What Is Data Analysis? Definitions, Skills, History, and Practical Steps

1 Data Analysis Definition

Data analysis is the targeted collection, processing, and interpretation of data using statistical and mining techniques to extract useful information and draw conclusions.

Data Types

Data can be qualitative (attributes, names) or quantitative (numeric values). Quantitative data further divides into ordinal, interval, and ratio types.

2 Historical Milestones

Computer Era : Early censuses took years; computers reduced processing time dramatically.

Relational Databases : Invented by Edgar Codd in the 1970s, enabling SQL queries for data retrieval.

Data Warehouses : Emerged in the late 1980s to support decision‑making systems.

Business Intelligence (BI) : Grew in the 80s‑90s with ETL and OLAP technologies.

Data Mining : Developed in the 1990s as data volumes outgrew simple queries.

Big Data : 2000‑2010 saw the rise of Google’s File System, MapReduce, BigTable, and Amazon’s Dynamo, enabling massive data processing.

Data Scientist : Term popularized in 2012 as a high‑impact role.

Augmented Analytics : Introduced in 2017, combining machine learning and natural language generation to automate insights.

3 Data Analysis vs Data Science vs Business Intelligence

Business Intelligence : Uses tools to turn raw data into actionable information for business decisions.

Data Science : A multidisciplinary field that extracts actionable insights from large, structured and unstructured data using mathematics, statistics, machine learning, visualization, and high‑performance computing.

Data analysis is a broad area that includes data science; not all analysis is commercial, and the two intersect only partially.

4 Roles: Data Analyst vs Data Scientist

Data analysts focus on exploring large datasets, visualizing trends, and supporting strategic decisions, while data scientists build predictive models, algorithms, and custom analytical pipelines.

5 Why Analyze

Analysis clarifies business situations, identifies opportunities, mitigates risks, diagnoses problems, and drives actionable decisions.

6 How to Analyze – Six Steps

Define analysis purpose and understand the business context.

Choose analysis thinking models and frameworks (e.g., structured, time, logical, importance models).

Apply classic analysis models such as SWOT, STP, RATER, etc.

Design indicator systems: define metrics, ensure they are comparative, simple, actionable, and behavior‑changing.

Build indicator hierarchies from business goals to key process metrics.

Follow indicator design principles (OMTM, MECE, etc.).

7 Key Thinking Models

Structured Models : Pyramid principle, 5W2H.

Time Models : User behavior stages.

Logical Deduction : Three‑premise reasoning, 4W model.

Importance Models : MECE, priority matrices.

8 Indicator Design Principles

Good metrics are comparative, simple, reflective of reality, and capable of changing behavior. Use the One Metric That Matters (OMTM) principle to focus on the most critical metric.