How AI Is Redefining Data Analyst Skills and What You Must Learn

01 AI Lowers Data Analysis Barriers

AI technologies such as DeepSeek reduce the entry threshold for data analysis, allowing anyone to become a data analyst; the tools can handle routine tasks like table processing, model code generation, and data cleaning, even for users without coding experience.

However, AI cannot replace human abilities to understand business needs, judge whether analysis results meet objectives, and translate insights into actionable decisions, which require deep business insight and critical thinking.

02 Essential Data‑Application Skills for Enterprises

The CDA Data Science Institute revised the 2022 "Red Book" to create a forward‑looking, practice‑oriented curriculum that meets the talent requirements of digital‑intelligent enterprises. It proposes a data‑talent capability model covering macro business analysis to micro individual prediction.

The CDA certification system defines three learning levels (Level 1‑3) based on the depth of attribution analysis and strategy optimization, ranging from macro business analysis to detailed modeling, and sets tool‑mastery requirements for each level.

CDA Level 1 – Commercial Strategy Data Analysis covers metric system construction, user tagging, profiling, qualitative and attribution analysis, and business‑strategy reporting.

CDA Level 2 – Quantitative Strategy Analysis teaches advanced user tagging, model attribution, and predictive modeling that combine macro business insights with micro customer behavior.

CDA Level 3 – Agile Data Mining focuses on building and deploying mining models within an MLOps framework, using SQL and Python for implementation.

The core talent model can be summarized in three statements: describe the current state, analyze causes, predict behavior; distinguish macro (metrics) and micro (tags) perspectives; and combine them for comprehensive analysis.

03 Key Skills for Data Analysts

Skill 1: Business Status Visualization – Build indicator systems, use spreadsheets, BI tools, and SQL for visual dashboards that follow the "metric + dimension" principle.

Skill 2: Business Attribution Analysis – Perform qualitative, metric, and model attribution (e.g., DuPont metric decomposition, linear regression) to uncover root causes.

Skill 3: Quantitative Strategy Formulation – Define quantitative user profiles and rule‑based strategies; apply algorithmic models for probability‑based product recommendations.

Retirees (65+): long capital retention cycle

Small‑business owners (30‑40): high monthly cash flow, short turnover

Recent graduates (21): low monthly income, short retention

04 Highlights of the New CDA Textbooks

First Edition Highlights

1. Strengthen uniquely human business and data thinking that machines cannot replace.

2. Incorporate cutting‑edge industry technologies to keep the knowledge base current.

3. Expand breadth (more industry cases) and depth (detailed technical guidance, cross‑database query syntax, model tuning, and validation).

Code‑related knowledge is dynamically updated; AI‑generated code is verified through testing to ensure reliability.

05 CDA Level 2 New Textbook Highlights

The Level 2 book integrates enterprise‑level data project workflows, teaching when and how data analysis should intervene in product development—from requirement discovery to post‑launch evaluation—bridging the gap between technology and business.

Overall, the optimized CDA Level 1 and Level 2 textbooks address the concrete skill gaps of professionals, enhancing their market competitiveness.

For those interested in testing their data‑analysis abilities, the CDA certification mini‑program offers numerous practice questions—scan the QR code to start.