.. _dpa-why-do-we-analyze-data:

========================================================================
Why Do We Analyze Data?
========================================================================

**Stage 1 · 📋 Foundations**  ·  Lesson 01 of 56  ·  *beginner*

:doc:`Next · The Process of Data Analysis ▶ <02-the-process-of-data-analysis>`   ·   :doc:`↑ Section <index>`


.. important::

   **✨ AI-generated content.** This page was written with the assistance of an
   AI language model and is provided as a learning aid. Despite careful
   review, it may still contain mistakes, omissions, or out-of-date
   information. Whether you are new to the topic, a team lead, or a senior
   practitioner, treat it as a starting point rather than an authoritative
   reference: read it critically and independently verify anything you act on
   (code, commands, figures, and factual claims) against official
   documentation and primary sources before relying on it.

From data to decisions
------------------------

Data on its own is inert — rows and numbers. **Analysis** is what turns it into something useful:
patterns you can see, questions you can answer, and, ultimately, **decisions** you can act on. Every
technique in this course exists to move data one step further along that chain, from raw records
toward a choice someone has to make.

Four kinds of question
------------------------

It helps to name **what** you are asking. Analysts group questions into four types, each harder and
more valuable than the last:

* **Descriptive** — *what happened?* (how have sales trended this year);
* **Diagnostic** — *why did it happen?* (why did churn rise last quarter);
* **Predictive** — *what is likely to happen?* (which customers will leave next);
* **Prescriptive** — *what should we do?* (which customer to call to keep them).

Choosing the right one
------------------------

The right type is set by the **business problem**, not by the fanciest tool available. A dashboard
answering "what happened" may be all a question needs; a churn-prevention campaign needs a
**predictive** model and, ideally, a **prescriptive** recommendation on top. Much of this course is
about matching the question to a method — and then, in the final stage, **checking the answer is
trustworthy** before anyone relies on it.

.. hint::

   **Related lessons:** :doc:`The Process of Data Analysis <02-the-process-of-data-analysis>`  ·  :doc:`CRISP-DM for Data Science <03-crisp-dm-for-data-science>`  ·  :doc:`Assessing the Quality of Prediction Models <50-assessing-the-quality-of-prediction-models>`  ·  :doc:`The First Step in Knowing Your Data <05-the-first-step-in-knowing-your-data>`

.. seealso::

   **Source article** Adapted (context, re-expressed) in our own words from: `https://insightful-data-lab.com/2026/01/14/why-do-we-analyze-data/ <https://insightful-data-lab.com/2026/01/14/why-do-we-analyze-data/>`__ (insightful-data-lab.com).

.. tags:: purpose: reference, topic: data analysis, topic: data preparation, level: beginner