Indirect Spend Category Strategy Framework
“A category strategy earns its keep when someone owns the decision it points to, and can be asked about the result a quarter later.”
| Statistic or documented observation | Source | How to use it |
|---|---|---|
| US federal guidance issued in 2019 defines category management as buying common goods and services as an enterprise, and its maturity model opens at unaligned, decentralized spending | OMB guidance | Frame indirect spend's default state and the direction of travel, not a savings promise |
| A survey of sourcing projects classified on the Kraljic matrix found the full range of tactical sourcing levers used in every quadrant | Hesping and Schiele | Treat the matrix as a way to structure debate about a category, not a fixed prescription of tactics |
| A 2025 US federal audit found category managers generally did not fulfill their responsibilities where senior leaders did not hold them accountable | GAO | Read it as evidence that decision rights and accountability, not the plan itself, decide whether category management sticks |
| A practitioner review describes a steady decline in organizations' ability to deploy category management effectively even as teams still claim to do it | O'Brien | Use it as a caution that the capability erodes quietly when the discipline is not maintained |
Sources span US federal guidance, a peer-reviewed portfolio study and practitioner commentary; their sectors and methods are not comparable.
Why does indirect spend resist a one-time fix?
Indirect spend covers what a company consumes to run itself, from IT and facilities to marketing, travel, professional services and MRO, and it is bought by many budget owners across many sites, so it fragments by default. US federal guidance defines category management as the practice of buying common goods and services as an enterprise, and its own maturity model opens at unaligned, decentralized spending. That is the state most indirect portfolios start in, which is why a single sourcing wave rarely holds, because the fragmentation regrows unless something keeps managing it.
How do you baseline spend that nobody fully owns?
The first move is a reconciled baseline rather than a strategy, since everything downstream depends on trustworthy numbers. Pull accounts payable, purchase-order history, corporate-card transactions, the vendor master and the active contract repository into one view, then clean and classify it so each dollar has a category and an owner. This is unglamorous spend-analysis work, and it is where you find out whether the categories you argue about later are real or imagined.
How should you segment without over-fitting the model?
Most teams reach for a purchasing-portfolio model, sorting categories by value and supply risk into the four Kraljic quadrants of non-critical, leverage, bottleneck and strategic, so the high-stakes categories get bespoke strategies and routine buys get catalogs. That instinct is sound, and the evidence sharpens it, because classifying 107 real sourcing projects on the Kraljic matrix, Hesping and Schiele found the full range of tactical sourcing levers is used in each of the four quadrants and argued for moving away from a strict adherence to a limited number of recommended tactics. Their projects were direct materials, so read it as a caution rather than a rule, and let a category's own specifics choose the tactics rather than the quadrant it happens to land in.
A five-move roadmap you can actually run
| Move | What you actually do | Who holds the decision | Review cadence |
|---|---|---|---|
| Baseline | Reconcile AP, POs, card spend, vendor master and contracts into one classified, owned view | Procurement operations, with finance signing off on the spend cut | Refresh quarterly |
| Segment and prioritize | Rank categories by value and supply risk, then pick the few worth a strategy this year | Category lead proposes; the CPO and business sponsor agree the shortlist | Reset annually |
| Control demand | Set specification standards, approval thresholds and catalog routing so unnecessary buying stops at the source | Budget owner, within policy set by procurement | Review each planning cycle |
| Source and contract | Run the sourcing event, negotiate and award against the category strategy | Category lead runs it; a named signatory owns the award | Per event |
| Govern and review | Track compliance to the plan, surface leakage and decide corrective action | Cross-functional council with clear decision rights | Quarterly category review |
Disclosed expert analysis: an authored sequence and decision-rights template, not a benchmark; the owners and cadences are illustrative and should be fitted to each organization's structure.
