Snowflake · DAA-C01
SnowPro Advanced: Data Analyst (DAA-C01) Practice Tests
Validates advanced skills in applying comprehensive data analysis principles using Snowflake, including data ingestion, advanced SQL, UDFs, and predictive and diagnostic analytics. Requires 1+ years of hands-on Snowflake experience and an active SnowPro Core certification.
Exam Details
Practice Questions
600
≈ 6 practice exams
Duration
115 minutes
Passing Score
750/1000
Difficulty
ProfessionalLast Updated
Jun 2026
Topics Covered
Data Ingestion and PreparationData TransformationAdvanced SQL Query DevelopmentSnowflake Built-in Functions and UDFsDescriptive and Diagnostic Data AnalysisPredictive Data AnalysisData Presentation and Business Requirements
SnowPro Advanced: Data Analyst (DAA-C01) Practice Exam Preparation
Use this DAA-C01 practice exam to prepare for SnowPro Advanced: Data Analyst (DAA-C01) with realistic questions, detailed explanations, and focused study modes. The practice bank includes 600 questions for Snowflake DAA-C01, so you can review the exam steadily instead of relying on one long cram session.
As you practice, pay extra attention to recurring topics such as Data Ingestion and Preparation, Data Transformation, Advanced SQL Query Development, Snowflake Built-in Functions and UDFs, and Descriptive and Diagnostic Data Analysis. Start with short sessions to identify weak areas, then move into timed quizzes once your accuracy is consistent.
The explanations are especially useful when you want to connect exam wording to the responsibilities and scenarios described in the official certification guidance. Use the free preview first, then unlock the full question bank when you are ready to build a complete study routine.
Exam Overview
The SnowPro® Advanced: Data Analyst (DAA-C01) certification is Snowflake's advanced-tier credential for data professionals who apply comprehensive analytical techniques within the Snowflake Data Cloud. The exam validates expertise across the full analytical workflow: acquiring and loading structured, semi-structured, and unstructured data; transforming and preparing it for analysis; writing and optimizing complex SQL; leveraging Snowflake built-in functions and user-defined functions (UDFs); and performing descriptive, diagnostic, and predictive analytics to surface business insights.
Unlike entry-level Snowflake certifications, this exam expects candidates to demonstrate production-grade fluency with Snowflake-specific features such as Snowsight dashboards, Time Travel, semi-structured data handling (JSON, Parquet, Avro, ORC, XML), query optimization via partition pruning and materialized views, and integration with BI tools including Tableau, Looker, and Power BI. Delivered by Pearson VUE, the exam consists of 65 scored multiple-choice and multiple-select questions to be completed in 115 minutes, with a passing score of 750 on a scaled 0–1000 range. The registration fee is $375 USD per attempt (a 20% regional discount applies in India).
Who Should Take This Exam
This certification is designed for working data professionals who spend the majority of their time querying, transforming, and analyzing data inside Snowflake. Relevant job titles include Data Analyst, Analytics Engineer, Business Intelligence Developer, ELT Developer, and BI Professional. Candidates typically have responsibility for building analytical pipelines, authoring complex SQL reports, creating dashboards in Snowsight or connected BI tools, and communicating data-driven findings to business stakeholders.
The exam is appropriate for those who have moved beyond foundational Snowflake usage and are ready to demonstrate advanced competency in forecasting, UDFs, and diagnostic analytics. It is not intended for platform administrators or data engineers focused primarily on infrastructure — those roles are better served by the SnowPro Advanced: Administrator or SnowPro Advanced: Data Engineer tracks.
Prerequisites
Snowflake requires candidates to hold an active SnowPro Core Certification before sitting for the DAA-C01 exam. The SnowPro Core validates foundational knowledge of Snowflake architecture, data loading, querying, and account management, and its concepts are assumed knowledge for the advanced exam. There is no formal academic prerequisite, but Snowflake recommends candidates have at least one year of hands-on, production-level experience working as a Data Analyst in the Snowflake Data Cloud.
Beyond the formal requirement, candidates should be fluent in advanced SQL (window functions, CTEs, recursive queries, analytic functions) and have practical experience with at least one scripting or programming language for UDF development (Python, JavaScript, or Java are common choices). Familiarity with data modeling concepts, semi-structured data formats (JSON, Parquet), and at least one BI or visualization tool is strongly recommended before attempting the exam.
