Data & Decision-Making

Having data is not enough. You need to know which question to ask and what evidence the answer supports.

From Data to Decision

Databricks in practice for investigating real business problems

Learn to turn scattered data into evidence that supports more responsible business decisions and actions.

Notify me when it is available

This book is currently available in Brazilian Portuguese.

EditionIn preparation2026Book languagePortugueseBrazilFormatsComing soonComing soon

About the book

A complete view of the book.

Having data is not enough to make sound decisions. Orders, inventory, costs, failures, sensors, and spreadsheets only become useful when they answer a clear question, pass validation, and gain context. From Data to Decision presents this path with Databricks without turning the tool into an end in itself.

Through the Atlas Case, readers investigate production, quality, inventory, suppliers, sales, and margin. Queries, exercises, SQL patterns, data contracts, and an implementation roadmap help distinguish fact, evidence, hypothesis, and gap before proposing action. The goal is not to find a magic cause, but to build responsible, contestable decisions followed by new measurements.

The challenge

Does this situation sound familiar?

Companies accumulate orders, inventory, costs, failures, sensors, and spreadsheets, but isolated numbers do not explain what is happening. Without clear questions, validation, and context, an analysis can look convincing while still leading to a fragile decision.

The book’s approach

A path from intention to practice.

Learn a practical path—problem, question, data, investigation, information, evidence, decision, action, and a new measurement—using Databricks and the Atlas Case to connect analysis to useful, responsible, and contestable choices.

“Deciding means choosing an action despite uncertainty—and accepting that the next set of data will assess the quality of that choice.”

By the end of the book, you will be able to:

✓Formulate business questions before opening a dashboard or writing a query

✓Explore data in Databricks without losing sight of context, granularity, and limits

✓Validate keys, cardinality, and reconciliations before trusting a number

✓Investigate production, quality, inventory, suppliers, sales, and margin as connected subjects

✓Distinguish fact, evidence, hypothesis, gap, and decision

✓Turn an analysis into an action followed by a new measurement

Questions that guide the reading

Answers begin with the right questions.

Questions explored through context, criteria, and examples.

How can you tell whether a number actually answers the question that matters?

Why can an apparently correct join or metric lead to a wrong conclusion?

How can production, quality, inventory, and margin problems be investigated without searching for one single cause?

What needs to accompany a decision so it can be verified later?

Inside the book

A structured journey that helps you move forward.

Concepts and applications presented in a coherent sequence.

From data to information

Understand why owning records does not yet mean understanding a business situation.

Databricks without complication

Use Databricks as an environment for investigation without making the tool an end in itself.

Trustworthy data

Check quality, keys, joins, counts, and reconciliations before advancing a conclusion.

Investigating the company

Investigate production losses, scrap, failures, inventory, suppliers, and margin through progressive questions.

Information and decision

Connect analyses to responsibilities, alternatives, limits, and later follow-up.

The Atlas Case

Work through a complete case with teaching data, queries, exercises, and response criteria.

Content in preparation

Planned book structure

Explore the 3 parts and 14 planned chapters. Titles may be refined during editorial review.

View chapters14 chapters

Part I — From Data to Information

You have data. But do you have information?
Databricks without complication
The first Atlas data
When data cannot be trusted

Part II — Investigating the Company

Why do we produce less than planned?
Where does scrap come from?
Were there signs before the failure?
How can a product be missing with so much inventory?
What are suppliers telling us?
Does selling more mean earning more?

Part III — From Information to Decision

When production, inventory, and sales talk to each other
From analysis to follow-up
AI: when data starts working for you
The Atlas Case: from question to decision

What makes this book different

Depth to understand. Clarity to keep moving forward.

Databricks presented as an environment for investigation rather than a collection of screens or features

An integrated Atlas Case with fictional operational data and problems that cross business areas

Validation, reconciliation, and evidence limits addressed before any conclusion

Exercises, SQL patterns, data contracts, and an implementation roadmap for putting the method into practice

Who this book is for

Knowledge that becomes action.

Professionals, analysts, managers, and students who want to investigate real business problems with data instead of treating dashboards as ready-made answers will find a direct approach connected to concrete problems.

✓ Professionals who need to turn operational data into business decisions✓ Analysts and managers who want to investigate before explaining a problem✓ People who want to learn Databricks through real situations rather than tool features alone✓ Students and teams who need to connect analysis, evidence, action, and follow-up

Before you begin

What you need to benefit from the book

✓No previous Databricks experience is required

✓Familiarity with spreadsheets, business data, or basic queries helps, but the concepts begin with the question

An informed choice

This may not be the right book if…

—Readers looking for a formula that produces correct decisions without investigation and human judgment

—Readers expecting a tool or dashboard to reveal a root cause on its own

Why this author, on this subject

Experience that gives the reading context.

A teacher, developer, and corporate systems professional, Régys brings data analysis, operations, and decision-making together to show how to investigate real problems without confusing numbers with explanations.

Upcoming release

Receive resources and updates about From Data to Decision.

Upcoming release