Capstone: End-to-End Analysis

Bringing It All Together

In the previous lesson, we explored building repeatable analysis. Now let’s build on that foundation. You’ve learned the pieces. Now let’s see the complete picture.

This lesson walks through a full analysis—from business question to recommendation. You’ll see how each skill from this course connects into one workflow.

The Complete Analysis Workflow

QUESTION → EXPLORE → ANALYZE → VISUALIZE → INSIGHT → COMMUNICATE
    ↓          ↓         ↓          ↓          ↓           ↓
 Lesson 2   Lesson 3  Lesson 3   Lesson 4   Lesson 5   Lesson 6
            Lesson 4                                    Lesson 7

Each step builds on the previous one. Skip a step and the analysis suffers.

Capstone Scenario

The situation:

You work at a SaaS company. The CEO asks: “Why did revenue drop last month? What should we do about it?”

Let’s walk through the complete analysis.

Step 1: Frame the Question (Lesson 2)

Don’t just accept the question at face value.

The CEO asked about revenue. But what specifically?

• Revenue dropped—compared to what? Last month? Last year? Target?
• What’s the magnitude? 5% or 50%?
• Is this one-time or a trend?

Clarifying questions:

AI: "Help me develop clarifying questions for this analysis request:
'Why did revenue drop last month and what should we do?'

What do I need to understand before starting the analysis?"

Refined question: “Revenue dropped 12% month-over-month. Is this due to fewer customers, lower spend per customer, or both? What segments are most affected? What’s driving the decline?”

Step 2: Explore the Data (Lesson 3)

Profile before analyzing.

You pull revenue data by customer, product, and time period.

AI: "Profile this revenue data. Tell me:
- Structure and time range
- Any data quality issues
- Initial patterns you notice

[Paste sample data]"

What you discover:

• Data covers 18 months
• 3,200 customers
• Missing data for 12 customers (investigate later)
• Revenue by segment: Enterprise (45%), Mid-market (35%), SMB (20%)

Step 3: Analyze the Pattern (Lesson 3-4)

Break down the revenue drop.

Revenue can decline because of:

• Fewer customers (churn)
• Lower spend per customer (contraction)
• Mix shift (losing high-value segments)

AI: "Analyze this revenue decline by decomposing it into:
1. Customer count change
2. Average revenue per customer change
3. Segment mix change

[Paste data]

Which factor is the primary driver?"

Findings:

• Customer count: down 3% (lost 96 customers)
• Average revenue per customer: down 9%
• Mix: Enterprise flat, SMB down significantly

The story emerges: Revenue dropped mainly because existing customers spent less. SMB segment is hurting most.

Step 4: Visualize the Insight (Lesson 4)

Choose charts that show the pattern clearly.

For this analysis:

• Line chart: Revenue trend over 6 months (shows it’s not random)
• Bar chart: Revenue change by segment (shows SMB problem)
• Waterfall chart: Bridges from last month to this month (shows components)

AI: "I found that revenue dropped 12%, primarily due to:
- 3% from customer losses
- 9% from existing customer spend reduction
- SMB segment most affected

What visualizations would best communicate this to executives?"

Key visual: A simple bar chart showing revenue by segment, with annotations highlighting the SMB decline.

Step 5: Extract the Insight (Lesson 5)

Move from description to diagnosis to prescription.

Description: Revenue dropped 12%.

Diagnosis: Existing SMB customers reduced spending. Investigation shows they’re downgrading plans due to price sensitivity—competitors recently cut prices.

Prediction: Without intervention, revenue will likely drop another 5-8% as more SMB customers downgrade or churn.

Prescription: Consider targeted retention discounts for at-risk SMB customers. Evaluate pricing competitiveness.

AI: "Help me develop this finding into a full insight:

Finding: SMB revenue dropped 15% due to plan downgrades.
Context: Competitors reduced prices 10-15% last quarter.

Structure this as: what happened, why it matters, what's causing it,
what might happen next, what we should do."

Quick check: Before moving on, can you recall the key concept we just covered? Try to explain it in your own words before continuing.

Step 6: Communicate the Findings (Lesson 6)

Structure for your audience—the CEO.

Executive summary (one page):

REVENUE ANALYSIS: APRIL DECLINE

THE SITUATION
Revenue dropped 12% month-over-month ($180K), primarily driven by
SMB customer spend reduction, not customer losses.

ROOT CAUSE
SMB customers are downgrading plans. Exit surveys and competitor
analysis suggest price sensitivity—competitors cut prices 10-15%.

RECOMMENDATION
1. Implement retention discounts for at-risk SMB accounts (est. $50K investment)
2. Review SMB pricing strategy vs. competitors
3. Increase customer success outreach to at-risk accounts

EXPECTED IMPACT
Retention offers could recover 40-50% of lost revenue based on
similar interventions last year.

RISKS
- Price cuts may not be the only factor; product satisfaction may also contribute
- Deeper competitor analysis recommended before pricing changes

Step 7: Build for Reuse (Lesson 7)

This analysis will repeat.

The CEO will ask about revenue again next month. Build a template:

Monthly Revenue Analysis Template:

1. Pull revenue data by segment and customer
2. Compare to previous month and previous year
3. Decompose into: customer count, spend per customer, mix
4. Identify largest drivers
5. Investigate causes for major changes
6. Generate executive summary with recommendations

AI: "I just completed a revenue decline analysis. Help me turn this into
a reusable monthly template I can run quickly each month.

My process was:
[Describe your steps]

Create a prompt template I can reuse with fresh data."

The Analysis Mindset

Throughout this course, you’ve developed more than skills. You’ve developed a mindset.

Ask before analyzing: What question are we really answering? What decision will this inform?

Explore before concluding: Understand your data before drawing conclusions. Profile first.

Simplify always: One chart, one message. Clear visualizations beat complex ones.

Push to insight: Descriptions aren’t enough. Diagnose, predict, prescribe.

Match to audience: Executives get one page. Peers get context. Technical audiences get methodology.

Build for reuse: If you’ll do it again, template it.

Your Analysis Toolkit

You now have:

Exercise: Your Own Capstone

Think of a question your organization faces. Work through the full process:

1. Frame it: What’s the real question? What decision does it inform?
2. Get data: What data would you need? Where would it come from?
3. Explore: What would you look for first?
4. Analyze: How would you break down the problem?
5. Visualize: What charts would tell the story?
6. Extract insight: What’s the “so what” and “now what”?
7. Communicate: Who’s the audience? What do they need?

Even without data, walking through this framework builds analytical muscle.

What’s Next?

You’ve completed the course. Here’s how to keep building:

Practice immediately: Apply these skills to a real analysis this week. Learning solidifies through use.

Build your library: Start creating templates for analyses you repeat. Each one saves future time.

Iterate with AI: The more you practice AI-assisted analysis, the better your prompts become. Experiment.

Go deeper: For advanced topics—statistical testing, predictive modeling, visualization design—seek out specialized resources.

Course Summary

Across eight lessons, you’ve learned to:

1. Think like an analyst: AI augments your thinking, doesn’t replace it
2. Frame questions properly: The right question is half the answer
3. Explore data rapidly: Profile before analyzing, spot issues early
4. Create clear visualizations: Match chart type to message
5. Extract real insights: Move beyond description to prescription
6. Communicate for impact: Structure reports for your audience
7. Build repeatable workflows: Template what you’ll do again
8. Execute end-to-end: Combine all skills into complete analyses

Data analysis isn’t about having the fanciest tools. It’s about asking good questions, finding meaningful patterns, and communicating them clearly.

You have the framework. Now go analyze something.