The third move is the one most often skipped, because it is internal and awkward: demand management. Federal guidance lists it as a named discipline, to implement demand management strategies to eliminate inefficient purchasing and consumption behaviors, and it works on the need itself rather than the unit price. Specification standards, approval thresholds and default catalogs shape demand before a supplier is ever involved, and in categories like marketing agency retainers or contingent labor a tighter brief usually saves more than the negotiation does.
Where do decision rights fail quietly?
This is where category strategies decay, and there is direct evidence of it. Auditing the Department of Veterans Affairs in 2025, the GAO found that category managers and their leads generally did not fulfill their responsibilities because senior leaders responsible for oversight did not take steps to ensure they did so. The plans were written and then drifted, because nobody senior was on the hook for whether category managers actually did the work, so the auditors went after the structure that let it drift.
Their recommendations were to establish performance requirements to hold category managers accountable for their responsibilities, set category-specific goals and make sure the officials get the training their roles need. The lesson generalizes well past the public sector, because a category plan needs a named owner, a cadence that puts them in front of results, and a council that can decide corrective action, or it reverts to unmanaged spend. A written purchasing policy only starts to matter once someone is named to enforce it and is asked about the results each quarter.
When does governance become the thing that stalls you?
There is a failure mode on the other side too, where governance turns into its own bureaucracy. Jonathan O'Brien, a veteran category-management practitioner and author, describes a steady decline in the ability of organisations to deploy Category Management effectively even as teams still say they do it. Read alongside the accountability evidence, the practical response is to keep the cadence light enough that people actually run it, because a short quarterly review that changes a decision or two gets you further than a framework that looks impressive in the deck and never gets opened again.
Frequently asked questions
What actually counts as indirect spend?
Indirect spend is everything a company buys to operate rather than to build its product, such as IT, facilities, marketing, travel, professional services and MRO, as opposed to the direct materials that go into what it sells. Because it is spread across many budget owners, it is usually where visibility is weakest.
How is category management different from strategic sourcing?
Strategic sourcing is an event, where you run a tender, negotiate and award. Category management is the standing discipline around it: baseline, segment, control demand and govern a category over years, with sourcing events as one recurring step inside it. The strategic sourcing method sits within the category plan.
What if a category doesn't fit one Kraljic quadrant cleanly?
That is common, and the evidence says not to force it. Studying real sourcing projects of direct materials, researchers found the full range of tactical sourcing levers is used in each of the four quadrants, and although that study was direct spend the caution carries: treat the matrix as a way to frame the conversation, then let the category's own specifics choose the tactics, whichever quadrant it nominally sits in.
How many categories should we start with?
Fewer than you want to. Prioritize the handful where value and supply risk are highest and you can realistically maintain a plan, track them with real procurement analytics, and leave routine tail spend on catalogs and automation until the discipline is proven.
Sources
- M-19-13: Category Management - Making Smarter Use of Common Contract Solutions and Practices — Office of Management and Budget, Executive Office of the President (OMB/OFPP), 2019. Historical evidence (official report): Official definition of category management, demand management as a named discipline, and the unaligned/decentralized starting tier.
- VA Acquisition: Steps Needed to Hold Category Management Officials Accountable (GAO-25-107398) — U.S. Government Accountability Office, 2025. Current empirical evidence (official report): Current evidence that decision rights and senior-leader accountability, not the plan itself, determine whether category management is sustained.
- Matching tactical sourcing levers with the Kraljic matrix: Empirical evidence on purchasing portfolios — Frank Henrik Hesping; Holger Schiele, Journal of Purchasing and Supply Management (Elsevier), 2016. Foundational evidence (peer reviewed journal): Peer-reviewed evidence that portfolio segmentation should structure debate rather than dictate a fixed set of tactics per quadrant.
- Category management entropy - Why procurement is failing to deliver the benefits it once did — Jonathan O'Brien, CPOstrategy, 2025. Contextual evidence (practitioner article): Contextual counterweight describing how category-management capability erodes when the discipline is not maintained.