# Data analyst roadmap 2026

Turn a small dataset into a defensible decision using spreadsheets, SQL, Python and clear reporting.

## Weeks 1–2 · Questions before charts
Identify the decision, audience, unit of analysis and definition of success.

- [ ] Build: Ask which campus events should run again based on attendance and costs.
- [ ] Review: Define attendance rate and distinguish registrations from check-ins.

## Weeks 3–4 · Clean and audit
Understand missing values, duplicates, inconsistent dates and outliers. Preserve the original input.

- [ ] Build: Clean a synthetic event CSV and record every transformation.
- [ ] Review: Report rows before and after cleaning and justify exclusions.

## Weeks 5–6 · SQL
Learn filtering, grouping, joins and window functions. Think about table grain before joining.

- [ ] Build: Calculate attendance by month and event type.
- [ ] Review: Check that a join has not accidentally multiplied records.

## Weeks 7–8 · Python analysis
Use pandas for repeatable transformations. Keep business definitions in one place.

- [ ] Build: Reproduce your spreadsheet results in a script or notebook.
- [ ] Review: Reconcile totals and explain differences rather than hiding them.

## Weeks 9–10 · Visualization
Choose charts around comparisons. Label axes, units and time ranges; avoid misleading scales.

- [ ] Build: Create a bar chart of attendance rate and a monthly trend.
- [ ] Review: A reader can identify the denominator and whether samples are small.

## Weeks 11–12 · Decision memo
Separate observations from causal claims. Include uncertainty and alternative explanations.

- [ ] Build: Recommend two events to repeat with a one-page evidence memo.
- [ ] Review: State what additional data would change your recommendation.

Portfolio outcome: Campus events analysis with a synthetic dataset, cleaning log, SQL queries, dashboard and decision memo.

https://www.egotechworld.com/recourses/data-analyst-roadmap-2026.php