If the price of cocoa moves on Monday, when can you see which products are exposed?
The answer depends on more than whether a new price has reached your system. You need to know which sites buy at that price, which product versions use the affected material, and whether the increase changes a decision about purchasing, reformulation, or profitability.
A product record can remain unchanged while the conditions that justified its release have moved on. The formulation is still approved; the specification is still current. But the next production run may face a different cost, a different supplier constraint, or a new regulatory requirement.
That gap deserves more attention. We measure how long it takes to develop and release a product. We should also measure how long it takes to recognize that something affecting a released product has changed, understand the consequences, and make an approved decision.
That is the case for a living PLM: a product lifecycle system that keeps released products connected to the inputs and conditions that affect them.
What changes after release
Release establishes an approved product definition. It does not make the assumptions behind that definition permanent.
Consider a raw material price increase. A movement in the market does not necessarily change your purchase price immediately. Contracts, inventory, and regional sourcing may limit or delay the effect. The useful question is therefore more specific than “what is the new price?” It is “where does this price apply, which products use that material, and what does it change?”
Answering that question requires the price to connect to the relevant raw material, purchasing context, site, and product version. A number arriving in a field is only the beginning. Someone still needs to assess its effect on material cost and, where the necessary commercial and manufacturing inputs are available, on product margin.
Regulatory changes create a similar problem. A new restriction may require a fresh assessment of products released years earlier. The result depends on what those products contain, where they are sold, and which requirements apply. An existing compliance assessment remains useful evidence of a previous decision, but it cannot answer a new question without being re-evaluated.
Supplier changes also reach beyond the supplier record. A discontinued grade, a revised specification, or an audit finding may affect multiple formulations across several plants. The organization needs to identify those relationships before it can decide whether to qualify an alternative, revise a formulation, or investigate a production risk.
In each case, the challenge is the same: connect a changed input to the product versions it affects, then bring that impact into a controlled decision process.
What the product record needs
A responsive lifecycle depends on the relationships the system can follow.
For formulation-based products, those relationships extend through raw materials, intermediates, finished formulations, site-specific versions, and substance composition. They also need to connect to the context that makes an assessment meaningful: applicable prices, supplier status, market requirements, and approved manufacturing conditions.
The distinction is not simply between a bill of materials and a bill of substances. A discrete PLM can hold structured material and substance information and support compliance rollups. The more useful distinction is between information that can be evaluated through connected records and information that remains isolated in documents.
A supplier declaration may provide essential evidence. But if the relevant composition exists only in an attachment, someone must extract and interpret it before the organization can assess exposure. Structured data makes that assessment repeatable, provided the relationships, calculation rules, and triggers are in place.
Different questions require different inputs. A substance restriction calls for composition data and applicable regulatory rules. A supplier disruption calls for supplier-to-material and material-to-product relationships. A cost assessment calls for quantities and applicable prices; a broader production-cost calculation may also require resources, overheads, and yield.
This is why flattening a formulation into a finished-product substance list is useful but insufficient as the whole product model. That list can support a composition assessment, but it does not preserve the stages, intermediates, order of addition, or processing conditions needed to describe how the product is made.
Local versions matter just as much. Consider a product manufactured in Lyon and Edison, New Jersey. The two sites may use different approved raw materials, sourcing arrangements, and process conditions. A price increase affecting one site should reach the versions exposed to that price, rather than trigger an investigation across every version carrying the same product name.
A living PLM preserves both the approved product definition and the context needed to reassess it. It should be possible to trace an exposure back to the source data, the applicable version, and the calculation or rule that produced the result.
From detection to approved action
A living PLM should shorten the path from a changed input to an informed response. That path has several distinct steps.
First, the system identifies exposure. When a raw material changes price, specification, or availability, the organization should be able to find the formulations that use it, including their quantities, sites, and approved versions. When a substance requirement changes, the assessment should reach the relevant compositions and markets.
Next, it recalculates what can be recalculated. Material-cost estimates should use the applicable quantities and prices. Compliance assessments should use the relevant composition, requirements, and effective dates. Where inputs are missing or uncertain, the result should make that limitation visible rather than present an unsupported answer.
Then the system prepares a response. An affected formulation might require a cost review, a supplier investigation, or a reformulation task. Candidate alternatives can be drawn from the organization’s material library, but their suitability still needs to be checked against the product’s performance requirements, manufacturing conditions, and approval context.
AI can help prepare that work, for example by organizing evidence or suggesting candidates for evaluation. It does not remove the need for qualification or authorize a product change.
The same boundary applies to lifecycle status. A changed input should trigger review; it should not silently turn a released formulation into a different approved product. Authorized people decide whether to revise a formulation, change an approval, update a controlled document, or retire a product.
Documents need to follow that controlled process. Connected formulation data can support the generation of draft specifications and data sheets. Issuing a new version remains a separate action, with the appropriate review, approval, and revision history. The goal is to reduce re-entry while keeping the published document tied to the correct approved product version.
The response also needs to reach the systems responsible for carrying it out. Approved formulation changes must reach manufacturing and ERP. Quality findings must connect back to the affected product record. R&D needs access to the cost, supply, and production context that prompted the review.
Those exchanges should have defined ownership, timing, and approval rules. Otherwise, the organization still depends on someone exporting a spreadsheet, reconciling versions, and explaining what changed.
That person has become the integration layer. His time is better spent assessing the consequences than reconstructing the relationships.
Portfolio decisions belong in this process, too. A material shared across many products may create a wider exposure than any single formulation suggests. Current assessments can help leaders see that concentration and decide where to investigate, reformulate, or review commercial viability. A proposed retirement still requires judgment about customers, commitments, alternatives, and the cost of making the change.
Measure your response lag
Before debating architecture, measure a real event.
Choose a raw material price increase, a supplier discontinuation, or a regulatory change that affected your organization. Establish when it became relevant to your products, when your team identified the affected versions, and when an authorized person made a decision based on that assessment.
Those dates reveal different sources of delay. The information may have arrived late. It may have arrived promptly but remained disconnected from the product records. The affected products may have been identified quickly, only for the review to stall because the evidence was incomplete or the decision owner was unclear.
For a price change, distinguish the market movement from the date it affected your purchasing context. For a regulatory change, distinguish publication from applicability and any transition period. The aim is to measure the time available to respond, not create a dramatic number from dates that mean different things.
Then examine where the work happened. How many records did people search? Which calculations had to be rebuilt? How many handoffs were needed before someone could approve an action? Could the team explain which product versions were assessed and why others were excluded?
This provides a practical test for a living PLM. Can the organization trace a changed input to its affected products, prepare a reliable assessment, and bring the result into an approved decision without reconstructing the portfolio by hand?
The answer may involve improving the PLM already in place, connecting it more effectively to other systems, or adding formulation capabilities that it does not currently provide. The requirement is the same: preserve the approved record while making changed conditions visible and actionable.
A lifecycle earns its name when it remains useful after release. Measure how quickly yours helps you respond.
.png)
.png)
