Notes

Make-to-stock vs make-to-order: what happens before the order?

A customer wants 20 lights dispatched in three days. Compare finished stock, common components and production from raw material to see which promise each can support.

A customer asks for 20 industrial bench lights, dispatched within three days. Sales knows how long assembly takes: two days. The order sounds achievable.

That answer depends on what is ready now. Twenty finished lights, twenty sets of common parts and enough raw material for twenty lights are three different starting points.

Make-to-stock (MTS) finishes products before a customer order arrives. Make-to-order (MTO) makes the product after the order. Assemble-to-order (ATO) stocks common components and completes assembly after the order. These are the distinctions described in ASCM's supply chain dictionary.

The choice changes which work the customer waits for. It also changes what the business has already committed to making.

One product, three starting points

Our fictional manufacturer offers lights assembled from a common housing and standard selectable fittings. Making the housings takes four working days. Assembly takes two. Final checking, packing and dispatch take one more.

These are elapsed times for the whole 20-light order, not times per light. Transport to the customer is excluded because the request is for dispatch within three days.

Assume the correct stock is usable, available and not allocated to another order. For this first comparison, production capacity is ready when needed.

Approach Available when the order arrives Work still required Earliest dispatch
Make-to-stock 20 finished lights of the requested version Check, pack and dispatch Day 1
Assemble-to-order 20 common housings and all required fittings Assemble, then check, pack and dispatch Day 3
Make-to-order Raw material and purchased parts for 20 lights Fabricate, assemble, then check, pack and dispatch Day 7

The work hasn't disappeared in MTS or ATO. Some of it happened before the customer committed. That is why the response can be shorter.

The point where work switches from preparation for expected demand to fulfilment of a specific order is often called the customer order decoupling point. Here it sits at finished stock, common components or the raw-material stage.

See what a queue does to the promise

All three approaches remain visible below. Add a wait for the first production stage that still needs doing. The finished-stock case does not enter that queue because its fabrication and assembly are already complete.

20 lights / one customer request

Work before and after the order

Day 0 is the order. The customer asks for dispatch by day 3. All three rows include the same fabrication, assembly and final dispatch work.

Work completed before the orderWork after the orderF: Fabricate, 4 daysA: Assemble, 2 daysP: Check, pack, dispatch, 1 day

Scroll the timeline sideways to see every stage and the day scale.

Dispatch timing with a 0-day production waitMTS: dispatch day 1, meets day3 request. ATO: dispatch day 3, meets day3 request. MTO: dispatch day 7, misses day3 request. Before-order stage positions are illustrative offsets, not stock ages.-6-4-2024681012OrderBy day 3DayMTSFAP1ATOFAP3MTOFAP7Outlined gap: wait for capacity. Graphite work was completed before the order.
ATO / available at day 0

20 housings and complete fitting sets

Dispatch at day 3. Meets the three-day dispatch request. The customer waits for the remaining production work and the selected queue.

Separate alternatives for the same order. All stock is usable and uncommitted.
ApproachStock at orderWait after orderDispatch dayBy day 3?
MTS20 correct finished lights01Yes
ATO20 housings and complete fitting sets03Yes
MTORaw material and purchased parts for 20 lights07No
Calculation and boundary

MTS: one day for final check, packing and dispatch. ATO: queue + two assembly days + one dispatch day. MTO: queue + four fabrication days + two assembly days + one dispatch day. Durations apply to the full order, not each light.

No material shortages, other orders, failures, transport or later replenishment are modelled. The negative day positions simply place completed work to the left of the order. They do not describe inventory holding time.

Fictional 20-light order placed at day 0. Fabrication takes 4 working days, assembly 2 and final check/pack/dispatch 1. Correct uncommitted stock is available at the stated stage. The slider adds a wait before the first production stage after the order, with no other delays. Negative positions illustrate work completed before the order; they do not measure inventory age. Transport is excluded.

With no queue, ATO reaches dispatch on day 3. Add a two-day wait for assembly and it moves to day 5. The parts can be ready while the order is still too late for the customer's request.

MTO moves from day 7 to day 9 with the same added wait. Its raw material is available throughout. Availability of material and availability of production capacity are separate checks.

The APICS master-planning excerpt connects order promises with uncommitted stock and capacity. A label such as “assemble-to-order” cannot establish a date by itself.

For a factory application, Vrolen's production scheduling guide explains how machines, people, materials and existing commitments affect the executable plan.

Moving work earlier also moves the stock decision

In the MTS case, the manufacturer chose the finished version before knowing this customer's order. If customers buy a different fitting, those finished lights may wait while the requested version is unavailable.

ATO leaves that final choice open for longer. The common housing can serve several standard versions, provided the required fittings and assembly capacity are available. It still needs a decision about how many housings and fittings to hold.

MTO postpones more production until demand is confirmed. In our example it still holds raw material and purchased parts. “Made to order” does not mean “no inventory”, and it doesn't mean every order requires a new design. Engineer-to-order adds a separate design requirement.

The ABC/XYZ inventory example helps review item value and demand patterns. Use that evidence alongside the customer's acceptable wait, the cost of holding each stock form and the chance that a version becomes obsolete. There is no universal best approach for every item in the catalogue.

Make-to-order still needs advance planning

Waiting for a firm order before manufacturing doesn't remove the need to prepare people, equipment, supplier capacity or long-lead parts.

Suppose every confirmed order needs the same housing material. The business can plan that shared requirement before it knows the exact finished-product mix. When orders arrive, the dated production plan and bill of materials translate them into component requirements.

The MRP worked example explains that calculation. Forecast demand can inform the wider plan; component demand for a particular build comes from what that build contains.

That is also a useful connection to Foreviq, the forecasting project I'm working on. The forecast supports preparation. The order promise still needs the stock, capacity and dates behind it.

Choose by product family, then check a real order

A business can stock a few popular finished versions, assemble less common standard combinations and make unusual products to order. Its planning rules should make the difference clear to sales and production.

For the bench-light order, the next conversation is specific. Are the finished lights the requested version? If using ATO, are every required fitting and an assembly slot available? If using MTO, will the customer accept the longer dispatch time?

Agree that dispatch promise before accepting the order. Then measure against it. The OTIF article follows the next question: did the full order actually reach the customer within the agreed delivery window?

Sources and example notes

The product, stock positions and durations are fictional. Each approach is a separate alternative for the same order. The comparison isolates the work remaining after receipt of the order; it does not optimise inventory, estimate stockout probabilities or include replenishment, cost, yield loss or transport.

NextMRP explained: what to order and when

Working on something like this?

I work with operations and transformation teams on operational excellence, digital transformation programs, supply chains and industrial AI. If this sounds like your line, your program or your problem, I’d be glad to compare notes.

Abolfazl ShirkavandImage preview unavailable