ARA-C01 Question 143
Single answerSharing using Snowflake Data Clean RoomsA retail company wants to collaborate with a media partner to measure overlap between customer purchases and ad exposure, but neither party is allowed to share raw customer-level data with the other. The retail company also wants to ensure the media partner can run only approved analyses and see only privacy-safe aggregated results. Which Snowflake architecture best meets these requirements?
- A
Create a Snowflake Data Clean Room collaboration so each party contributes governed datasets and only preapproved analysis templates can be executed, with results returned in an approved privacy-safe form.
- B
Use a standard secure data share to give the media partner direct read access to the retailer's customer table, then rely on masking policies to prevent exposure of sensitive rows during joins.
- C
Replicate both parties' datasets into a shared database owned by the retailer and restrict access with role-based access control so analysts can build custom SQL as needed.
- D
Publish the retailer's data in a private listing and allow the media partner to subscribe, because listings automatically prevent all access to row-level data and limit consumers to overlap analysis only.
Show answer and explanation
Correct answer: A
Explanation
The best answer is to use Snowflake Data Clean Rooms because the scenario requires controlled collaboration between two parties that do not want to reveal raw customer-level data. This is a classic clean-room use case: identity or audience overlap, measurement, and joint analysis under strict governance. In Snowflake, standard data sharing and listings are powerful distribution mechanisms, but they are not substitutes for a clean room when the business requirement is governed multi-party analysis with privacy protections and restricted outputs. From an architect perspective, the key decision point is matching the collaboration pattern to the platform capability: use secure shares or listings for data distribution, and use Data Clean Rooms when you need privacy-enhanced analysis, governed execution, and limited result exposure across organizations.
- A. Correct.
Correct. Snowflake Data Clean Rooms are designed for privacy-enhanced collaboration where multiple parties can contribute data without exposing raw records directly to one another. A clean room can enforce approved analysis patterns and governance controls so collaborators execute only sanctioned queries or templates, with output limited to permitted results such as aggregated, privacy-safe measurements. This aligns with the stated need for overlap analysis, no raw data sharing, and controlled execution.
- B. Incorrect.
Incorrect. A standard secure share is for sharing database objects, but it does not by itself create a privacy-preserving collaborative environment for mutual analysis. Even with masking policies, the consumer is still being given access to shared objects from the provider. This does not satisfy the requirement that neither party expose raw customer-level data to the other, especially in a bidirectional matching or overlap analysis scenario.
- C. Incorrect.
Incorrect. Copying both datasets into a shared database increases data movement and does not provide the governance model expected for privacy-enhanced cross-party collaboration. RBAC can restrict who accesses objects, but it does not inherently enforce clean-room-style controls such as limiting collaborators to approved analytical workflows and preventing raw data exposure between parties.
- D. Incorrect.
Incorrect. Listings are used for data distribution and discovery, including private data products, but they are not the same as Snowflake Data Clean Rooms. Subscribing to a listing gives access to the published product under the provider's sharing model; it does not automatically constrain collaboration to only overlap analysis or guarantee the type of query-governance and privacy-preserving execution model required here.