Exam Format
The DAA-C01 exam consists of 65 questions in multiple-choice (single correct answer) and multiple-select (two or more correct answers) formats. Candidates have 115 minutes to complete the exam. It is delivered online through Pearson VUE, which supports both remote proctored (OnVUE) and in-person testing center formats. The exam is currently offered in English only.
Scoring uses a scaled system ranging from 0 to 1000; a minimum scaled score of 750 is required to pass. Snowflake does not publish a fixed percentage-correct passing threshold, as the scaled score accounts for question difficulty. There is no published limit on retake attempts, and each attempt requires the full $375 USD registration fee. Snowflake recommends downloading the official exam study guide from the certification page prior to registration, as it contains the most current domain objectives and any updates to exam content.
Skills Measured
- 1.Domain 1 – Data Ingestion and Preparation (approx. 15–20%): Retrieving structured (CSV), semi-structured (JSON, Parquet, Avro, ORC, XML), unstructured, and synthetically generated data; performing data discovery to assess available datasets and identify elements required for business goals; profiling data for quality, granularity, and completeness; enforcing data integrity constraints; and using Snowflake's COPY INTO, Snowpipe, and internal/external stages for data loading.
- 2.Domain 2 – Data Transformation (approx. 15–20%): Cleaning and reshaping raw data using SQL transformations and Snowflake's FLATTEN, PARSE_JSON, and LATERAL JOIN capabilities for semi-structured sources; designing and implementing scalable transformation pipelines; applying data normalization and denormalization techniques; and building reusable transformation logic using views, stored procedures, and tasks.
- 3.Domain 3 – Advanced SQL Query Development (approx. 20–25%): Writing and troubleshooting complex queries using window functions, CTEs, subqueries, and recursive SQL; optimizing query performance through partition pruning, clustering keys, materialized views, and result caching; handling data anomalies and NULLs; and interpreting query profiles in Snowsight to diagnose performance bottlenecks.
- 4.Domain 4 – Snowflake Built-in Functions and UDFs (approx. 15–20%): Applying date/time, string, numeric, conditional, and aggregate functions; working with semi-structured data functions (GET, GET_PATH, OBJECT_CONSTRUCT); creating and invoking scalar and tabular User-Defined Functions (UDFs) in SQL, Python, JavaScript, or Java; and using User-Defined Table Functions (UDTFs) for advanced analytical use cases.
- 5.Domain 5 – Descriptive and Diagnostic Data Analysis (approx. 10–15%): Summarizing datasets using statistical measures (mean, median, standard deviation, percentiles); identifying outliers, trends, and anomalies; using exploratory data analysis techniques within Snowflake; applying Snowflake's TABLESAMPLE for large-dataset profiling; and building descriptive reports and interactive dashboards in Snowsight.
- 6.Domain 6 – Predictive Data Analysis (approx. 10–15%): Applying time-series forecasting methods available natively in Snowflake (including Snowflake ML-powered forecasting functions); using regression techniques for predictive modeling; interpreting model outputs and communicating forecast uncertainty; and understanding when to leverage Snowpark for more advanced ML workflows.
- 7.Domain 7 – Data Presentation and Business Requirements (approx. 10–15%): Translating business questions into analytical requirements; exporting data in formats appropriate for downstream consumers (CSV, JSON, Parquet); connecting Snowflake to BI tools such as Tableau, Power BI, and Looker via Partner Connect or direct ODBC/JDBC; building stakeholder-facing dashboards and reports; and applying data storytelling principles to communicate insights effectively.
Study Tips
- Download and work through the official Snowflake SnowPro Advanced: Data Analyst Study Guide (available at learn.snowflake.com) before anything else — it defines the exact domain objectives and is the authoritative source for what will be tested.
- Build hands-on practice in a free Snowflake trial account: load semi-structured JSON and Parquet files using COPY INTO and Snowpipe, practice FLATTEN and PARSE_JSON queries, and profile the results in Snowsight's query editor to understand explain plans and partition pruning.
