Data & Tools
Every company has data and few people who can read it. Learn the handful of ideas and free tools that cover most beginner analyst work.
1Tidy tables
Analysis starts with a tidy table: one row per thing, one column per fact, one value per cell. Most data problems are table problems.
- One header row with short, clear names (order_date, not 'Date of the order!!')
- One data type per column: all dates, all numbers or all text
- No merged cells, no blank rows, no totals mixed into the data
- Keep the raw file untouched and clean a copy
2Formulas that matter
Ten formulas cover most of what a junior analyst does in a sheet.
=SUM(B2:B100) total
=AVERAGE(B2:B100) mean
=COUNTA(A2:A100) how many filled cells
=IF(B2>100, "big", "small") a decision in a cell
=SUMIF(C:C, "Berlin", B:B) total for one city
=COUNTIF(C:C, "Berlin") how many in one city
=XLOOKUP(E2, A:A, B:B) find a value in another table
=TRIM(A2) remove stray spaces
=ROUND(B2/C2, 2) a rate to two decimals3A SQL cheat sheet
SQL asks a database a question. The clauses always come in this order, and you can add only the ones you need.
SELECT city, COUNT(*) AS orders, SUM(total) AS revenue
FROM orders
WHERE order_date >= '2025-01-01'
GROUP BY city
HAVING SUM(total) > 100
ORDER BY revenue DESC
LIMIT 5;- WHERE filters rows before grouping; HAVING filters groups after
- JOIN ... ON combines tables on a shared id
- Use aliases (AS) so results have readable names
- Test a query on a few rows with LIMIT before running it on everything
4Choosing the right chart
| Your question | Chart |
|---|---|
| How does it change over time? | Line chart |
| Which category is biggest? | Bar chart |
| What share does each part have (few parts)? | Pie or stacked bar |
| Are two numbers related? | Scatter plot |
| What are the exact values? | A table |
- Start bar charts at zero, or the differences look bigger than they are
- Title every chart with the finding, not just the topic
- Use one highlight colour and grey for the rest
5The tools analysts use
You do not need all of these. Start with Sheets, one SQL tool and Looker Studio. Free options are enough to build a full portfolio.
Google Sheets
FreeClean data, formulas, pivot tables and quick charts.
Looker Studio
FreeFree dashboards that read from Sheets, SQL and more.
DB Browser for SQLite
FreeRun SQL on a database file on your own computer.
SQLite Online
FreePractise SQL in the browser with nothing to install.
Python and pandas
FreeSeason 2 of the Data Trip: automate cleaning and analysis on bigger data.
Kaggle / Our World in Data
FreeFree real datasets to practise on.
Excel / Power BI
PaidCommon in offices. Same ideas as Sheets and Looker Studio.
Tableau
Free planPopular dashboard tool. Tableau Public is free for portfolios.
6Practice projects
- Clean a messy public spreadsheet and write down every change you made
- Answer five business questions about a small shop database with SQL
- Build a one-page Looker Studio dashboard from a Kaggle dataset
- Write a one-page case study: question, method, finding, recommendation
- Offer a free mini report to a local business and ask for feedback
Keep learning data
- Our World in Data and Kaggle Datasets for free real data
- Google's Looker Studio help center for dashboard tutorials
- SQLBolt and Mode's SQL tutorial for more SQL practice
- The Data Trip lessons, one stop a day
Knowledge Check
Test what you learned. Each correct answer earns XP. up to 55 XP
1. A sales sheet has the same order listed twice. What should you do first?
2. Which SQL clause keeps only the groups whose total is above 100?
3. You want to show how one number changes across 12 months. Which chart fits best?
4. Ice cream sales and sunburn both rise in summer. What can you say?
