Indian insurers have roughly six months to get their data practices in order. Most of the DPDP Act's duties, rights and penalty provisions take effect on 13 May 2027, and the Act allows penalties of up to ₹250 crore for failing to maintain reasonable security safeguards.
Few sectors hold data as sensitive as insurance. A single policy file can contain medical history, income, family details, nominee information and claim records. This guide covers what needs to change and in what order.
Where the DPDP timeline stands
The DPDP Rules, 2025 were notified in November 2025 and are being phased in:
- November 2025: the Data Protection Board and the definitions came into force.
- 13 November 2026: Rule 4 begins, so Consent Managers can now register with the Board.
- 13 May 2027: the main obligations apply, including notice, consent, data principal rights, breach reporting and penalties.
Ordinary companies have no new duty on the November date, and no insurer is required to use a Consent Manager. The real deadline for insurers is May 2027.
Why insurance is harder than most sectors
Insurers don't collect data in one place. It moves across proposal forms, agents, bancassurance partners, medical examiners, hospitals, third-party administrators (TPAs), surveyors and reinsurers.
Under the Act, the insurer is the Data Fiduciary and stays accountable for personal data its processors handle. A TPA's weak controls become your problem.
Health underwriting adds a second tension. The Act expects data minimisation and purpose limitation, but underwriting depends on collecting detailed information. Bar & Bench has examined this in detail. You can't remove the tension, but you can document why each data point is necessary.
Step 1: Map every data flow
You can't protect or justify data you haven't found. Start with an inventory that covers:
- What personal data you collect at each stage, from lead to renewal to claim
- Where it sits: core policy systems, CRMs, email, call recordings, WhatsApp, spreadsheets held by sales teams
- Who receives it: TPAs, brokers, POSPs, surveyors, hospitals, reinsurers, cloud and analytics vendors
- Which purpose each data point serves
Legacy systems and agent-held files are where most gaps turn up. If data discovery looks heavy, run it in phases, starting with health and claims data.
Step 2: Rebuild consent around purpose
The Act requires consent that is free, specific, informed, unambiguous and easy to withdraw. A single "I agree to everything" tick box on a proposal form won't meet that standard.
Practical changes include:
- Separate purposes. Underwriting, claims processing, marketing and cross-selling each need their own clearly described purpose.
- Plain-language notices. Say what you collect and why, and explain how to withdraw consent and complain. Offer the notice in the customer's language.
- Withdrawal as easy as giving consent. If someone can agree in one tap, they should be able to withdraw in about the same effort.
- A consent record you can produce. You need to show who consented, to what, when, and which notice version they saw.
Not every processing activity needs consent. Some are covered by "legitimate uses", such as complying with a legal obligation. Have counsel classify each purpose rather than treating everything as consent-based.
This is where a consent management platform helps in practice. It captures consent at each touchpoint, keeps an auditable trail and carries withdrawals through to downstream systems. Without that automation, manual consent tracking across several channels becomes hard to defend in an inquiry.
Step 3: Reconcile retention with erasure
Insurers must retain some records for regulatory and claims purposes, while the Act expects personal data to be erased once its purpose is served, unless law requires otherwise.
Build a retention schedule by data category and record the legal basis for keeping each one. For example:
- Claims files kept for statutory periods, with the supporting provision noted
- Lapsed-lead data with no legal retention basis, deleted on a fixed timeline
- Marketing data, removed when consent is withdrawn
Auditors will ask for this schedule, so document it now.
Step 4: Tighten your processor chain
Treat each TPA, broker and vendor as part of your own compliance perimeter:
- Review contracts for security safeguards, purpose limits, sub-processor controls and breach-notification duties.
- Require processors to alert you of an incident quickly, since your own reporting clock starts when you become aware.
- Audit high-risk processors, starting with those that hold health data.
- Ask whether each one needs the data it receives at all.
Step 5: Prepare for breaches before they happen
The Rules require notifying affected individuals and sending the Board a detailed report within 72 hours. Failure to notify can attract penalties of up to ₹200 crore.
IRDAI's cyber-security norms have their own reporting expectations, so build one incident workflow that satisfies both regulators. At minimum:
- Encrypt personal data at rest and in transit, with role-based access.
- Keep logs long enough to investigate an incident.
- Name who decides, who drafts and who notifies.
- Run a tabletop exercise before May 2027.
Step 6: Operationalise data principal rights
Policyholders can request access, correction and erasure, and can nominate someone to exercise their rights. They can also raise grievances. Set up:
- A single intake channel, such as a portal or email, that logs every request
- Identity verification before data is shared
- A defined owner and turnaround time for each request type
- A grievance process that ends with escalation to the Board
Step 7: Handle minors' data correctly
Child and endowment plans involve data about minors. The Act requires verifiable parental or guardian consent before processing a child's personal data, and tracking and behavioural monitoring of children is restricted. Review your onboarding flows for these products first.
Step 8: Assign ownership
Compliance fails when nobody owns it. Appoint an accountable leader, define who handles privacy reviews for new products, and train frontline staff and agents.
Larger insurers could be designated Significant Data Fiduciaries, which carries extra duties such as a Data Protection Officer based in India, periodic audits and impact assessments. Plan as if that may apply to you.
A six-month plan
MonthFocusNov–Dec 2026Data mapping, gap assessment, processor inventoryJan 2027Consent redesign, new notices, retention scheduleFeb 2027Consent tooling live, processor contracts updatedMar 2027Rights-request and grievance workflows, breach playbookApr 2027Testing, staff and agent training, tabletop exerciseMay 2027Final audit and go-live readinessOne distinction to get right
A Consent Manager is a company registered with the Data Protection Board that lets individuals manage consent across organisations. A consent management platform is the tooling you use internally to collect, record and honour consent. They are different things. Insurers need the second, while the first is optional.
Takeaway
Insurers that do this well start with data mapping and consent, then work outward to retention, vendors and incident response. Pick the three riskiest data flows, usually health, claims and TPA sharing, and fix those first. Then widen the programme before May 2027.
This article is for general information and isn't legal advice. Consult qualified counsel for your organisation's specific obligations.