Business Analytics: Framework & KPIs
402 words
2 min read
Visual companion
Python
Type and operator map
Python Week 1: the first filter for runtime behavior
View
Revision summary
What this note is really saying
Short form
# Business Analytics: Framework & KPIs ## 🎯 Learning Objectives - Define the business analytics framework and its components - Identify and design KPIs aligned with business strategy - Apply A/B testing for data-driven decisions - Analyze customer behavior using RFM and CLV ## 📖 Core Content ### 1.1 What is Busine...
Business Analytics: Framework & KPIs
🎯 Learning Objectives
- Define the business analytics framework and its components
- Identify and design KPIs aligned with business strategy
- Apply A/B testing for data-driven decisions
- Analyze customer behavior using RFM and CLV
📖 Core Content
1.1 What is Business Analytics?
Business Analytics (BA) is the practice of using data, statistical analysis, and quantitative methods to drive business decisions. It sits at the intersection of data science, business strategy, and decision-making.
(Diagram)
1.2 Types of Analytics
| Type | Question | Example |
|---|---|---|
| Descriptive | What happened? | "Sales dropped 15% in Q3" |
| Diagnostic | Why did it happen? | "Sales dropped due to competitor price cut" |
| Predictive | What will happen? | "Sales will recover by 5% next quarter" |
| Prescriptive | What should we do? | "Reduce price by 10% and increase ad spend by 20%" |
1.3 KPI Design Framework
SMART KPIs:
- Specific: Clearly defined
- Measurable: Quantifiable
- Achievable: Realistic target
- Relevant: Aligned with strategy
- Time-bound: Has a deadline Example KPIs by Domain:
| Domain | KPI | Formula | Benchmark |
|---|---|---|---|
| Sales | Revenue Growth | (Rev₂ - Rev₁)/Rev₁ | Industry average |
| Marketing | CAC | Marketing spend / New customers | < 3× CLV |
| Customer | Churn Rate | Customers lost / Total customers | < 5% monthly |
| Operations | On-time delivery | On-time orders / Total orders | > 95% |
| Finance | Gross Margin | (Revenue - COGS)/Revenue | Industry-dependent |
1.4 A/B Testing Framework
(Diagram)
Key considerations:
- Sample size: Power analysis to detect effect
- Duration: Minimum 1-2 weeks (captures weekly cycles)
- Multiple testing: Bonferroni correction for multiple metrics
- Practical significance: Lift > minimum detectable effect
1.5 Customer Analytics: RFM Model
RFM (Recency, Frequency, Monetary) segments customers:
| Segment | Recency | Frequency | Monetary | Action |
|---|---|---|---|---|
| Champions | Recent | High | High | Reward, retain |
| Loyal | Recent | High | Medium | Upsell |
| At Risk | Long ago | High | High | Win back |
| Lost | Long ago | Low | Low | Reactivation campaign |
| New | Recent | Low | Low | Nurture |
📝 Practice Questions
Q1: What's the difference between a leading and lagging KPI?Leading KPI predicts future performance (e.g., number of demos scheduled → predicts future sales). Lagging KPI measures past performance (e.g., revenue). Good dashboards have both: leading indicators tell you what's about to happen, lagging indicators confirm what happened. Q2: Why randomize in A/B testing?Randomization ensures the treatment and control groups are comparable on both observed and unobserved variables. Without randomization, differences in outcomes could be due to pre-existing differences (confounding). For example, showing the new website to tech-savvy users would bias results. Q3: How does RFM help in customer retention?RFM identifies which customers need attention: "At Risk" customers (formerly high-value, now inactive) can be targeted with win-back offers. "Champions" can be thanked and retained. "Lost" customers might not be worth the reacquisition cost. It prioritizes retention spending based on expected value. Join Discord NextA/B Testing