- Master window functions, CTEs, and analytic SQL — these appear heavily across multiple domains. Practice writing RANK, DENSE_RANK, LAG, LEAD, NTILE, and RATIO_TO_REPORT functions against realistic datasets, and focus on understanding when each is appropriate.
- Experiment with Snowflake's native ML forecasting functions (introduced in 2023–2024) and build at least one UDF in Python or JavaScript. The predictive analytics and UDF domains require applied experience that is difficult to gain through reading alone.
- Use Snowflake's free Hands-On Essentials and Data Warehousing Workshop labs on Snowflake University (learn.snowflake.com) to reinforce practical skills, and supplement with the Pluralsight SnowPro Advanced: Data Analyst learning path, which maps directly to DAA-C01 objectives.
- Review the SnowPro Core Certification study material as a baseline — the DAA-C01 builds on Core concepts such as micro-partitions, clustering, Time Travel, data sharing, and Snowflake architecture. Gaps in Core knowledge will surface during the advanced exam.
- Practice with timed mock exams (65 questions in 115 minutes ≈ 1.75 minutes per question). Focus especially on multiple-select questions, which require identifying all correct answers with no partial credit — a common source of lost points for candidates who know the material but rush.
Career Benefits
Earning the SnowPro Advanced: Data Analyst credential signals to employers that a candidate can operate independently at an advanced level on the Snowflake platform — a differentiated skill as Snowflake's market share in the cloud data warehouse space continues to grow across financial services, retail, healthcare, and technology sectors. Job titles commonly held by certified candidates include Senior Data Analyst, Analytics Engineer, BI Engineer, and Data Scientist, with US salaries for Snowflake-proficient analysts typically ranging from $95,000 to $145,000 depending on seniority, industry, and location.
The DAA-C01 sits within Snowflake's Advanced certification tier, positioning it above the SnowPro Core and SnowPro Associate credentials and alongside other Advanced specializations (Data Engineer, Architect, Administrator). Unlike vendor-neutral analytics certifications, the DAA-C01 directly validates platform-specific expertise that many employers explicitly list as a job requirement when Snowflake is the primary data platform. Holding both the SnowPro Core and an Advanced specialization is increasingly recognized as a strong signal of job-ready Snowflake proficiency in competitive data hiring pipelines.
Sample Questions
5 sample questions with answers and explanations. The full bank has 600 questions, enough for 6 full-length practice exams.
Preview — answers shown1. A data analyst at Tailspin Analytics runs a complex GROUP BY aggregation query on a 500 million row SALES_FACTS table that has had no DML activity in the past 6 hours. The first execution takes 38 seconds. Two hours later the analyst submits the identical unmodified query, but the virtual warehouse has been suspended in the interim and has not yet resumed. What result does the second execution produce, and why? (Select one!)
AThe query returns instantly from the 24-hour result set cache in the Cloud Services layer, consuming zero warehouse credits even with the warehouse suspended
BThe query returns instantly from the SSD warehouse cache, but the warehouse must be automatically resumed before cache contents stored on warehouse nodes can be accessed
CThe query returns instantly from the metadata cache, which stores pre-aggregated summaries of all GROUP BY combinations for tables with no recent DML
DThe query fails because no virtual warehouse is active to process the query and no cached result exists outside the warehouse tier
Explanation
Snowflake's result set cache (persisted query results cache) is managed entirely within the Cloud Services layer, which is architecturally separate from and independent of virtual warehouses. When an identical query is rerun against data that has not been modified (no DML or DDL changes since the prior execution), Snowflake serves the result directly from this cache without activating the virtual warehouse and without consuming any warehouse credits. The result set cache retains results for 24 hours from last use, and each reuse extends retention by another 24 hours up to a maximum of 31 days total. The SSD warehouse cache lives on the compute nodes of the virtual warehouse itself and is completely cleared when the warehouse is suspended — it cannot serve cached results after the warehouse goes offline. The metadata cache stores micro-partition statistics such as row counts, min/max values, and null counts to enable partition pruning; it does not store full query result sets or pre-aggregated summaries.
