1Course 1FoundationsBeginner
Data Analysis Foundations
From a messy spreadsheet to a clear answer: cleaning, summarising, and presenting data so a decision can be made.
- Chapters
- 10
- Time
- 8 hours
- Format
- Self-paced
- Tools
- Google Sheets or Excel, pivot tables, lookups, charts
What you will be able to do
- Clean a raw export: duplicates, blanks, inconsistent labels, dates stored as text.
- Summarise any table with pivot tables and the eight functions that cover most business questions.
- Choose the right chart for a comparison, a trend, a distribution, or a share.
- Write a one-page summary that states the answer first and the evidence second.
- Spot the four most common mistakes that make an analysis wrong before it reaches a meeting.
Chapters
10 chapters. Chapter 1 is free below.
- 1
What a good analysis looks likeFree to read
The question, the data, the answer, the caveats. A template you will reuse in every chapter.
- 2
Getting data in
Exports, CSVs, copy-paste from systems, and the problems each one brings.
- 3
Cleaning
Duplicates, blanks, inconsistent categories, dates as text, numbers as text. Fixing each without breaking the rest.
- 4
Lookups and joins in a spreadsheet
VLOOKUP, XLOOKUP, INDEX/MATCH, and when a spreadsheet join stops being enough.
- 5
Pivot tables
Counting, summing, averaging, and grouping by time. The fastest way to a first answer.
- 6
Averages lie
Mean versus median, outliers, and why a distribution beats a single number.
- 7
Charts that answer a question
Bar, line, histogram, scatter. Which one to use, and the formatting that keeps it honest.
- 8
Rates, ratios, and growth
Percentages of what? Month-on-month versus year-on-year, and base effects.
- 9
The one-page summary
Answer, evidence, caveats, next step. Writing for someone who has two minutes.
- 10
Capstone: a sales and refunds dataset
Clean it, summarise it, chart it, write it up. Reviewed against a model answer.
Free chapter 1 of 10
What a good analysis looks like
Most analysis at work fails before any number is computed. Someone is handed an export, opens it, and starts making charts. Two hours later there are eleven charts and no answer. The fix is a habit, not a tool: write the question down first, in one sentence, with the decision it feeds. 'Which product line lost the most margin in Q2, so we know where to renegotiate supplier terms' is a question. 'Look at Q2 margins' is not.
Once the question is written, the data usually turns out to be smaller than the export. You need three columns, not forty. You need one quarter, not three years. Cutting the data down early is not laziness; it is the step that makes the rest of the work checkable by someone else.
The answer comes next, and it should be a sentence a manager can repeat. 'Line B lost ₹14 lakh of margin, two thirds of it from one supplier's price increase in May.' Then the evidence: the table or chart that shows it. Then the caveats: what the data does not include, what you assumed, what would change the conclusion.
Every chapter in this course ends with this same four-part template. By the capstone you will produce it without thinking, and that is the point. The tools change every few years. The template does not.
Who it is for
Anyone who receives data at work and is expected to say what it means: operations, finance, marketing, founders, and analysts in their first year.
Before you start
None. You have used a spreadsheet to add up a column.
Taught by
Paul Montero
Founder and instructor
Questions about this course
How long do I have access?
Is the dataset included?
Can I get a refund?
incl. GST, 7-day refund