---
title: "Asking better analytical questions | Optimize All Academy"
description: "Good analysis starts with good questions AI can answer almost any question you ask of data, which makes the quality of the question the limiting factor…"
url: https://optimizeall.com/learn/ai-for-data-analysis-and-decision-making/asking-analytical-questions
updated: 2026-10-05
---

AI for Data Analysis & Decision Making · Exploring data and finding insights · lesson 5 of 16 · 10 min

# Asking better analytical questions

## Good analysis starts with good questions

AI can answer almost any question you ask of data, which makes the quality of the question the limiting factor. Vague questions ("What's interesting in this data?") produce long lists of observations. Decision-focused questions produce useful answers.

## From vague to decision-focused

| Vague | Decision-focused |
|---|---|
| How are our campaigns doing? | Which of our three paid social campaigns should get the extra budget next month, based on cost per qualified lead? |
| What do customers think? | What are the top three reasons for cancellation among customers who left in the last 90 days, and how common is each? |
| Is the new feature working? | Did users who adopted the new checkout complete purchases at a higher rate than similar users who didn't, and is the difference large enough to matter? |

## A question-framing template

```text
Decision: [what will we decide?]
Options: [A, B, C]
Metric that decides it: [e.g. cost per qualified lead]
Threshold: [what difference would change our mind?]
Time frame and population: [who, when]
Known confounders: [seasonality, price change, channel mix]
```

Share this frame with the AI before asking for analysis. It will choose better methods and flag relevant limitations.

## Types of analytical questions

1. **Descriptive:** what happened? (sales by month)
2. **Diagnostic:** why did it happen? (drivers of the drop)
3. **Predictive:** what is likely to happen? (forecast; module 5)
4. **Prescriptive:** what should we do? (recommendation)

Descriptive questions are safest to answer from observational data. Diagnostic and prescriptive questions need extra care, because correlation in historical data is not proof of cause. AI assistants often slide from "these moved together" to "this caused that". Watch for it.

## Metrics: define before you measure

Many disputes about analysis are really disputes about definitions:

- What counts as an "active user"? Logged in? Performed a key action?
- Is "revenue" gross or net of refunds and discounts? Before or after VAT?
- Is "conversion" per session, per visitor or per lead?

Write definitions down and give them to the AI. Ask it to restate your definitions at the top of its analysis, so misunderstandings surface early.

## Leading and lagging indicators

Lagging indicators (revenue, churn) confirm results after the fact. Leading indicators (trial activations, demo bookings, onboarding completion) give earlier signals. When exploring, ask the AI to test whether a candidate leading indicator actually relates to the lagging outcome in your historical data, while remembering that such relationships can change.

## Worked example: a B2B SaaS pricing question

Vague: "Should we raise prices?"

Framed:

```text
Decision: raise the Team plan price for new customers from next quarter?
Options: keep, +10%, +20%
Metric: expected new MRR from Team plan over 6 months
Threshold: proceed if expected MRR rises and win-rate drop is modest
Population: new Team plan deals in UK, UAE, KSA, last 12 months
Confounders: sales team changes in Q2, a competitor price cut in March
```

With this frame, the AI proposes looking at win rates by deal size, discounting behavior and the limits of historical data (no past price tests), and suggests a controlled price test in one region as better evidence than historical analysis alone.

## Questions to ask the AI about its own analysis

- "What assumptions did you make?"
- "What would change your conclusion?"
- "Which numbers are computed and which are estimates?"
- "What is the simplest alternative explanation?"

## Hands-on: a metrics glossary file the AI reads first

Keep definitions in a small file and paste it (or attach it) at the start of every analysis:

```yaml
# metrics_glossary.yml  (owner: analytics; last reviewed 2026-09)
active_customer: "placed >= 1 completed order in the last 90 days"
revenue: "completed orders, net of refunds and discounts, excluding VAT; converted to GBP at monthly average FX"
conversion_rate: "orders / sessions (not per visitor)"
qualified_lead: "lead accepted by sales within 7 days (CRM stage >= SQL)"
churned_subscriber: "no active subscription 30 days after last renewal date"
```

Then instruct: "Restate the definitions you are using at the top of your answer; if a question requires a metric not in the glossary, ask me before defining it."

