster, and counting repeats in your own traffic.
Partner-side screening adds a check before the lender decides
How we screen covers the checks CYBERTECH reads together before an applicant reaches the lender, who keeps the final credit decision.
Outcomes come back by postback where the integration allows it, which helps tune the checks over time. First-payment default by source is worth reading for sources that bring repeat borrowers, though it cannot tell stacking from other causes; see lead quality by source. Agree the billable event and any validation window with the bureau lag in mind; pricing shows where each falls.
General information about regulation, not legal advice. Reporting deadlines change, so check the current RBI text and your bureau agreement before you design around them.
Related guides
- Duplicate, recycled and synthetic identities hide between applicationsHow duplicate applications, recycled identities and synthetic identities appear in lead flows, and why de-duplication should run before billing.
- Velocity and behavioural signals expose scripted loan applicationsHow velocity windows and form-fill behaviour expose bots and rings, where they misfire on shared devices and networks, and how to tune them.
- Measure loan lead quality by source, from approval to delinquencyHow a lender measures lead quality by source: approval, disbursal, first-payment default, 30+ DPD, fraud rate and cost per disbursed loan.
- Screen loan leads for fraud and fit before they reach youFive checks, including fit to your own criteria, screen applicants before handoff. No single signal decides, and a stopped applicant is not delivered to you.
Sources
- Reserve Bank of India: Frequency of reporting of credit information by credit institutions to credit information companies, circular RBI/2024-25/60 (2024)
- Reserve Bank of India: Non-Banking Financial Companies - Credit Information Reporting Amendment Directions, 2025, RBI/DOR/2025-26/117 (2025)
- Reserve Bank of India: Credit Information Companies Amendment Directions, 2025, RBI/DOR/2025-26/119 (2025)
- Reserve Bank of India: Non-Banking Financial Companies - Credit Information Reporting Directions, 2025, RBI/DOR/2025-26/349 (2025, updated 2026)
- Reserve Bank of India: Digital Lending Directions, RBI/2025-26/36 (2025)
- Consumer Financial Protection Bureau: Buy Now, Pay Later, market trends and consumer impacts (2022)
- Innovations for Poverty Action: New consumer protection and competition policies are needed for Kenya's digital credit market (2021)