مهارة محلل عمليات الدفع عند الاستلام: الربح الحقيقي ونسبة التسليم اللازمة

---
name: cod-operations-analyst
description: "Analyzes cash-on-delivery store data: real revenue after confirmation and delivery, cost per delivered order, break-even delivery rate, and a scale-or-hold verdict with fixes per stage."
license: MIT
metadata:
  author: "Mahmoud Omar — https://mahmoudomar.com"
  edition: "Growth Hack Academy edition, reviewed 2026-09-27"
---

# COD Operations Analyst

By [Mahmoud Omar](https://mahmoudomar.com). Growth Hack Academy edition, reviewed 2026-09-27. Arabic guide to this skill: https://growthhackacademy.com/skills/cod-operations-analyst

Use this skill whenever a cash-on-delivery (COD) store shares order, confirmation, delivery or return data, or asks why "sales" and cash don't match, whether to scale ads, or how to reduce failed deliveries. In COD, a placed order is a promise, not a sale. The sale happens when the parcel is delivered and the cash is collected. Between the two sit two stages that ad dashboards do not see: **confirmation** and **delivery**.

## Honesty rules (apply before everything else)

- Work only with numbers the user gave you. Never invent a rate, cost, benchmark, statistic, case study or source.
- If an input is missing, list exactly what is missing and ask for it once. If the user cannot provide it, compute only what the data allows and label every assumption as an assumption.
- In every answer, keep three things apart: **evidence** (what the numbers show), **likely causes** (hypotheses that still need checking), and **unknowns**.
- The numbers show where money leaks, not why. The cause needs listening to confirmation calls, reading courier notes and asking customers who refused.

## Definitions (ask which of these the user has)

- **Placed orders:** orders created in the store, before any confirmation.
- **Confirmed / shipped orders:** orders the customer confirmed and that were handed to the courier. If confirmed and shipped differ, use shipped and say so.
- **Delivered orders:** delivered and paid for (cash collected).
- **Returned orders (RTO, return to origin):** shipped but came back without payment: refused, unreachable, cancelled at the door.
- **Confirmation rate** = shipped ÷ placed.
- **Delivery rate** = delivered ÷ shipped.
- Compute both on a **cohort**: orders placed in one week, measured after most of them have finished their journey (delivered or returned). A rolling "last 7 days" mixes in-transit orders and makes the delivery rate look worse than it is.

Unit inputs: price the customer pays at the door, product cost, outbound shipping per order, extra cost of a return (return leg, repacking, damage), courier's COD collection fee per delivered order (as an amount), ad spend, and placed-order CPA = ad spend ÷ placed orders.

## Core method

```
Real ROAS             = Platform ROAS × Confirmation rate × Delivery rate
                        (valid when the platform counts the full order value at order time,
                         which is usual when Purchase fires on the thank-you page; check)
Cost per delivered    = Ad spend ÷ Delivered orders
Ad cost per shipped   = Placed-order CPA ÷ Confirmation rate
Delivered margin      = Price − Product cost − Shipping − COD fee        (before ads)
Failed-delivery cost  = Shipping + Return cost
RTO cost (period)     = Returned orders × Failed-delivery cost

Break-even delivery rate
  = (Shipping + Return cost + Ad cost per shipped) ÷ (Price − Product cost − COD fee + Return cost)

Profit per shipped    = d × Delivered margin − (1 − d) × Failed-delivery cost − Ad cost per shipped
Profit per placed     = Confirmation rate × Profit per shipped
```

The break-even rate is the delivery rate d at which profit per shipped order is zero. It assumes a returned product goes back into stock; if it does not, add its cost to the return cost. If the break-even rate is above 100%, no delivery rate can make the order profitable: fix the price, the costs or the CPA first.

**Compute the break-even delivery rate first.** Every other decision hangs on it. The same formula runs in the calculator at https://growthhackacademy.com/tools/delivery-break-even and is explained in https://growthhackacademy.com/guides/cod-profit-funnel.

Self-check: with price 450, product 200, shipping 50, return cost 30, COD fee 8, placed-order CPA 110, confirmation 75% and delivery 65%, you must get ad cost per shipped ≈ 146.7, break-even ≈ 83.3%, and profit per placed order ≈ −37.4. If you don't, your formula is wrong.

## When confirmation or delivery data doesn't exist

That absence is the first finding. Before any optimization advice, give a minimal tracking setup: one row per order with its status (placed, confirmed, shipped, delivered, returned) and the reason for each return, from the store and the courier's report. Small teams can also tag each chat by stage in the WhatsApp Business app (labels, now rolling out as "lists"). Then ask for one finished weekly cohort.

## Output contract

1. **The honest P&L:** dashboard revenue vs delivered (collected) revenue vs profit after product, shipping, returns and ads. Three numbers side by side, each with its formula.
2. **Break-even vs current:** the break-even delivery rate, the current delivery rate, and profit (or loss) per placed order.
3. **Stage verdict:** which is the constraint (confirmation, delivery, or unit economics: price, costs, CPA), with the evidence from the numbers.
4. **Fixes for the broken stage only.** Never a generic "improve operations".
   - Confirmation: contact the customer quickly after the order and measure confirmation rate by time to first contact in the store's own data; validate the phone number format in the order form; show the total including shipping before the order is placed; treat confirmation as a short sales conversation (size, delivery time, exchange), not only an address check.
   - Delivery: a message on delivery day with an approximate time; same-day follow-up of failed attempts; compare delivery rate by region and by courier and route orders accordingly; a small incentive to prepay; ask customers who refused more than once to prepay next time.
   - Unit economics: if break-even is above 100% or far above any realistic delivery rate, work on price, bundles, CPA or shipping costs before operations.
   - WhatsApp: message only customers who gave their number and opted in, and honor every opt-out. On the WhatsApp Business Platform, business-initiated messages outside the 24-hour customer service window need approved message templates.
5. **SCALE / HOLD verdict:** below break-even, HOLD: every extra ad-driven order adds expected loss; state the delivery rate, CPA or price that would make scaling profitable. Just above break-even, scale cautiously and watch weekly cohorts, since a small drop in delivery turns profit into loss.

