Cohort Analysis: Tracking Delivery Satisfaction Over Time
- Cohort analysis segments customers by acquisition window, revealing how delivery satisfaction evolves across life‑cycles.
- In Tier‑2/3 Indian markets, metrics like On‑Time Delivery Rate (OTDR) and Return‑to‑Seller Rate (RSR) directly impact COD‑refund churn.
- Edgistify’s EdgeOS, Dark Store Mesh, and NDR Management provide real‑time data pipelines that transform raw logistics data into actionable cohort insights.
Introduction
Indian e‑commerce has expanded beyond metros, now reaching Tier‑2 and Tier‑3 cities like Guwahati, Bhopal, and Surat. Consumers in these regions still favor Cash‑on‑Delivery (COD), making delivery reliability a critical differentiator. Yet, most retailers still look at aggregate KPIs—average delivery time, total orders—without understanding how satisfaction behaves over customer lifetime. Cohort analysis, a staple in SaaS analytics, can close this gap by mapping satisfaction trends to specific acquisition windows and delivery patterns.
What is Cohort Analysis?
A cohort groups customers who share a common attribute—in e‑commerce, usually the month or week of first purchase. By tracking each cohort’s performance over subsequent months, we can identify retention drivers, churn triggers, and, crucially for logistics, how delivery experiences shape long‑term loyalty.
| Cohort | Acquisition Date | Month 1 OTDR | Month 2 OTDR | Month 3 OTDR | Avg. RSR |
|---|---|---|---|---|---|
| Jan‑22 | 01‑Jan‑2022 | 92% | 90% | 88% | 4% |
| Feb‑22 | 01‑Feb‑2022 | 95% | 94% | 93% | 3% |
| Mar‑22 | 01‑Mar‑2022 | 90% | 88% | 85% | 5% |
Why Delivery Satisfaction Matters in India
| Metric | Impact on Indian Market |
|---|---|
| OTDR | Directly influences COD‑refund requests; each missed delivery can cost ₹10–₹30 per order in refunds. |
| RSR | High return rates inflate warehousing costs; in Tier‑2 cities, returns often require re‑delivery, doubling logistics effort. |
| Delivery Time | A 24‑hour delay can push a buyer from “buy” to “abandon” during festive rushes (Diwali, Christmas). |
Problem: Traditional dashboards report only *overall* OTDR, masking cohort‑specific dips.
Solution: Cohort analysis pinpoints where and when satisfaction erodes—be it a particular courier (e.g., Shadowfax) or a city hub (e.g., Guwahati).
Key Metrics for Delivery Cohort Analysis
| Metric | Definition | Target (India) |
|---|---|---|
| OTDR | % of orders delivered on or before promised date | ≥ 94% |
| Avg. Delivery Time | Mean days from order to handover | ≤ 2.5 days |
| RSR | % of orders returned within 30 days | ≤ 3% |
| COD Refund Rate | % of COD orders refunded | ≤ 1.5% |
| NDR (No‑Delivery‑Report) | % of orders with no delivery confirmation | ≤ 0.5% |
These metrics feed into a *Cohort Satisfaction Index* (CSI) calculated as: \[ CSI = 0.4 \times OTDR + 0.3 \times (1 - RSR) + 0.2 \times (1 - NDR) + 0.1 \times (1 - COD\_Refund\_Rate) \]
A CSI above 0.85 indicates high cohort satisfaction.
Problem‑Solution Matrix
| Pain Point | Root Cause | Cohort‑Level Insight | Strategic Intervention |
|---|---|---|---|
| Sudden drop in OTDR for Mar‑22 cohort | Shadowfax shift to night‑shift, causing missed windows | OTDR fell from 90% to 85% in month 2 | Implement NDR Management to auto‑retry delivery slots |
| High RSR in Tier‑3 cities | Inadequate last‑mile connectivity | RSR spiked to 6% in Guwahati cohort | Deploy Dark Store Mesh to create micro‑hubs |
| Elevated COD Refunds in Feb‑22 cohort | Poor courier communication about COD status | Refund rate went from 1% to 1.8% | Integrate EdgeOS real‑time alerts to courier teams |
Implementing Cohort Analysis with Edgistify
EdgeOS acts as the analytics backbone—collecting real‑time data from couriers, dark stores, and customer touchpoints. By feeding this into a cohort engine, EdgeOS produces monthly dashboards that automatically flag anomalies.
Dark Store Mesh is the distribution layer that reduces last‑mile distance. By aligning cohort data with mesh node performance, we can isolate whether a delivery dip is due to node proximity or courier inefficiency.
NDR Management automatically escalates undelivered orders to alternative carriers. Cohort analysis shows that after NDR implementation, OTDR improved by 3% in affected cohorts.
Case Study: Guwahati Delivery Cohort (Jan‑22)
- Initial OTDR : 88%
- Intervention : Introduced a Dark Store Mesh node in North Guwahati; integrated NDR with Shadowfax API.
- Result after 2 months : OTDR rose to 94%; RSR dropped from 5% to 2.5%; CSI improved from 0.78 to 0.86.
This demonstrates how cohort‑driven decisions translate into measurable logistics KPIs.
Best Practices for Sustainable Cohort Analysis
- 1. Segment by Acquisition Window : Monthly cohorts capture seasonal effects (e.g., festive rush).
- 2. Align Metrics with Business Goals : Tie OTDR and RSR to revenue loss thresholds.
- 3. Automate Data Pipelines : EdgeOS should pull data from couriers, dark store APIs, and payment gateways every 12 hours.
- 4. Set Alert Thresholds : NDR > 0.3% or OTDR < 92% triggers a workflow in the logistics ops dashboard.
- 5. Iterate Quarterly : Re‑define cohorts and metrics every quarter to capture evolving consumer behavior.
Conclusion
In India’s rapidly evolving e‑commerce landscape, understanding *how* delivery satisfaction changes over time is as crucial as measuring it. Cohort analysis turns raw logistics data into strategic intelligence, pinpointing the exact moments and segments where customer delight falters. When powered by Edgistify’s EdgeOS, Dark Store Mesh, and NDR Management, retailers can proactively engineer smoother deliveries, lower returns, and higher loyalty—especially in Tier‑2/3 markets where COD remains king.