All insights
Security

Apr 28, 2026 · 11 min read

Sri Lanka PDPA 2022: What Software Teams Must Do Before the Next Compliance Window

A practical breakdown of the Personal Data Protection Act for developers, architects, and product owners building or operating software in Sri Lanka.

CE

CodeLK Engineering

Sri Lanka's Personal Data Protection Act (No. 9 of 2022) moves personal data from a contractual nicety to a statutory obligation. For software teams, this means data architecture decisions that were previously engineering concerns are now compliance concerns. The gap between "we know where the data is" and "we can prove it, report on it, and respond to a data subject request in 14 days" is where most organisations discover their actual exposure.

What the PDPA requires from software

The Act requires controllers (organisations that determine the purpose and means of processing) to have a lawful basis for every collection and processing activity. For most software systems this means: (1) identifying all data flows, (2) mapping each flow to a lawful basis (consent, contract, legal obligation, or legitimate interest), and (3) documenting this in a Record of Processing Activities (ROPA).

  • Consent mechanisms must be specific, informed, and freely given. Pre-ticked boxes, bundled consent, and vague "we may use your data to improve services" language do not meet the standard.
  • Data subject rights — access, rectification, erasure, restriction, and portability — must be actionable. If your database design makes it impossible to locate all records relating to one person, you have a compliance gap independent of your privacy policy.
  • Breach notification within 72 hours to the Data Protection Authority (when required) demands an incident response plan that engineers, not just legal, have rehearsed.

Practical software changes PDPA drives

  • 1. Data discovery and mapping: tag sensitive fields in your schema (NIC, phone, health data, financial records) and document what system processes them and why.
  • 2. Retention enforcement: implement automatic deletion or anonymisation when retention periods expire. Manual processes for data deletion do not scale and create audit risk.
  • 3. Consent audit trails: log when, how, and what version of consent was given — not just a boolean "consented" flag.
  • 4. Erasure workflows: design deletion to cascade correctly across primary tables, backups, and audit logs — distinguishing what must be deleted from what must be retained for legal hold.
  • 5. Cross-border transfer controls: if your application sends data to cloud services, analytics platforms, or APIs hosted outside Sri Lanka, document and assess this under the Act's transfer restrictions.

PDPA compliance is an architecture exercise, not just a legal one. Start with a data map, add retention enforcement, and build consent audit trails — these create the foundation everything else depends on.

Get in touch

Let's build your next system together.

Tell us about the system you need — we'll reply within one business day. Or write to biz@codelk.com.

Contact us