Job descriptions for data analysts are full of tool names: SQL, Excel, Python, Power BI. They rarely describe the actual work. If you understand what an analyst does through a normal week, you will know exactly why each skill matters and how to prepare.
The core job in one sentence
A data analyst helps people make better decisions by turning data into clear answers. Everything else (tools, dashboards, models) is a way of doing that.
Where requests come from
Most work starts with a question from someone who is not technical:
- A sales head asks why one region missed its target last month.
- An operations manager wants to know which warehouses dispatch late most often.
- The marketing team wants a list of customers who have not ordered in 90 days.
- Finance wants the monthly numbers in a fixed format by the third working day.
These requests are often vague. A big part of the job is asking follow-up questions: what period, which customers, what counts as "late", and what decision the answer will feed into.
A typical week
Recurring reports
Many analysts own a few regular reports or dashboards: daily sales, weekly operations, monthly management packs. Over time, good analysts automate these so they take minutes instead of hours.
Ad-hoc analysis
The interesting part: one-off questions that need investigation. You pull data with SQL, clean and combine it, look at it from several angles, and write a short summary with a chart or two.
Data cleaning and checking
Real data has duplicates, missing values, wrong codes and changing definitions. Analysts spend a lot of time checking that numbers are right before anyone sees them. A wrong number in front of a senior manager damages trust quickly.
Meetings and explaining
You present findings, answer "why" questions and sometimes push back when a request would give a misleading answer. Communication is as much a part of the job as SQL.
Documentation
Writing down how a metric is calculated, where data comes from and what assumptions you made saves everyone time later.
Tools and where they fit
| Task | Common tools |
|---|---|
| Getting data from databases | SQL |
| Quick checks and small tables | Excel or Google Sheets |
| Cleaning, combining, repeatable analysis | Python with pandas |
| Dashboards for managers | Power BI, Tableau, Looker |
| Sharing results | Slides, documents, email, chat |
Skills that separate good analysts
- Asking the right question before writing any query.
- Sense-checking numbers: does this total look plausible compared with last month?
- Knowing the business: what drives revenue, cost and customer behaviour in that industry.
- Writing clearly: a three-line summary that a busy manager can act on.
What this means for your preparation
Practise the full loop, not just syntax: start from a vague business question, get and clean the data, find the answer and write it up. That is exactly what guided projects and live projects are designed to train. For the skills order, see the data analyst roadmap for freshers.
Practise what you just read
Reading helps; writing the code yourself is what sticks. Start here: