COMMERCE StoriesSubscribe
← Commerce courseCOMMERCE FROM ZERO · MODULE 8 · 8 min READ · FULL COURSE

Order Management

From "Buy" Button to Promise Kept

About access

Buyers receive the course-start link after checkout. This lesson link is unlisted, not account-based access control; anyone with the direct link can open it.

Labels: [Sourced] checked against a source listed at the end. [General] standard industry knowledge. [Our view] opinion, labelled as such. [Illustrative] made-up numbers to teach an idea.

1. The story: an airlift that taught the world to send orders to computers

After World War II, the Allied sectors of Berlin were cut off by the Soviet blockade, and for 13 months supplies came in by air. The X12 standards body says Air Force Colonel Edward Guilbert, as traffic manager of the airlift, decided the key was to send standardized information about each cargo ahead of the planes, in effect electronic manifests. He later developed standard electronic cargo messages at DuPont and led the Transportation Data Coordinating Committee for 19 years, and is known as the "Father of EDI." [Sourced: X12 awards page, which draws partly on Ralph Notto's 2005 book; X12 is an interested party] By 1968 railroads, airlines, truckers and shippers formed the Transportation Data Coordinating Committee, and in 1975 it released the first US electronic data interchange (EDI) standard for rail. [Sourced: X12; iWay history page] Before that, each company used its own formats, with conflicting definitions of the same data. [Sourced: X12]

Why it matters: an order is a promise written in a shared language. EDI was the first shared language, and many large retailers still use it for orders. [General]

2. The one idea

Order management answers: who ordered what, can we deliver it, from where, by when, and did we? It runs the whole life of an order from placing it to payment. [General]

3. The kitchen-table version

A restaurant order. The waiter writes it (capture), the kitchen checks it has the ingredients (availability), the cook makes it (fulfillment), the waiter brings it (delivery), you pay (cash). Order management is the waiter, the ticket rail and the promise "ten minutes".

4. The order's life, in six steps [General]

  1. Capture: web, app, store, phone, EDI from a retailer.
  2. Validate: address, payment, credit, fraud checks.
  3. Promise: available-to-promise (ATP) is the quantity that can be promised for a date, counting uncommitted stock and planned receipts, per Microsoft's documentation. [Sourced: Microsoft Dynamics docs] Capable-to-promise adds production capacity for made-to-order goods.
  4. Source: pick the location that will fulfill it.
  5. Fulfill and deliver: Modules 10-12.
  6. Invoice and collect: the "order to cash" cycle ends when payment is recorded. [Sourced: Stripe/Microsoft overviews; General]

5. Sourcing: the decision that makes or loses money

"Distributed order management" systems choose a fulfillment node using rules such as inventory availability, proximity, shipping cost and location capacity, and decide whether to split an order. [Sourced: Fluent Commerce, IBM, Salesforce docs; vendor descriptions] Splits raise delivery and handling cost, which academic papers on omnichannel split orders discuss. [Sourced: JKU paper abstract]

Worked example [Illustrative, computed by script]. Customer orders 3 items. Costs are made up for teaching.

OptionCost
East DC ships all 3$4.50 + 3 x $0.60 pick = $6.30
West DC ships all 3$6.80 + 3 x $0.60 = $8.60
Split: 2 from East DC, 1 from a store($4.50 + $1.20) + ($7.00 + $0.60) = $13.30
Cheapest is not always possible: if East DC lacks one item, you pick between splitting ($13.30) or waiting. A good system also weighs speed and stock risk, not only cost. [Our view]

6. Keeping the promise: OTIF

Big retailers measure suppliers on on-time, in-full (OTIF). Walmart's requirement was 98% from September 2020, and from February 1, 2024 it moved to 90% on time and 95% in full, per Forbes. [Sourced: Forbes, Feb 1 2024] Earlier, a 2017 policy (announced July 2017) fined suppliers 3% of the value of a late or early order for fast-turning items below 75% on-time-in-full; those terms have since changed. [Sourced: IndustryWeek, Jul 14, 2017; Retail Dive] Combined, 90% x 95% means at most about 85.5% of orders could be both on time and in full if the two failures never overlap, a point to check against how Walmart actually measures it. [Illustrative arithmetic; our interpretation]

7. Where AI fits

  • Promise dates that learn from real carrier and warehouse performance. [General]
  • Sourcing choices that weigh cost, speed, stock risk and markdown risk together, not rule by rule. [General]
  • Agents that place orders on behalf of buyers (Module 21): the order arrives as a machine message, so clean item data (Module 1) and clear policies matter even more. [Our view]

8. Where it breaks

Phantom inventory (Module 9), cancelled orders after payment, partial shipments, address errors, fraud, systems that can't see each other (Module 3).

9. Exercise

Make up three fulfillment locations and costs for a 2-item order. Which is cheapest? Which is fastest? Would you pay $2 more for a day sooner?

10. Quiz

  1. What does ATP tell you? The quantity you can promise for a date.
  2. Why can splitting an order cost more? Extra shipments and handling.
  3. What is OTIF? On-time, in-full measure of delivery to a customer.
  4. What was EDI replacing? Company-specific, conflicting formats.

Sources

CURIOUS? TEST THE CLUES

Curiosity check

Pick an answer and see why. No scores, no pressure. All shop examples are invented practice scenarios.

01 What does ATP (available to promise) tell you?
02 What is OTIF?