Data Analysis with Clear Questions

So far you've experimented. Now you'll get focused. You'll feed AI precise questions and learn what matters: a good question is half the analysis.

Why Intention Matters

In K01 (Text) you just wrote freely. In K02 (Music) you experimented. That was important — to understand what AI can do. Now comes the next stage: analytical intention.

That means: before you ask AI, you yourself must know WHAT you want to know. Not "analyze," but "where do I spend the most money — and compare that with last year." That's a huge difference.

A clear question leads to actionable answers. A vague question leads to interesting but useless observations. That's not your fault — it's just the reality of data analysis.

Three Real Scenarios

Let me show you three scenarios many people have:

Scenario 1: The Household Budget (personal finance)

You've kept track of your expenses for a year: groceries, transportation, hobbies, clothing. Now you have 50 entries. The weak version of the question would be:

Analyze my spending.

The strong version would be:

Which category do I spend the most on percentagewise? Which category went over budget this month — compare with the last 12 months. Visualize for me: bar chart for category comparisons, line chart for my trends over months.

The difference: the strong question tells AI exactly what you need. You save time AND get results you can use directly.

Scenario 2: Survey Results (research, customer feedback)

You surveyed 200 people: "What is the most important feature in a fitness app?" You categorized their answers (tracking, music, community, gamification, etc.). The weak question:

What do people say about the best fitness app feature?

The strong question:

Which features were mentioned most often? Are there differences between male and female respondents? What are the Top 3 features — give me absolute numbers and percentages. Show me a bar chart with the Top 5 features.

Again: specificity makes the difference between information and noise.

Scenario 3: Business Data (sales, traffic, growth)

You sell online and have data: daily sales, product category, discount yes/no. Weak question:

When do I sell the most?

Strong question:

In which product category is my revenue highest? On which days of the week do I sell the most? Does a 10% discount make a difference in sales volume? Give me a line chart for daily trends, and tell me if there are anomalies (unusual days).

The Pattern: Weak vs. Strong

You see the pattern? Strong questions have these characteristics:

1. Specific: Not "analyze," but "compare X with Y."
2. Limited: Not everything, but Top 3, or a time period, or one question per prompt.
3. Demanding: You ask for concrete outputs — not just description, but numbers and visualizations.
4. Context: You briefly explain WHY you're asking (household budget over 12 months, survey with 200 respondents). This helps AI choose appropriate methods.

Weak vs. Strong: Prompt Pairs

Here are real examples you can use directly:

Scenario 1: Household Budget Analysis

Weak:

I've recorded my spending over 12 months. What stands out?

Strong:

I've recorded my spending (groceries, transportation, hobbies, clothing, other) over 12 months. In which category do I spend the highest percentage? Show me a comparison from January to December for each category. In which month were my expenses highest? Create a line chart for monthly total spending.

Scenario 2: Blog Traffic Analysis

Weak:

My blog traffic is irregular. What's the problem?

Strong:

My blog has 50 articles. I have the data: publication date, topic (Technology, Travel, Recipes), page views, comments. Which topics have the most views on average? Is there a difference in traffic between weekdays and weekends? What are my 3 most successful articles? Show me a bar chart (Top 10 articles by traffic) and a line chart (traffic over time).

Scenario 3: Product Sales Analysis

Weak:

My sales are different. Why?

Strong:

I sell 3 products (A, B, C) online. My data for 6 months: daily sales per product, price, discount. Which product has the highest average daily revenue? In which month were sales best? Does a 10% discount correlate with higher sales volume? Create two charts: 1) bar chart for revenue per product, 2) line chart for daily sales over 6 months.

The Checklist: Before You Ask AI

Use this checklist before you write a prompt:

□ Do I have a concrete question, not an open exploration? □ Have I told AI what's in my data (how many data points, which categories)? □ Have I asked for specific outputs (numbers, Top 3, comparisons)? □ Have I asked for visualizations that make sense? □ Have I asked one question per prompt (not 5 questions at once)?

If you can check all boxes, you're ready. Otherwise, rethink the question.

What This Lesson Isn't

This lesson isn't about you becoming a data expert. It's not about you becoming an Excel genius either. It's about the minimum: understanding how to use AI with intention.

That's the bridge between experiment and practice. And the bridge is surprisingly short — if you know how to ask.

A Thought to Take Away

A good question is a gift you give to AI (and yourself). A bad question is wasted time for both. That's why: take time to formulate the question. Write it down. Check it. Then ask AI.

This sounds like more work. It's less work — because you get answers you can actually use.