Challenging a supplier price increase: cost decomposition method
A supplier announcing an increase and citing inflation is making a claim about their own cost structure, and that claim decomposes. Establish the weight of each cost component in the supplier’s base, apply the movement of the relevant published index only to the component it affects, and sum the result. The justified increase is the weighted sum of index movements, not the headline rate of any single index. Where materials account for 47% of total cost, a 15% rise in material prices justifies about 8% at product level. The remaining share is a claim about something else, and the supplier can be asked what.
- Applies to
- Strategic and Bottleneck categories, and Leverage categories where commercial mechanics dominate
- Unit of analysis
- One contract or one renewal, not a category portfolio
- Position in the lifecycle
- Before the increase is accepted
A general claim about inflation is not a claim about your price
A price increase letter usually references an aggregate: consumer price inflation, producer price inflation, or a general statement about market conditions. None of these describes what happened to the supplier’s costs.
Aggregate indices measure a basket. The supplier’s cost base is a specific set of components in specific proportions — a particular metal, an energy contract in one country, a wage base under one collective agreement, a freight lane. Movement in the basket and movement in that set are different quantities, and they routinely diverge in direction as well as magnitude. Aggregate inflation can fall in a quarter in which the supplier’s dominant input rises. Eurostat publishes producer price indices by detailed product group precisely because the aggregate conceals this dispersion.
This matters commercially because the burden of specificity sits with whoever raises it first. A buyer who responds to a general claim with a general objection has an argument about credibility. A buyer who responds with a decomposition has an argument about arithmetic, and the second is easier to win because it can be checked.
Cost decomposition and should-cost analysis answer different questions
The two are frequently conflated and the distinction determines which one to run.
Should-cost analysis constructs the price from the bottom up — materials, labour, overhead, logistics, margin — to establish what a product or service should cost in absolute terms. It tests the baseline. It requires a bill of materials structure or an equivalent, and it is the heavier exercise.
Cost decomposition tests a change in price against movement in the components that produced it. It takes the existing price as given and asks only whether the increase on top of it is supported. It is faster, needs less data, and is the correct instrument when a letter has arrived and a response is due.
A buyer who suspects the baseline itself is wrong needs should-cost. A buyer facing an announced increase needs decomposition first, and should-cost afterwards if the decomposition suggests the base was already high.
Decompose the claim, one component at a time
Step 1 — Establish the weight of each cost component
Break the supplier’s cost base into components and assign each a share of total cost. A workable set for most manufactured goods and services:
Cost components and the evidence that establishes their weight| Component | Typical evidence source |
|---|
| Direct materials | Bill of materials structure, material families, published cost benchmarks for the sector |
| Direct labour | Regional labour rate benchmarks, sector wage settlements |
| Energy | Energy intensity of the process, national industrial energy tariffs |
| Logistics and freight | Route, mode, freight index for the lane |
| Overhead and indirect | Sector norms; the least observable component |
| Margin | Residual, or sector benchmark |
The weights do not need to be exact. They need to be defensible and stated as estimates. A decomposition presented as approximate and open to correction invites the supplier to supply better figures, which is itself a useful outcome — the alternative is that the supplier supplies nothing and the general claim stands.
Where a bill of materials structure is available, material weight can be established directly. Where it is not, sector cost-structure benchmarks give a starting position that the supplier will correct if it is wrong in their favour.
Step 2 — Match each component to a published index
Each component takes the index that actually governs it, not a general one. The consumer price index is the wrong instrument for every component in this table.
Cost components and the published index families that govern them| Component | Index family | Publisher |
|---|
| Direct materials | Producer price indices by product group; commodity indices for the specific material | Eurostat, national statistical offices, commodity exchanges |
| Direct labour | Labour cost index by economic activity and country | Eurostat, national statistical offices |
| Energy | Industrial energy price indices by country and consumption band | Eurostat |
| Logistics | Freight rate indices by mode and lane | Route-specific published indices |
| Overhead | Producer price index for services, or the general PPI as a proxy | Eurostat |
The point of specificity is that indices for different components move independently and often in opposite directions within the same period. Applying a single index to the whole cost base is the same error as accepting the supplier’s aggregate claim, performed in the other direction.
Step 3 — Compute the justified increase
The justified increase is the sum of each component’s weight multiplied by that component’s index movement:
Justified increase (%) = Σ ( weight of component i × index movement for component i )
Worked example, using illustrative weights:
Illustrative worked example — the weights below are examples, not benchmarks| Component | Weight | Index movement | Contribution |
|---|
| Direct materials | 47% | +15.0% | +7.05% |
| Direct labour | 18% | +4.5% | +0.81% |
| Energy | 9% | −6.0% | −0.54% |
| Logistics | 7% | +3.0% | +0.21% |
| Overhead | 12% | +2.0% | +0.24% |
| Margin | 7% | 0.0% | 0.00% |
| Total | 100% | | +7.77% |
A supplier citing a 15% rise in material prices in support of a 15% price increase is, on these weights, justified in approximately 8%. Note that the energy line reduces the total: a decomposition is not a device for producing a lower number, it is a device for producing the right one, and it will sometimes support the supplier.
Step 4 — Put the residual to the supplier as a question
The gap between the claimed increase and the justified increase is not necessarily unjustified. It is unexplained, which is a different thing and a more productive way to raise it.
