Data Analysis: Steps, Methods, and Thinking Frameworks

Data analysis has become increasingly popular, influencing everything from daily operations to strategic decisions, yet many people are unsure how to actually perform it, what methods to use, and what thinking process to follow.

01 Data Analysis Steps

1. Accurately view data – Both the numeric value and its descriptive context are essential; without a clear scenario the meaning of the number is ambiguous. For example, "Xiao Ming earned 1 million" could mean a single year's income or a total over 20 years.

2. Objectively judge whether there is a problem – Establish standards for problem identification, such as experience‑based benchmarks, expert evaluations (e.g., scoring, AHP), or trend‑based comparisons. A sales‑cycle trend chart illustrates how a downward point may be normal (point A) while a deeper dip (point B) signals an issue.

3. Choose appropriate analysis methods – Common techniques include matrix analysis, funnel analysis, DuPont analysis, hierarchical analysis, and cross‑analysis, each helping to pinpoint the core contradictions of a problem.

02 Data Analysis Thinking

Practitioners often collect excessive dimensions (user profiles, product profiles, scenario profiles) and generate massive Excel files without extracting valuable conclusions.

1. Business‑process driven – Map the business flow to identify which steps produce the result and what factors influence each step. For example, analyzing a product stock‑out requires tracing supply‑chain nodes to determine whether forecasting, ordering policies, or inventory controls caused the shortage.

2. Department‑function oriented – Align analysis to the responsible department to facilitate implementation. In activity analysis, grouping metrics by department helps pinpoint each team's performance and suggest targeted strategies.

3. Strategy‑goal oriented – Clarify the business objectives and strategies before analysis. For a promotion review, understanding the campaign goals and tactics allows analysts to frame the evaluation correctly and produce insights that align with business expectations.

Questions such as “What mindset is required for data analysis? Which tools improve efficiency?” are raised, followed by a call to scan a QR code to reply “Data Analysis”, download the e‑books Data Analysis Path and Big Data Analysis Platform , and continue learning.

References: 1) “Data Analysis Has So Many Tricks!” (https://mp.weixin.qq.com/s/iNxczk2sbP2AzbjH‑B71bQ); 2) NetEase “Data Analysis Pitfall Guide: Thinking Edition” (https://mp.weixin.qq.com/s/UbttRq6q9jriKPt7ayVq_Q).