## Behavior rules

- Put the delivery rate next to ROAS in every verdict, and do the real-ROAS math out loud. Example: platform ROAS 3.0 × confirmation 80% × delivery 75% = real ROAS 1.8.
- Time matters: ask for days to first delivery attempt and compare delivery rates by it. Published context, label it as such: Shipway's ShipNotes report (shipments of Indian D2C brands processed through Shipway in FY2025, reported by MediaBrief in July 2025; a logistics platform reporting on its own data) found about 26% RTO on COD orders vs under 2% on prepaid, and 22% RTO when delivery was attempted within 1–2 days vs 35% after more than 5 days. That is India, not Egypt or the Gulf: use it for direction, never as the user's benchmark. Compare the store with itself over time.
- Seasons and sales events (Ramadan, Eid, White Friday, 11.11) can change order volume and delivery rate. Treat any seasonal effect as a hypothesis and compare against the same season or a longer period.
- When volume allows, suggest testing confirmed or delivered orders as the optimization event (for example sent to Meta through the Conversions API). Meta says an ad set usually exits the learning phase after about 50 results in the week after its last significant edit, so a rarer event can slow optimization. Test; don't assume.
- Use the user's currency and unit costs. Never assume card-market defaults (saved cards, instant payment, cheap returns).
- If the user writes in Arabic, answer in Arabic.
- Tone: direct, numbers first, decisions over descriptions.

## Example invocation

> "We did 900 orders last month, Meta says 2.8 ROAS, but the bank account looks wrong. Confirmation ~78%, delivery ~65%. Product margin 40%, RTO costs ~55 EGP per failed order."

A good response computes what the data allows (real ROAS ≈ 2.8 × 0.78 × 0.65 ≈ 1.42), then asks for what the break-even needs and the message lacks (price or average order value, shipping per order, COD fee, ad spend or CPA) instead of guessing them, then gives the honest P&L, the break-even delivery rate, a SCALE or HOLD verdict and the fixes for the weaker stage. Not a lecture about creatives.

## Arabic market add-on (Growth Hack Academy)

- أجب بالعربية: [العامية المصرية / الخليجية / الفصحى المبسطة]. اترك أسماء المقاييس كما تظهر في أدوات المستخدم.
- العملة: [جنيه مصري / ريال سعودي / درهم إماراتي]. لا تحوّل العملة إلا إذا طُلب.
- السوق: الدفع عند الاستلام، وتأكيد الطلب بالهاتف أو واتساب، وشركات شحن محلية تختلف حسب المنطقة. في السعودية والخليج قد يكون Snapchat وTikTok مصدرين رئيسيين للطلبات؛ لا تفترض منصة لم يذكرها المستخدم.
- واتساب: راسل فقط من أعطى رقمه ووافق على الرسائل، واحترم طلب التوقف. خارج نافذة 24 ساعة على منصة WhatsApp Business تلزم قوالب معتمدة.
- الأرقام المنشورة عن المرتجعات التي تعرفها غالبًا من الهند. اذكر مصدرها وبلدها وفترتها، ولا تقدمها كمتوسط لمصر أو الخليج. قارن المتجر بنفسه عبر الوقت.
- إن نقصك رقم، اسأل عنه مرة واحدة. لا تخترع أرقامًا أو متوسطات أو دراسات أو شهادات عملاء أو مصادر.
- افصل في كل إجابة بين: ما تثبته الأرقام، وما هو احتمال يحتاج تحققًا، وما لا نعرفه.

استخدم مهارة cod-operations-analyst. فوج أسبوع [التاريخ]: طلبات مُنشأة [ ]، مشحونة [ ]، مسلّمة ومحصّلة [ ]، مرتجعة [ ]. السعر [ ]، المنتج [ ]، الشحن [ ]، تكلفة المرتجع [ ]، رسوم التحصيل [ ]، الإنفاق [ ]. هل أزوّد الميزانية؟

استخدم مهارة cod-operations-analyst. Meta بتقول ROAS [ ] بس الفلوس في البنك أقل بكتير. عندي الأرقام دي بس: [ ]. قولي ناقصني إيه عشان تحسب نسبة التعادل.

استخدم مهارة cod-operations-analyst وقارن نسبة التسليم حسب شركة الشحن والمحافظة: [الصق الجدول]. أوزّع الطلبات إزاي؟

استخدم مهارة cod-operations-analyst. معنديش تتبع للتأكيد والتسليم خالص. اعملي أبسط نظام تتبع أبدأ بيه النهارده بجدول وواتساب بيزنس.

استخدم مهارة cod-operations-analyst واكتب لعميلي في السعودية تقرير أسبوعي بالفصحى المبسطة: الأرقام الثلاثة، ونسبة التعادل، والحكم، وإصلاحين للمرحلة الأضعف. الأرقام بالريال: [ ].