Data Mapping and Records of Processing
Data mapping, the inventory of what personal information an organization holds and why, underpins almost every Canadian privacy obligation. It supports Law 25 assessments and portability, breach reporting, 30-day access responses and retention schedules, but only if the map is kept current as systems change.

Key takeaways
- PIPEDA's accountability principle requires policies and practices that a data map makes demonstrable.
- Law 25 assessments, transfer agreements and portability all depend on knowing where data sits.
- Breach records must be kept for 24 months under the federal regulations.
- Access requests must be answered within 30 days under PIPEDA.
- Retention schedules should set minimum and maximum periods.
- Maps decay quickly unless updates are built into procurement and change management.
Why data mapping matters
Data mapping is the process of building and maintaining an inventory of the personal information an organization holds: what it is, where it comes from, why it is collected, where it is stored, who can access it, who it is shared with and when it is destroyed. Canadian privacy statutes rarely use the phrase, but almost every obligation they impose assumes that the organization knows the answers. An organization cannot identify its purposes, respond to an access request within 30 days, assess a breach or meet retention rules for information it cannot find.

This insight explains how records of processing support compliance with Quebec's Law 25 and the federal Personal Information Protection and Electronic Documents Act (PIPEDA), and how to build a map that stays accurate as systems change. It is general information, not legal advice.
The legal hooks for data mapping
PIPEDA's accountability principle (Principle 4.1.4 in Schedule 1) requires organizations to implement policies and practices that give effect to the fair information principles, including procedures to protect personal information, respond to complaints and inquiries, and train staff. Principle 4.2 requires purposes to be identified at or before collection, and Principle 4.5 limits use, disclosure and retention to those purposes. Each of these is far easier to show with a documented inventory.
Quebec's Law 25 goes further in practice. It requires businesses to adopt governance policies and practices, appoint a person in charge of the protection of personal information, conduct privacy impact assessments for information system projects and before communicating personal information outside Quebec, keep a register of confidentiality incidents and, since 22 September 2024, provide computerized personal information in a structured, commonly used technological format on request. All of these depend on knowing what is held and where.
What a records of processing inventory should capture
| Field | Why it matters |
|---|---|
| Data category and examples | Identifies sensitive information such as health, financial, biometric or children's data |
| Individuals concerned | Customers, employees, applicants, patients, users or minors |
| Source | Collected directly, from third parties or generated by the organization |
| Purpose and lawful basis | Links each use to identified purposes and consent or an exception |
| Systems and locations | Applications, databases, file shares, cloud services and jurisdictions |
| Access | Teams and roles with access, including administrators |
| Recipients | Service providers, affiliates, partners and any transfers outside Canada or Quebec |
| Retention and destruction | Minimum and maximum periods and method of destruction or anonymization |
| Safeguards | Encryption, access controls and other measures appropriate to sensitivity |
| Owner | The business person accountable for keeping the entry up to date |
Data mapping for Law 25 compliance
For organizations subject to Quebec's private-sector law, data mapping is the practical foundation of the Law 25 program. The map shows which projects trigger a privacy impact assessment, which data flows leave Quebec and need a transfer assessment and written agreement, where sensitive information requires express consent for secondary uses, and which technology collects information through identification, location or profiling functions that must be disclosed in advance. The Commission d'accès à l'information summarizes these obligations, and its inspectors are likely to ask for evidence of how an organization identified them.
Portability is a particular driver. To provide computerized personal information collected from an individual in a structured, commonly used format, an organization needs to know which systems hold that individual's information and how to extract it consistently.
Breach response
When an incident occurs, the first questions are what information was affected, whose it was and how sensitive it is. Under PIPEDA, an organization must report to the Commissioner and notify individuals of breaches that create a real risk of significant harm, and the Breach of Security Safeguards Regulations require records of every breach to be kept for 24 months. Quebec requires notice to the CAI and to affected individuals of incidents presenting a risk of serious injury, and a register of incidents. Without an up-to-date map, incident teams spend critical days reconstructing data flows that should already be documented. See our insight on breach notification.
Build your data map so it can answer breach questions quickly: for each system, record the categories of personal information, the approximate number of individuals, the provinces and countries where they live, and whether the data is encrypted. These are the facts regulators ask for first.
Access and correction requests
Under PIPEDA subsection 8(3), an organization must respond to an access request with due diligence and in any case within 30 days, subject to limited extensions. Quebec also imposes a 30-day response period. Meeting these deadlines requires a reliable way to find an individual's information across systems, including email, customer relationship management platforms, support tickets, backups and vendor systems. Data mapping identifies those systems in advance and allows organizations to build repeatable search procedures rather than improvising each time.
Retention and destruction schedules
PIPEDA Principle 4.5.2 says organizations should develop guidelines and procedures for retention, including minimum and maximum retention periods, and Principle 4.5.3 says information no longer required for the identified purposes should be destroyed, erased or made anonymous. Quebec's law similarly requires destruction or anonymization once the purposes are achieved, and anonymization must meet criteria set by regulation. A data map links each data set to a retention rule and an owner, which turns a paper policy into something that can be enforced and audited.
Retention schedules should reflect legal retention requirements, such as tax, employment and limitation periods, as well as business needs. Over-retention is a common finding after breaches: old data that should have been deleted is often the bulk of what was exposed.
Gap analysis
Once the map exists, it becomes the basis for a gap analysis against applicable laws. Typical findings include purposes that were never communicated to individuals, sensitive data collected without express consent, shadow IT systems outside security controls, service providers without adequate contracts, cross-border transfers without assessments, and data kept far beyond its retention period. Prioritizing gaps by risk, combining sensitivity, volume and likelihood of harm, gives management a defensible remediation plan.
How to build the map
- Secure executive sponsorship and appoint a privacy lead and business owners.
- Start from systems: use IT asset inventories, cloud consoles and procurement records.
- Interview business units with a short questionnaire based on the fields above.
- Use discovery tools to find unstructured data in file shares and collaboration platforms.
- Validate entries with system owners and security teams.
- Record the map in a tool that supports updates, ownership and reporting.
- Run the gap analysis and agree a remediation plan.
Organizations do not need to map everything at once. Starting with the highest-risk systems, such as customer databases, HR systems and any repository of sensitive data, delivers most of the value early.
Maintaining the map as systems change
The most common failure in data mapping is decay. A map completed for a compliance project quickly becomes outdated as new tools are adopted, vendors change, products launch and teams reorganize. Keeping it current requires embedding updates into existing processes: procurement and vendor onboarding, privacy impact assessments, change management, product launches and system decommissioning. Annual attestation by business owners, with spot checks by the privacy team, helps keep entries accurate.
Measuring the program
Boards and senior management increasingly ask for evidence that privacy programs work. Useful measures drawn from the data map include the percentage of systems with a named owner and a current entry, the number of data sets past their retention period, the number of vendors without privacy terms, the time taken to locate data for access requests, and the number of high-risk gaps still open. Reporting these figures regularly keeps the map visible and funded.
AI tools and new data flows
Generative AI tools, analytics platforms and software development kits often create data flows that bypass traditional inventories. Employees may paste personal information into public AI tools, and vendors may add AI features that send data to new sub-processors. Data mapping should capture these flows and identify whether information is used to train models. See our insight on generative AI policies for employees.
Data mapping and cybersecurity
Privacy and security teams benefit from a shared map. Security teams need to know where the most sensitive information resides to prioritize controls, monitoring and testing. Privacy teams need to know which safeguards protect each data set. Combining the privacy inventory with the IT asset register avoids duplicated effort and gives incident response teams a single source of truth. See our cybersecurity practice.
Using the map for privacy notices
Privacy policies are only as accurate as the knowledge behind them. PIPEDA's openness principle requires organizations to make information about their policies and practices readily available, and Quebec requires organizations that collect personal information by technological means to publish a privacy policy in clear and simple terms and to notify individuals of changes. A current data map allows the privacy team to check that published notices describe the actual categories collected, the purposes, the use of service providers, transfers outside Canada or Quebec and retention practices. When the map changes, for example because a new vendor is onboarded or a new analytics tool is deployed, the notice should be reviewed at the same time.
Employee and HR data
HR data is often overlooked in customer-focused mapping exercises, yet it includes some of the most sensitive information an organization holds: health and accommodation records, background checks, payroll and banking details, performance assessments and, increasingly, monitoring data from devices and collaboration tools. Which law applies depends on the employer and province. PIPEDA covers employee information of federal works, undertakings and businesses, while Quebec, Alberta and British Columbia have private-sector statutes that reach provincially regulated employers. A complete map should cover the whole employee life cycle, from recruitment to post-employment retention.
Mergers, acquisitions and restructurings
In a transaction, a target's data map is a valuable diligence document. It allows a buyer to assess privacy risk, understand what personal information will transfer, identify consents and notices that may limit future use, and plan integration. Organizations that maintain a reliable map can answer diligence questions faster and with more credibility, which can matter in negotiations over representations, indemnities and purchase price. After closing, the map guides integration and the retirement of duplicate systems.
Common pitfalls in data mapping
- Treating the map as a one-time project rather than a living record.
- Mapping only structured databases and missing email, file shares and backups.
- Recording systems without purposes, making it impossible to test purpose limitation.
- Omitting service providers and sub-processors.
- Failing to assign owners who are accountable for updates.
- Collecting so much detail that the map is never finished.
How TRW & Co helps
We help organizations design and run data mapping programs that support Law 25, PIPEDA and provincial privacy obligations. Our work includes inventory templates and questionnaires, gap analyses, retention and destruction schedules, cross-border transfer assessments, access request procedures and breach readiness. See our privacy and data protection page or contact us.
Frequently asked questions
Is a data map legally required in Canada?
Private-sector statutes do not generally use the term, but obligations under PIPEDA and Quebec's Law 25, including accountability, purpose identification, privacy impact assessments, breach reporting, access and portability, are very difficult to meet or demonstrate without one.
What should a data map include?
Data categories, individuals concerned, sources, purposes and lawful basis, systems and locations, access, recipients including service providers, retention periods, safeguards and an accountable owner.
How does data mapping help with breaches?
It lets incident teams quickly identify what information was affected, whose it was and how sensitive it is, which drives the real risk of significant harm assessment and notices to regulators and individuals.
How often should the map be updated?
Continuously through procurement, privacy impact assessments, product launches and decommissioning, with an annual attestation by business owners.
Where should we start?
With the highest-risk systems: customer databases, HR systems and any repository of sensitive information such as health, financial or biometric data.
Sources
General information only, not legal advice. References to statutes, courts and authorities describe published law and practice at the date shown.


