ScrutoraCode, cloud & consent
Back to Blog
|6 min readCompliance Guide

DPDP Rules 2025, Rule 6(1)(f): What “Appropriate Provision in the Contract” With a Data Processor Actually Requires

Part 6 of 7. The clause most directly tested by breaches that originate through a third party rather than your own systems.

By Vimalendukumar Dwivedi, Co-founder - Scrutora

Part of a series on DPDPA Section 8(5) and DPDP Rules 2025, Rule 6. Read the pillar post first if you haven’t.

The Clause, Verbatim

“(f) appropriate provision in the contract entered into between such Data Fiduciary and such a Data Processor, wherever applicable, for taking reasonable security safeguards; and”

DPDP Rules 2025, Rule 6(1)(f)

Every other clause in Rule 6(1) is about controls inside your own environment. This one is about controls inside someone else’s, your vendors, your subprocessors, anyone processing personal data on your behalf, governed only by what you managed to put in a contract. It’s also the clause most directly tested by real-world breaches that originate through a third party rather than through the Data Fiduciary’s own systems.

What It Technically Requires

A written data processing agreement (DPA) with every Data Processor handling personal data on your behalf, not a general commercial contract with a data clause bolted on as an afterthought.

Security obligations mirroring Rule 6 itself. A DPA that requires the processor to implement reasonable security safeguards should, in practice, flow the substance of Rule 6(1)(a) through (e) and (g) down to the processor: their own encryption and access-control standards, their own logging and monitoring, their own incident response commitments.

Breach notification obligations. A contractual requirement that the processor notify you promptly if a breach occurs on their side, with a defined timeframe, because your own obligation to assess and potentially report a breach under DPDPA starts running from when you become aware of it, and you can’t become aware of what a processor doesn’t tell you.

Audit and verification rights. The right to audit, request evidence of, or otherwise verify the processor’s actual security posture, not just take their word for the contractual promise. A right to audit that’s never exercised is functionally the same as not having the right.

Subprocessor flow-down. Where the processor itself uses subprocessors, contractual requirements that the same obligations extend down that chain, so the safeguard doesn’t stop at the first link.

Termination and data-return or deletion terms.Clear provisions for what happens to personal data when the relationship ends: secure deletion or return, and confirmation of it, not an assumption that a lapsed contract means the data quietly disappears.

Where This Gets Subjective

Does a boilerplate clause satisfy “appropriate”? A one-paragraph clause stating the processor “shall maintain reasonable security measures” is, technically, a provision in the contract for taking reasonable security safeguards. A detailed DPA with named technical requirements, audit rights, breach SLAs, and subprocessor terms is also a provision in the contract for the same thing. Both satisfy the clause’s bare language. Only one of them gives you any actual leverage or visibility if the processor’s security turns out to be inadequate, and the EY breach earlier this year, where exposure ran through a third-party IT support platform rather than EY’s own systems, is a live example of why that gap matters in practice, not just in theory.

Enforcement versus existence. A contract clause is a piece of paper until it’s tested. An organisation that signs a strong DPA and never exercises its audit rights, never reviews evidence of the processor’s actual controls, and never follows up on the processor’s own compliance, has the clause but not necessarily the safeguard. Whether “appropriate provision in the contract” implies an ongoing verification obligation, or is satisfied purely by the contract’s existence, is a real point of disagreement, and the more defensible reading leans toward the former given the clause sits inside a rule about actually preventing breaches, not about paperwork.

“Wherever applicable” and vendor tiering. Not every vendor touching your infrastructure processes personal data, and not every processor handles equally sensitive data. The clause’s “wherever applicable” qualifier gives room to tier contractual rigor by risk, a payment processor handling financial identifiers reasonably gets a heavier DPA than an email delivery vendor handling only opt-in marketing addresses, but where exactly that tiering line sits is left to the Data Fiduciary’s own judgment, with no floor specified in the clause itself.

What the Breakdown Really Means

Rule 6(1)(f) is asking whether your organisation’s security safeguards genuinely extend through every hand personal data passes through, not just the hands you control directly. A DPA that exists but was never negotiated for substance, never checked against how the processor actually operates, and never revisited after signing, is the contractual equivalent of a checklist item ticked without the underlying control actually being verified.

The clause’s real test isn’t whether the contract has the right words in it. It’s whether, if that processor had a breach tomorrow, the contract would have done anything to prevent it, catch it early, or hold the processor accountable for it.

Vimalendukumar Dwivedi is Co-founder of Scrutora, a Code Compliance Platform.

See where personal data actually flows to third parties

Scrutora traces personal data to every external API call, webhook, and third-party integration in your codebase, so your DPA list matches what your code actually sends out.

Share this article