How do you write a brief or RFP for a design thinking consultancy in India?
November 19, 2026
July 2, 2026
Free, specific, informed, unconditional and unambiguous, given by a clear affirmative action. That means no bundling consent with unrelated services, one consent per purpose, a notice available in English or any of the 22 scheduled languages, and no pre-ticked boxes or dark patterns. Withdrawal must be as easy as consent, and you must be able to prove what each user consented to, when and under which notice version.
consent = true column cannot answer an inspection; an append-only, tamper-evident consent ledger can.Ask most Indian businesses how they track user consent and the honest answer is some combination of a signup checkbox, a terms-of-service page, and — if anyone asks — a database column that says consent = true. Under the Digital Personal Data Protection Act, 2023, that answer is now a liability measured in crores.
Consent under the DPDPA must be free, specific, informed, unconditional, and unambiguous, signaled by a clear affirmative action. Unpacking each word reveals an operational requirement:
And the obligation that changes system design the most: withdrawal must be as easy as consent. When a user withdraws, processing must stop, downstream processors must be told, and data no longer needed must go — all provably.
Here is the question that kills spreadsheet-based compliance: *for any given user, at any given moment, can you prove what they consented to, when, through which notice text, in which language, and whether that consent is still valid?*
An inspection or a Data Protection Board inquiry will not ask whether you have a privacy policy. It will ask for the consent record behind a specific complaint — the notice version shown, the timestamp, the affirmative action taken, every purpose attached, and the full lifecycle since: renewals, expiries, withdrawals, and what your systems did about each. A spreadsheet cannot answer that. Neither can a boolean column. What answers it is a consent ledger: an append-only, tamper-evident record tied to notice versions and purposes, queryable in real time by every system that processes personal data.
Phase 2 of the DPDP Rules activates Rule 4: Consent Managers — registered platforms through which individuals can give, review, and withdraw consent across many Data Fiduciaries from one place, on the account-aggregator pattern already familiar in Indian fintech.
From November 13, 2026, if you rely on consent, you should assume some of your users will manage it through a registered Consent Manager rather than your own UI. Your systems must be able to ingest externally managed consent state, honor withdrawals that arrive via API rather than your app, and keep your ledger consistent across both channels. Retrofitting that onto a checkbox architecture is far more expensive than building on a consent service from the start.
A DPDPA-grade consent architecture has five layers:
You can build this — several quarters of engineering plus permanent maintenance as rules evolve. Or you can connect it. Data Adhikaar by Qodequay Technologies ships consent as a fabric: the Sammati agent captures, refreshes, expires, and proves consent across channels; notices are drafted in English plus all 22 scheduled languages; consent checks run at p95 latency under 100 ms; and every event lands in a tamper-evident Evidence Vault. SDKs, 25+ connectors, and an MCP server mean the first passing consent check takes minutes, not months.
November 2026 is the consent deadline that matters. Run the free readiness assessment and find out how far your current consent stack is from provable.
Because an inquiry asks for the record behind a specific complaint: the notice version shown, the timestamp, the affirmative action, every purpose attached, and the full lifecycle of renewals, expiries and withdrawals. A spreadsheet or boolean column cannot answer that; an append-only, tamper-evident consent ledger tied to notice versions can.
No. Consent must be specific, which means one consent per purpose. A single checkbox covering marketing, analytics and third-party sharing fails, as do pre-ticked boxes, dark patterns, and consent inferred from silence or inactivity. Each request must also carry a notice describing the data, purpose, withdrawal route and complaint route.
Withdrawal must be as easy as giving consent. When a user withdraws, processing must stop, downstream processors must be told, and data no longer needed must be deleted, all provably. This is the obligation that changes system design the most, which is why consent needs a real-time check every processing system calls.
Phase 2 of the DPDP Rules activates Rule 4: registered Consent Managers through which individuals can give, review and withdraw consent across many Data Fiduciaries. From that date you should assume some users manage consent outside your UI, so your systems must ingest external consent state and honour withdrawals arriving via API.
Five layers: versioned notice management per purpose and language with an approval trail; capture at the moment of affirmative action across web, app, call centre and offline channels; a low-latency real-time consent check; lifecycle automation for renewal, expiry and withdrawal propagation; and hash-chained, exportable evidence for an auditor or the Board.
General guidance on the DPDP Act, 2023 — not legal advice.
Monthly insights on AI, VR and DPDPA compliance — straight from our team to your inbox.
Free 30-minute consultation with our team — or see our products in action.