## Hands-on: turning a vague request into a frame, with AI

```text
My manager asked: "How are our campaigns doing?"
Rewrite this as a decision-focused analysis frame: decision, options, deciding metric
(with definition from the glossary), threshold that would change our mind, population and
time frame, known confounders. Then list the 3 questions I should ask my manager to confirm the frame.
```

## Second worked example: a Lahore ed-tech's "engagement" debate

Two teams disagreed about whether engagement was rising. Product counted "active users" as anyone who logged in; marketing counted anyone who watched a lesson for five minutes. The analyst wrote both into the glossary with separate names (`logged_in_users`, `learning_users`), reran the analysis, and showed that logins were rising while learning activity was flat. The debate ended in ten minutes, and the next question became how to turn logins into learning.

## Going further

Maintain a shared metrics glossary for your team and include it (or the relevant part) in every analysis prompt. Consistent definitions across analyzes are one of the highest-leverage improvements in any data-driven organization, AI-assisted or not.

## Video lecture: Asking better analytical questions

Lecture coming soon · 16 chapters · about 8 minutes. Read the full transcript below.

1. Asking analytical questions
2. Why it matters
3. The framing template
4. Four question types
5. Define before measuring
6. Leading and lagging
7. Thresholds first
8. Example 1: 'How are campaigns doing?'
9. Example 2: Lahore ed-tech
10. Watch me do it, part 1
11. Watch me do it, part 2
12. Questions for the AI
13. Evidence fit
14. Common mistakes
15. Recap
16. Try this now

## Lecture transcript

### Asking analytical questions

What's interesting in this data? It's the most common question people ask AI about data, and it's almost always the wrong one. You'll get a long list of observations and no decision. Now try this instead: which of our three paid social campaigns should get the extra budget next month, based on cost per qualified lead? Same data, completely different answer. In this lecture you'll learn to turn vague questions into decision-focused ones, to define metrics before measuring, and to use a glossary that keeps AI honest.

### Why it matters

Why does this matter? Because AI can answer almost any question you ask of data, which makes the quality of the question the limiting factor. Vague questions produce long lists. Decision-focused questions produce useful answers, better methods, and relevant limitations, because the model knows what's at stake. And many arguments about analysis are really arguments about definitions. Fix the question and the definitions first, and half the work is done.

### The framing template

Here's the template for framing. Decision: what will we decide? Options: A, B, C. The metric that decides it, like cost per qualified lead. The threshold: what difference would change our mind? Time frame and population: who and when. And known confounders: seasonality, a price change, a shift in channel mix. Think of it like a flight plan. A pilot doesn't just say fly somewhere nice. They file the destination, route, fuel and alternates before take-off. Share the frame with the AI before asking for analysis.

### Four question types

Know which type of question you're asking. Descriptive: what happened, like sales by month. Diagnostic: why did it happen, like drivers of a drop. Predictive: what's likely to happen, which we'll cover with forecasting. And prescriptive: what should we do. Descriptive questions are safest to answer from observational data. Diagnostic and prescriptive questions need extra care, because correlation in historical data isn't proof of cause. AI assistants often slide from these moved together to this caused that. Watch for it.

### Define before measuring

Define metrics before you measure. What counts as an active user: logged in, or performed a key action? Is revenue gross or net of refunds and discounts, before or after VAT? Is conversion per session, per visitor or per lead? Write definitions down, give them to the AI, and ask it to restate them at the top of its analysis. Misunderstandings then surface in the first paragraph instead of in the board meeting.

### Leading and lagging

Leading and lagging indicators help too. Lagging indicators, like revenue and churn, confirm results after the fact. Leading indicators, like trial activations, demo bookings and onboarding completion, give earlier signals. When exploring, ask the AI to test whether a candidate leading indicator actually relates to the lagging outcome in your history, while remembering that relationships can change. A leading indicator that doesn't predict anything is just another vanity metric.

### Thresholds first

And set the threshold before you look. It's tempting to analyze first and decide afterward what difference counts as meaningful. But then any difference you find will feel meaningful, because you've already seen it. So ask: what's the smallest difference that would change our decision? Maybe it's a twenty percent lower cost per qualified lead, or two percentage points of conversion. Write it down. It protects you from over-reacting to noise, and it helps the AI choose the right method and sample size.