2. A data engineer at Litware Analytics has a VARIANT column named campaign_labels that stores JSON arrays of string values. Some rows contain populated arrays while others contain empty arrays ([]). The following query silently omits rows with empty arrays without raising any error: SELECT c.campaign_id, f.value::STRING AS label FROM campaigns c, LATERAL FLATTEN(INPUT => c.campaign_labels) f; The engineer needs all campaign rows to appear in the output, including those with empty arrays. Which change to the FLATTEN call will produce the correct result? (Select one!)
AWrap the LATERAL FLATTEN in a LEFT OUTER JOIN clause so that standard outer-join semantics force Snowflake to retain rows where the array is empty
BAdd OUTER => TRUE to the FLATTEN call so that source rows with empty or NULL input arrays are preserved in the output with NULL values in the flattened columns
CPass the campaign_labels column through PARSE_JSON before FLATTEN so that Snowflake recognizes empty arrays as valid input and retains the source rows
DAdd a pre-filter WHERE ARRAY_SIZE(campaign_labels) >= 0 so that Snowflake evaluates the array size before the lateral join and retains zero-length arrays
Explanation
FLATTEN defaults to OUTER => FALSE, which means source rows whose input array is empty ([]) or NULL produce zero output rows — the source row is silently dropped with no error. Adding OUTER => TRUE instructs FLATTEN to emit exactly one output row for each such source row, populating the KEY, INDEX, VALUE, SEQ, THIS, and PATH output columns with NULL values. This preserves the source row in the result set. Wrapping in LEFT OUTER JOIN is not valid Snowflake syntax for the FLATTEN table function; the OUTER parameter on the FLATTEN call itself is the correct and only supported mechanism. A WHERE clause using ARRAY_SIZE does not change FLATTEN behavior because FLATTEN processes the arrays during the lateral join phase before WHERE is evaluated — rows are already dropped before the WHERE clause runs. Passing campaign_labels through PARSE_JSON is unnecessary because FLATTEN already accepts both VARIANT and ARRAY typed inputs directly and the empty array is already valid VARIANT; the problem is the default OUTER => FALSE behavior, not the data type.
3. A data analyst at Northwind Analytics needs to build a single SQL query that produces sales totals at exactly four aggregation levels: (1) region, product_category, and quarter combined; (2) region and product_category subtotals; (3) region-only subtotals; and (4) a single grand total row. Which GROUP BY clause produces exactly these four groupings using the most concise syntax? (Select one!)
AGROUP BY GROUPING SETS((region, product_category, quarter), (region, product_category), (region), (), (quarter))
BGROUP BY CUBE(region, product_category, quarter)
CGROUP BY ROLLUP(region, quarter, product_category)
DGROUP BY ROLLUP(region, product_category, quarter)
Explanation
ROLLUP(region, product_category, quarter) produces exactly four hierarchical grouping levels: (region, product_category, quarter), (region, product_category), (region), and () — the grand total. This matches all four required aggregation levels in the most concise form. CUBE(region, product_category, quarter) produces 2³ = 8 groupings, covering every possible combination of the three dimensions, which includes unwanted groupings such as (product_category, quarter) and (quarter) alone that are not requested. GROUPING SETS with five entries would produce an additional (quarter) subtotal not required and is unnecessarily verbose. ROLLUP(region, quarter, product_category) follows a different hierarchy — producing (region, quarter, product_category), (region, quarter), (region), () — which does not include the required (region, product_category) subtotal level and instead includes (region, quarter) which was not requested.
4. A developer at Contoso Analytics is building a pricing transformation pipeline that reads product prices stored as VARCHAR in a staging table. The developer writes SELECT TO_NUMBER(price_string) FROM staging_prices to convert the values. When the column contains the string '29.99', the query returns 30 instead of 29.99. What is the cause and the correct fix? (Select one!)
