Introduction to Data Analytics · Lesson 2 of 3

Turn a Business Request into an Analytics Question

Convert a vague request into a measurable question, metric definition, and analysis plan.

Lesson

1. Introduction

Stakeholders often start with a solution request such as “make a dashboard” or “find why sales are down.” Before choosing a tool, an analyst needs to understand the decision the work should support. A well-framed question saves time and prevents a technically correct answer to the wrong problem.

2. Understand the concept

Clarify the decision owner, population, time window, unit of analysis, comparison, and desired action. Define a metric as a precise numerator, denominator, filters, and time basis. “Conversion” might mean purchases divided by visitors, or purchases divided by product-page sessions; those are different metrics.

3. Syntax and interpretation

question = 'Did checkout completion change after the new flow?'
population = 'eligible checkout sessions'
metric = 'completed checkouts / eligible sessions'
comparison = 'four weeks before vs. four weeks after'

Writing the question and metric in plain language makes hidden assumptions visible before any query or chart is built.

4. Apply it to an analysis

Convert a vague request into a measurable question, metric definition, and analysis plan. Start with a small, known example and state what each row or value represents. After the transformation, check row counts, types, keys, and units against the original data. If an assumption fails, stop and investigate instead of presenting the output as a valid result.

5. Common mistakes

Using a metric name without agreeing its exact calculation.
Choosing a time window after seeing which period gives the preferred answer.

6. Practice task

A manager asks whether a campaign “worked.” Rewrite that as a question with a population, outcome, and comparison.

7. One possible solution

Example: Among customers eligible for the campaign, did the 14-day purchase rate differ between randomly assigned recipients and a comparable holdout group?

8. Summary and next step

Convert a vague request into a measurable question, metric definition, and analysis plan. Use the next lesson in this module to extend the idea with another technique and a more realistic analysis case.

Visual guide

Working through Turn a Business Request into an Analytics Question

Step 01 Define
State the question, data grain, and expected output.
Step 02 Apply
Use the method and check its assumptions on a small example.
Step 03 Verify
Reconcile the result and communicate limitations.

Code examples

Improve a vague request

TEXT

Vague: Why are customers unhappy?\nFocused: Did the 30-day refund rate change by product category after the policy update?

Expected output

The focused question identifies an outcome, segment, comparison, and event.

Define an operational metric

PYTHON

refunds = 24
eligible_orders = 800
refund_rate = refunds / eligible_orders
print(f'{refund_rate:.1%}')