### Example 1: 'How are campaigns doing?'

First example, a simple one. A manager asks how are our campaigns doing? The analyst reframes it. Decision: which of three paid social campaigns gets the extra budget next month. Metric: cost per qualified lead, where qualified means accepted by sales within seven days. Threshold: move budget only if one campaign is at least twenty percent cheaper per qualified lead, illustratively. Population: leads from the last sixty days. Confounder: one campaign ran during a holiday week. Now the analysis almost writes itself.

### Example 2: Lahore ed-tech

Second example, a business case. At a Lahore education technology company, two teams disagreed about whether engagement was rising. Product counted active users as anyone who logged in. Marketing counted anyone who watched a lesson for five minutes. The analyst wrote both into the glossary with separate names, logged in users and learning users, and reran the analysis. Logins were rising. Learning was flat. The debate ended in ten minutes, and the next question became how to turn logins into learning.

### Watch me do it, part 1

Watch me set up a glossary file. It's a short YAML file with an owner and a review date. Active customer: placed at least one completed order in the last ninety days. Revenue: completed orders, net of refunds and discounts, excluding VAT, converted to pounds at the monthly average rate. Conversion rate: orders divided by sessions, not per visitor. Qualified lead: accepted by sales within seven days. I attach it to every analysis prompt and add one instruction: restate the definitions you're using, and ask me before defining any metric that isn't in the glossary.

### Watch me do it, part 2

Then I let AI help with framing itself. I paste the vague request, how are our campaigns doing, and ask the assistant to rewrite it as a decision-focused frame with decision, options, deciding metric from the glossary, threshold, population, time frame and confounders, plus three questions I should ask my manager to confirm the frame. It suggests asking which budget decision is coming up, whether lead quality or volume matters more, and when the decision is due. I send those three questions before I touch the data.

### Questions for the AI

And ask the AI about its own analysis. What assumptions did you make? What would change your conclusion? Which numbers are computed and which are estimates? What's the simplest alternative explanation? Those four questions take thirty seconds and routinely surface assumptions that would otherwise ride silently into your report.

### Evidence fit

Finally, match the question type to the evidence you have. If your question is prescriptive, like should we raise prices, historical data alone is weak evidence, because you've probably never tested a price change. The frame should say so, and propose a better source of evidence: a controlled test in one region, a survey, or a staged rollout. Good analysts don't just answer the question they're given. They tell you what evidence would actually answer it.

### Common mistakes

Common mistakes. Starting with what's interesting. No threshold, so any difference seems meaningful. Undefined metrics that mean different things to different teams. Treating diagnostic correlations as prescriptive answers. And skipping the step of confirming the frame with the decision-maker. Each of these is easy to prevent with the checks you've just seen, and each one, left unchecked, eventually reaches a decision-maker.

### Recap

Recap. Decision-focused questions produce useful analysis. Frame the decision, options, deciding metric, threshold, population and confounders first. Know which type of question you're asking and watch the slide into causation. Define metrics in a shared glossary the AI reads first. And confirm the frame with the decision-maker before analyzing.

### Try this now

Try this now. Rewrite one vague question your team asks into the framing template, including metric definitions and known confounders. Start a glossary file with at least four definitions, and use it in your next AI analysis prompt. See whether the answer changes.

## Key takeaways

- Decision-focused questions produce useful analysis; vague questions produce lists.
- Frame the decision, options, deciding metric, threshold, population and confounders before analyzing.
- Distinguish descriptive, diagnostic, predictive and prescriptive questions; beware sliding from correlation to cause.
- Define metrics explicitly and have the AI restate them; maintain a shared glossary.

## Try it

Rewrite one vague question your team asks into the framing template, including metric definitions and known confounders.

- [Previous: Exploratory analysis with AI](https://optimizeall.com/learn/ai-for-data-analysis-and-decision-making/exploratory-analysis)
- [Next: A reusable prompt library for trustworthy AI analysis](https://optimizeall.com/learn/ai-for-data-analysis-and-decision-making/analysis-prompt-library)
- [All lessons of AI for Data Analysis & Decision Making](https://optimizeall.com/learn/ai-for-data-analysis-and-decision-making)