ATO_NUMBER without explicit precision and scale arguments defaults to scale 0, discarding all decimal digits with rounding; use TO_NUMBER(price_string, 10, 2) to preserve two decimal places
BTO_NUMBER cannot parse strings containing a decimal point character and requires REPLACE to remove the period before conversion; use CAST(price_string AS FLOAT) for decimal values instead
CTO_NUMBER applies ROUND half-up rounding by default; replace it with FLOOR(TO_DOUBLE(price_string)) to truncate rather than round decimal values
DThe VARCHAR column width exceeds the implicit numeric precision limit; pre-cast the column to FLOAT with CAST(price_string AS FLOAT) before passing it to TO_NUMBER
Explanation
TO_NUMBER in Snowflake defaults to a scale of 0 when called with a single string argument and no precision or scale specification. A scale of 0 means zero digits are preserved after the decimal point, so '29.99' is rounded to the nearest integer, producing 30. To preserve fractional digits, the function requires explicit precision and scale arguments: TO_NUMBER(price_string, 10, 2) specifies a total precision of 10 digits and a scale of 2 decimal places, correctly returning 29.99. TO_DOUBLE is a valid alternative for floating-point representation, but it introduces IEEE 754 binary floating-point imprecision that makes it unsuitable for financial values requiring exact decimal representation. There is no prohibition on parsing strings that contain a decimal point; TO_NUMBER handles them correctly when scale is specified. Pre-casting to FLOAT before calling TO_NUMBER adds unnecessary conversion steps and still requires the scale argument to avoid the default scale-zero behavior.
5. A data modeling team at Litware Commerce is building a reporting aggregation layer that requires joining three source tables — ORDERS, CUSTOMERS, and PRODUCTS — and pre-computing aggregated metrics that refresh automatically as source data changes. The team is comparing Snowflake Materialized Views and Dynamic Tables. Which statement accurately describes a key architectural difference that should drive their choice? (Select one!)
ADynamic Tables cannot reference other Dynamic Tables as upstream source objects, preventing the construction of multi-stage transformation pipelines
BMaterialized Views in Snowflake are restricted to queries against a single base table with no joins, while Dynamic Tables support arbitrary SELECT statements including multi-table joins and aggregations
CMaterialized Views support multi-table joins and use incremental refresh more efficiently than Dynamic Tables for complex pre-aggregated reporting
DDynamic Tables require a minimum TARGET_LAG of five minutes, making them unsuitable for reporting layers that require near-real-time refresh
Explanation
Snowflake Materialized Views carry a significant architectural restriction: they support queries against only a single base table. Joins, unions, subqueries referencing multiple tables, and other multi-source SQL constructs are not permitted in a Materialized View definition. This makes Materialized Views unsuitable for use cases requiring data from multiple sources — attempting to create a Materialized View with a multi-table join will fail at creation time. Dynamic Tables support arbitrary SELECT statements. A Dynamic Table definition can include multi-table joins, aggregations, CTEs, window functions, and FLATTEN operations. Snowflake automatically tracks upstream dependencies on all referenced source objects and refreshes the Dynamic Table using INCREMENTAL, FULL, or AUTO mode based on the configured TARGET_LAG. For the team's use case joining ORDERS, CUSTOMERS, and PRODUCTS, Dynamic Tables are the correct choice. Dynamic Tables can reference other Dynamic Tables as their upstream source objects. This chaining capability is a key design feature for building layered data architectures where bronze, silver, and gold transformation stages each build on the output of the previous layer. The minimum TARGET_LAG for Dynamic Tables is one minute, not five minutes. Near-real-time refresh with low-minute latency is achievable, though it increases compute costs proportionally as more frequent refreshes consume more warehouse credits. Materialized Views do not support multi-table joins in Snowflake. Any statement attributing that capability to Materialized Views is incorrect. Materialized Views are best suited for accelerating simple single-table projections or aggregations where their automatic background maintenance provides a query transparency benefit.
More Snowflake Practice Exams
SnowPro Advanced: Administrator (ADA-C02)
ADA-C02 · 600 questions
SnowPro Specialty: Gen AI (GES-C01)
GES-C01 · 600 questions
SnowPro Specialty: Native Apps (NAS-C01)
NAS-C01 · 600 questions
SnowPro® Advanced: Data Scientist (DSA-C03)
DSA-C03 · 600 questions
SnowPro® Specialty: Snowpark (SPS-C01)
SPS-C01 · 599 questions
SnowPro Advanced: Data Engineer (DEA-C02)
DEA-C02 · 597 questions
$17.99
One-time access to this exam
600 questions (6 practice exams' worth)
Unlimited timed exam simulations
Or $15/mo for all 253 exams
Detailed explanations
Free preview stays available