Data Analyst · Step 1 of 8 · 7 min
What data analysts actually do
You can describe the job in one sentence and turn a vague request into a clear question.
Question, data, answer, decision
A data analyst helps people make better decisions using data. The work follows a loop: someone has a question, you find and prepare the data, you work out an answer, and the answer feeds a decision. If no decision changes, the analysis was probably not needed.
Most of the skill is in the first step. “Why are sales down?” is a worry, not a question. “Did weekly orders fall in all regions since March, or only in one?” is something data can answer.
The job titles
The same work goes by many names: data analyst, business analyst, BI (business intelligence) analyst, reporting analyst, product analyst, marketing analyst, operations analyst. Read the job description, not the title. Look for the tools it lists and the decisions it supports.
Data scientists usually build predictive models and use more statistics and code. Data engineers build the pipelines that move and store data. Analysts sit closest to the business people asking the questions.
A normal week
Expect a mix: answering ad-hoc questions from colleagues, refreshing regular reports, fixing a number that looks wrong, writing SQL, building or updating a dashboard, and explaining findings in a meeting or a short written note.
A large share of time goes to finding and cleaning data, and to clarifying what people really want. That is normal, not a sign you are doing it wrong. Clear writing and calm communication matter as much as any tool.
The core toolkit
Almost every analyst job asks for three things: a spreadsheet (Excel or Google Sheets), SQL for pulling data from databases, and a visualisation tool such as Power BI, Tableau or Looker Studio. Python with the pandas library, or R, is a plus for many roles but rarely the first requirement.
This track follows that order. You will finish with a small public project you can show an employer.
Try it
Sharpen a vague question
- 1.Think of a vague question from your work or life, like “Are we spending too much on food?”
- 2.Paste it into a chatbot with the prompt below.
- 3.Pick the rewritten question you could answer with data you actually have.
- 4.Write down which decision the answer would change.
I am learning data analysis. Here is a vague question: [YOUR QUESTION]. Rewrite it as three specific questions that data could answer. For each, say what data I would need, what time period, and what decision the answer could change.