Certified Data Privacy Solutions Engineer (CDPSE)

Last checked October 22, 2026 against the official ISACA exam page.

1. Question 1Privacy Engineering

   Which privacy engineering practice reduces the amount of personal data a system holds?

   1. ACollecting every possible field at signup, so that future features never need new data collection
   2. BData minimisation by design, collecting only the fields that the feature actually requires
   3. CKeeping a copy of every API request indefinitely, in case the logs prove useful for a future audit
   4. DDuplicating the customer table in three locations, so that the personal data is safer from loss
   Show answer and explanationHide answer

   Correct answer: B

   Data minimisation by design collects only what the feature needs, which reduces the data held. Collecting everything, indefinite logging and duplicating tables each increase the data held.

2. Question 2Privacy Risk Management and Compliance

   A privacy risk assessment finds a high likelihood of harm to customers from a new feature. What is the appropriate response?

   1. AReduce the risk with controls that lower its likelihood or impact, and document any residual risk for approval
   2. BAccept the risk without review, since the feature is commercially important to the company's plans
   3. CLaunch the feature first, and assess the risk only after customers have reported harm from the feature
   4. DHide the assessment result from the product team, so that the launch date is not affected by the findings
   Show answer and explanationHide answer

   Correct answer: A

   Reducing the risk with controls and documenting residual risk for approval is the appropriate response. Accepting without review, launching first and hiding findings each ignore the risk.

3. Question 3Data Lifecycle Management

   Which principle should guide how long personal data is kept?

   1. AKeep personal data for exactly ten years, regardless of the purpose for which it was collected
   2. BKeep all personal data indefinitely, in case it becomes useful for an unknown purpose later
   3. CRetain personal data only as long as needed for the stated purpose, then securely delete it
   4. DDelete personal data on the day it is collected, even when the purpose requires it for years
   Show answer and explanationHide answer

   Correct answer: C

   Retention should match the stated purpose, followed by secure deletion. Indefinite retention, a fixed period regardless of purpose and immediate deletion each break the principle.

4. Question 4Privacy Governance

   A company is launching a new app that collects location data. Which governance step should come FIRST?

   1. ADesign the app's home screen and the colour scheme that the marketing team has proposed
   2. BChoose the database vendor that will store the location records for the first year of service
   3. CComplete a privacy impact assessment that defines the purpose, data needed and legal basis
   4. DStart collecting the location data, and document the purpose once the data volume is known
   Show answer and explanationHide answer

   Correct answer: C

   A privacy impact assessment defines the purpose, data needs and legal basis before collection, which governs every later decision. Vendor choice, interface design and collecting first each come before the purpose is defined.

5. Question 5Privacy Engineering

   Which technique replaces direct identifiers with a reference that cannot be linked back without a separate key?

   1. APseudonymisation, which replaces identifiers with pseudonyms held separately from the data
   2. BEncryption with the key stored beside the data, so that the data can be decrypted by anyone reading it
   3. CHashing with no salt on the names, which can be reversed easily using a precomputed table
   4. DDeleting the identifiers from the file and keeping the original file in the same folder as the data
   Show answer and explanationHide answer

   Correct answer: A

   Pseudonymisation replaces identifiers with pseudonyms, with the link held separately. Co-located keys, unsalted hashes of names and keeping originals beside the data all make re-identification easy.