What Is Data Analytics?
Kettle & Crumb is a small café. Its owner, Nisha, has a till that records every order, a spreadsheet of her menu, and a list of loyalty members. She also has questions: which days are busiest, why last Saturday was chaos, how much cold coffee to order. Data analytics is the craft of getting from those questions to answers she can act on. We will follow her café through the whole course.
The rule
Start with a question; end with a decision.
Data on its own answers nothing. An analyst's job is a loop with five stops:
- Ask: turn a vague worry ("sales feel slow") into a precise question ("How many orders did we take each day in June, compared with May?").
- Collect: find the data that can answer it, from a database, a file or an outside source.
- Clean: fix what is missing, doubled, mistyped or in the wrong format.
- Explore and analyse: summarise, compare, chart, and look for patterns.
- Communicate: explain the answer in plain words, with the right chart, so someone can decide.
Here is a taste of Nisha's orders sheet, the table you will see again and again:
| order_id | order_date | customer | item_code | qty | amount |
|---|---|---|---|---|---|
| 1001 | 2025-06-02 | Asha Rao | C01 | 2 | 240 |
| 1002 | 2025-06-02 | Vikram Shah | C02 | 3 | 120 |
| 1003 | 2025-06-03 | Meera Iyer | B01 | 1 | 90 |
An analyst is not the same as a data scientist or a data engineer. Engineers build the pipes that move and store data. Scientists often build predictive models. Analysts sit closest to the decision: they know the business question and make the data answer it honestly.
The shortcut
Before touching a spreadsheet, ask "What decision will this answer change?" If you cannot name one, sharpen the question first. A good question names a measure, a group and a time window.
Hear the hunch
Nisha says: "I think Mondays are a waste of time."
The common trap is jumping straight to charts. A beautiful chart of the wrong question wastes everyone's time.
Drills
Answer fast — trust the rule.
Which is the best-formed analysis question?
Three order amounts are blank and one order appears twice. Which stage of the loop deals with this?
Nisha asks "Should we stock more cold coffee?" What should the analyst clarify first?
Which task belongs most naturally to a data analyst rather than a data engineer?
Remember
- Start with a question; end with a decision.
- The loop: ask, collect, clean, explore and analyse, communicate.
- Trap: charts before a clear question. Name the measure, the group and the time window first.