Introduction to Descriptive Statistics in eCommerce

eCom Data 101

July 29, 2025

Welcome! This session is designed to help you understand the basics of descriptive statistics and how they apply to real-world eCommerce data like delivery times, shipping performance, and online testing results.

What You’ll Learn

By the end of this session, you’ll be able to:

• Understand key statistical terms like average, median, and percentiles
• Interpret patterns in your delivery and checkout data
• Identify anomalies and improve performance based on insights

1. What Are Descriptive Statistics?

Descriptive statistics help you summarize large sets of data so you can see patterns, trends, and areas for improvement.

Why it matters in eCommerce:

You can use stats to analyze:

• 🚚 Delivery times (e.g., how long it usually takes to ship an order)
• 🛒 Order values (e.g., average revenue per checkout)
• 📉 Test results (e.g., which shipping option performs best in an A/B test)

Example: If your SLA (Service Level Agreement) promises delivery in 4 days, descriptive stats can help you back that up: “80% of our orders arrive within 4 business days.”

2. Central Tendency: Mean, Median, Mode

These three terms help you understand what’s typical in your data:

• Mean (Average): Good for metrics like Average Order Value. But be careful—it’s sensitive to extreme values.

• Median (Middle Value): A better measure when you want to know what the typical customer experiences. It’s not thrown off by outliers.

• Mode (Most Frequent Value): Tells you what shows up most often—like your most popular shipping destination.

3. Variability: How Spread Out Is the Data?

Understanding the spread of your data shows how consistent (or inconsistent) your operations are.

• Range (Min to Max): Quick snapshot of best vs. worst-case delivery times.

• Standard Deviation (SD): Shows how tightly delivery times cluster around the average. Lower SD = more consistent performance.

• Outliers: Any data point more than 3 standard deviations from the mean is considered an anomaly. These could be system issues, regional delays, or one-off errors.

4. Percentiles & the TP80 Metric

Quantiles split your data into equal parts and help identify performance benchmarks.

• TP80 (80th Percentile): Tells you the time it takes for 80% of your orders to arrive. This is often used in SLAs and delivery promises.

• TP90 (90th Percentile): Captures edge cases—like the slowest 10% of orders. Useful for understanding operational risk.

Activity idea: Try calculating TP80 and TP90 using your delivery data. What does this say about your performance?

5. Wrap-Up & Key Takeaways

Here’s a quick summary of what we covered:

✅ Central tendency (mean, median, mode) shows typical performance

✅ Variability (range, SD) reveals consistency

✅ Percentiles like TP80/TP90 help set realistic SLAs

✅ Outliers (3 SD rule) are red flags for deeper investigation

✅ These tools also apply to conversion rates, CTR, CVR, and more

Need help applying these metrics to your store's performance?

Reach out, we’re here to help!