Legitimate explanations exist: a step change in volume that has moved the supplier off a pricing tier, a currency exposure not visible in local-currency indices, a capacity investment being amortised, a change in the specification, a regulatory cost such as the EU carbon border adjustment mechanism on an input. Each of these is checkable and each changes the negotiation into a discussion of a specific number.
The question that follows a decomposition is therefore not why are you increasing prices but our decomposition supports approximately 8% — what accounts for the remaining 7%. The first question invites a narrative. The second requires a figure.
Accuracy, and why the accuracy band is the point
A cost model built on clean data and category knowledge can be expected to fall within roughly ±10–15% of actual cost. That is an accuracy expectation under good inputs, not a claim about any particular model’s performance.
Stating the band is what makes the model usable in a negotiation. A figure presented without an error margin invites a dispute about whether it is right. A figure presented as approximately 8%, within a band of roughly ±10–15% on the underlying cost estimate invites a dispute about the inputs, which is the dispute worth having.
Models go stale. Annual revision is the minimum; quarterly for categories where the dominant input is volatile.
The result is cost avoidance, and it is systematically under-claimed
An increase reduced from 15% to 8% produces no line in the ledger that anyone can point at. Nothing got cheaper. What happened is that a cost which would otherwise have been incurred was not — which is cost avoidance in the standard classification, distinct from hard savings that reduce a budget line and from soft savings attributed to non-price improvements.
Two consequences follow, and buyers tend to get both wrong in the same direction.
The first is that cost avoidance has to be classified as such before the negotiation, not argued about after it. The comparator matters more than the figure: a claim measured against last year’s actual paid price is defensible, and a claim measured against the supplier’s opening offer is the version finance functions discount hardest. Choosing and recording the comparator up front is what determines whether the number survives.
The second is that a successfully defended increase is frequently not claimed at all. Where a category index rose and the negotiated price held, the difference is real avoided cost and it disappears because nobody wrote it down. Under-reporting of avoided cost and of non-price gains is the more common error in practice, not over-reporting.
What this method does not do
- It does not establish what the price should be — see the distinction from should-cost above.
- It does not work where the cost structure is genuinely unknowable. Highly integrated products, proprietary processes and services with no comparable market are the cases where component weights cannot be estimated with enough confidence for the arithmetic to carry weight. In those categories leverage comes from alternatives rather than from cost analysis.
- It does not address whether to accept the increase at all. A fully justified increase may still be worth refusing if an alternative exists, and an unjustified increase may be worth accepting if none does.
- It is not a substitute for an indexation clause. Where the contract already specifies an index and a formula, the decomposition tests whether the supplier has applied their own clause correctly — a narrower and usually faster exercise.
The demand is documented; the savings percentages are not
The demand side is well documented. In the RS and CIPS Indirect Procurement Report 2026 (n=448, UK and Ireland), inflation and higher costs were named by 68% of respondents as a top challenge for the coming twelve months. The same survey series recorded 37% in its 2024 edition. Concern has approximately doubled within two years in a like-for-like series.
The picture turns over quickly. CIPS Pulse data recorded concern about inflation falling from 59% to 41% between the first and second quarters of 2025 as availability and political risk displaced it, and by the 2026 edition of the CIPS Global State of Procurement & Supply US protectionism had left the top three supply chain risks for the first time in over a year. Any cost picture established at the start of a year has been reordered by the middle of it.
The decomposition arithmetic requires no source. It is a method rather than a measurement, which is why the worked example above is presented with its assumptions visible rather than cited.
Where this runs in EXOS
The method above is executed as a chain rather than a single analysis, because each step produces the input for the next.
The scenario chain that executes this method| Step | Scenario | What it produces |
|---|
| Decompose the supplier’s cost structure | S2 · Cost Breakdown / Should-Cost | Cost waterfall by component, should-cost gap against the quote, per-line-item confidence classification, negotiation leverage points, market benchmark comparison |
| Apply external index movement to the category | S6 · Forecasting & Budgeting | Market intelligence overlay layering CPI, PPI, commodity and FX movement onto the baseline; macro sensitivity ranking showing which external factor moves this category most |
| Convert the decomposition into a negotiating position | S21 · Negotiation Preparation | Power balance analysis, opening and concession sequence, counter-argument preparation against the supplier’s likely responses, walk-away trigger conditions |
| Prepare the outcome to be presented and defended | S4 · Savings Case | The claim rechecked against its comparator, the hard/soft/avoided split, the questions Finance will ask with answers to each, a finance-ready statement, and — the step most often missed — what may have been under-claimed |
The last step is where a defended increase becomes a recorded outcome. Its purpose is to strengthen the position rather than to test it: the comparator gets named, the avoided cost gets separated from hard saving, and the questions that will be asked get answered before they are asked.
Continuous component — Inflation Monitor. The chain above answers the question once, at the moment the letter arrives. The Inflation Monitor tracks the indices governing the category’s cost components on a scheduled basis, which changes the position at the start of the conversation: the supplier’s claim is tested against a picture that already exists rather than one assembled in response. It also supplies the category index movement that separates avoided cost from negotiated saving.
Monitoring in EXOS is scheduled across a defined supplier and category list. It is not continuous surveillance, and specialist monitoring platforms cover that ground more deeply.