Choose a phantom assembly when a component grouping should normally be built within the parent work order. Choose a stocked subassembly when the business needs to make, hold, count or allocate that intermediate item separately. In NetSuite, this decision belongs at the relevant BOM or work-order relationship; an item can be treated as phantom in one structure and stock in another.
The practical question is where you need an independently controlled inventory boundary. Saving a separate production step is useful only if the resulting material, scheduling and traceability records still match how the factory operates.
Walk through one batch and ask when the intermediate assembly becomes a recognizable object. Can it wait on a shelf? Is it inspected before the parent build? Can another production line take it? Does it have its own shelf life or serial identity? Would a warehouse employee count it separately?
A stocked structure is often worth considering when those answers require independent control. A phantom can fit when the grouping mainly organizes engineering content and is assembled as part of the parent process. These are design criteria, not absolute rules about what every NetSuite configuration permits.
Document exceptions. A normally unstocked module may occasionally be built for warranty replacements. NetSuite's phantom concept does not mean the item can never be produced as a stock item. Design that exceptional flow explicitly instead of selecting a structure from a simplified “stocked versus never stocked” definition.
NetSuite uses Item Source values on BOM-related records and work orders to describe the intended supply path. A phantom source expands the subassembly's components into the higher-level build. Stock sourcing instead represents drawing the intermediate assembly from stock.
Defaults deserve careful review. A blanket setting that marks all subassemblies phantom can change more relationships than the engineer intended. Clearing that setting does not necessarily reload or remove previously expanded components. Review the actual work-order lines rather than treating a checkbox change as proof of a clean structure.
For Advanced BOM accounts, examine the relevant revision as well as the assembly. A design decision recorded only in an item description will not reliably tell planners which relationship is supposed to consume stock.
| Decision | Phantom relationship | Stocked relationship |
|---|---|---|
| Production instruction | Intermediate components are part of the parent build path | Intermediate production can be managed separately |
| Buffer before parent assembly | Requires a deliberately designed physical and record approach | Can support a separately counted inventory buffer |
| Competing parent demand | Review shared component requirements | Review demand against intermediate stock and future supply |
| Quality release | Prove how inspection connects to the parent process | Prove how intermediate release affects later consumption |
| Cost explanation | Follow expanded components and applicable conversion costs | Follow intermediate production cost and later consumption |
| Traceability | Validate component-to-parent evidence | Validate intermediate identity and both production links |
The table describes questions to validate. It does not promise that enabling phantom sourcing configures quality, finite scheduling or genealogy automatically.
Two finished products each need one identical power module. A module contains one board and two connectors. Tomorrow's orders require 30 units of Product A and 20 of Product B. The component requirement for 50 newly built modules is 50 boards and 100 connectors.
Suppose the warehouse already holds 15 approved modules. Under a stocked design, the planner can evaluate consuming those 15 and building another 35, which requires 35 boards and 70 connectors. Under a phantom relationship, the parent orders may instead call for the underlying components. The team must not assume that existing modules will automatically satisfy that expanded demand.
Neither choice is inherently wrong. If the 15 modules are service stock, preserving them may be intentional. If they are meant for tomorrow's production, ignoring them can cause unnecessary work. The acceptance test must show which material is actually consumed and which balances remain.
Use one parent BOM with a stock relationship and another with a phantom relationship for the same intermediate item, if that reflects an authorized operating requirement. Compare the resulting component lists, planned supply and execution records.
Include an existing intermediate balance in the test. A zero-stock example cannot reveal whether the design accidentally bypasses usable subassemblies. Also test a nested structure in which the phantom itself contains a stocked subassembly. Confirm how far the expansion goes rather than assuming that every lower level behaves identically.
Keep a before-and-after extract when changing Item Source on a test work order. Have the planner explain the difference line by line. If no one can explain why a component appeared twice, resolve that before any production release.
Ask quality to identify the exact evidence needed for the intermediate module. The answer may involve test results, material lots, an individual serial number or approval by a qualified operator. A component grouping is not automatically a sufficient traceability record.
Run a backward trace from a finished parent and a forward trace from a controlled component. Establish whether the required intermediate identity survives the selected process and reports. If the business needs to recall a separately identifiable module across several parent products, that requirement should influence the structure decision.
For a phantom path, determine where inspection is triggered and how a failed intermediate prevents later use. For a stock path, determine whether a separately stored module can be consumed before quality release. Both designs need an enforceable answer, supported by the features actually installed.
List the real work performed to build the intermediate and the parent. Decide where setup, labor and machine effort will be scheduled or recorded. Moving a relationship to phantom does not make the physical work disappear.
Inspect the applicable routing and cost model in the test account. Verify that effort is neither omitted nor captured once on an intermediate order and again on the parent. Standard-cost and average-cost assemblies can produce different financial behavior; a cost accountant should approve the comparison and any inventory treatment.
Use a stable batch size for the comparison. If one scenario spreads setup across 100 modules and the other across ten, the cost difference may arise from the batch assumption rather than the phantom setting itself. Record that distinction before deciding one structure is cheaper.
Changing an active relationship requires a controlled cutover. List open parent orders, open intermediate orders, current intermediate stock, component commitments and unused printed travelers. Decide which population stays on the earlier structure and which uses the new one.
Do not delete an intermediate balance to make a new phantom model appear clean. Determine whether it will be consumed, sold as service stock, disassembled through an approved supported process or otherwise disposed of. Each choice has operational and accounting consequences.
After the first affected batch, reconcile parent output, intermediate movement and component consumption. A shorter work-order list is not the success measure; the success measure is an understandable production record with the required controls intact.
Keep the final decision at assembly, BOM revision and location level where relevant. Include the reason, permitted exceptions, quality requirements, inventory ownership and the approved test case. Revisit it when the factory introduces an intermediate buffer, new inspection point or different batch strategy.
For an existing structure that creates confusing supply or consumption, use CuriousRubik's NetSuite support services to scope a review around one mixed stock-and-phantom example. The most useful evidence is the expected physical flow alongside the actual work-order lines.
Yes. NetSuite supports an Item Source choice at the relationship level. Review defaults and the saved component lines so the intended choice is preserved on the relevant BOM revision and work order.
No. Phantom assemblies are typically used within a higher-level build, but a business may also build them as stock items for an exceptional purpose. Define and test that separate flow rather than treating phantom as a permanent prohibition.
Do not assume that it will. A phantom relationship expands underlying components for the higher-level build. Include existing intermediate stock in acceptance testing and verify the actual consumption path before releasing orders.
No. It can reduce separate production administration, but cost depends on material, batch sizes, conversion effort and the configured costing model. Compare the same physical work and have finance validate where costs are recorded.
A need for independent inventory control is a strong reason: separate counting, buffering, allocation, inspection or traceability. Confirm those requirements with production and quality, then test whether the chosen transaction flow preserves them.