DAA-C01 Question 38
Single answerFind external data sets that correlate with available dataA retail analytics team stores daily store-level sales in Snowflake and wants to determine whether local weather patterns and demographic trends correlate with revenue changes across regions. The team wants the fastest way to evaluate external data before committing engineering effort to ingest and model it. Which approach should the data analyst take in Snowflake?
- A
Use Snowflake Marketplace to discover weather and demographic data sets, query provider-shared data directly in the account, and join it to internal sales data on common dimensions such as location and date.
- B
Export internal sales data to CSV, download public weather files from external websites, and compare the files in a desktop BI tool before deciding whether to use Snowflake.
- C
Create external tables over arbitrary public URLs for weather and demographic websites, because Snowflake can automatically infer relational joins across those sites without provider sharing.
- D
Train a forecasting model on internal sales data first, because external data sets should only be evaluated after a machine learning baseline has been deployed to production.
Show answer and explanation
Correct answer: A
Explanation
The best answer is to use Snowflake Marketplace or direct data sharing to find external data sets that can be queried immediately alongside internal sales data. In real projects, analysts commonly enrich internal facts with third-party data such as weather, demographics, mobility, or economic indicators, then test correlation or business impact by joining on shared dimensions like date and geography. Snowflake's data sharing model lets consumers access provider data without copying or building custom ingestion pipelines, which is ideal for rapid evaluation. This aligns with Snowflake best practices around data collaboration, governed access, and minimizing unnecessary data movement. Relevant Snowflake documentation areas include Snowflake Marketplace, Secure Data Sharing, and working with shared databases. External tables are useful for files in cloud object storage, but they are not a substitute for discovering curated third-party data products or for automatic correlation logic.
- A. Correct.
Correct. Snowflake Marketplace and provider shares are designed to let consumers discover and access third-party data without building ingestion pipelines first. This is the most efficient way to evaluate correlation with internal data because the analyst can query shared data in place and join it with existing Snowflake tables using business keys such as geography, postal code, latitude/longitude-derived region, and date. This matches the scenario's goal of quickly assessing value before investing in engineering work.
- B. Incorrect.
Incorrect. Although exporting and manually comparing files is possible, it is not the fastest or most scalable Snowflake-native approach. It introduces unnecessary data movement, governance risk, and manual preparation. A common misconception is that external data evaluation must begin outside Snowflake, but Snowflake's data sharing and Marketplace capabilities are intended to reduce that friction.
- C. Incorrect.
Incorrect. External tables can reference data in cloud storage that your organization can access, but they do not work by pointing to arbitrary public websites and they do not automatically infer joins across unrelated web sources. This option mixes up external tables with data sharing and overstates Snowflake's capabilities. Correlation analysis still requires the analyst to identify join keys or derive matching dimensions.
- D. Incorrect.
Incorrect. Building a forecasting model first does not address the immediate need, which is to identify and assess potentially relevant external data sources. In practice, analysts often evaluate candidate external signals before model development. This option reflects a process misconception: machine learning does not replace the need to source and validate correlated external data.