The case for a procurement-led category taxonomy

UNSPSC codes are everywhere in procurement. They are embedded in ERP systems, required by public sector frameworks, and used as the default classification layer in spend analytics platforms. Most organisations that do any kind of structured spend analysis will have encountered them, and many will have built their category reporting directly on top of them.

The problem is not that UNSPSC codes are wrong. It is that they were designed to solve a data standardisation problem, not a procurement problem. And when organisations try to use them as the foundation for procurement decision-making: for sourcing, category strategy, supplier management and market engagement, the cracks start to show.

This article sets out why UNSPSC codes, used in isolation, are an insufficient basis for procurement, what goes wrong when organisations rely on them too heavily, and what a more effective approach looks like in practice.

What UNSPSC Was Designed to Do

It is worth being clear about the original purpose of UNSPSC, because understanding that purpose explains most of its limitations in a procurement context.

The UNSPSC was created in 1998 to solve a specific problem: different organisations were using different definitions for the same products and services, which made electronic data exchange between buyers and suppliers unreliable. The code system was designed to provide a universal reference, a shared vocabulary that could sit underneath procurement transactions and allow systems to communicate consistently.

That is a data infrastructure problem, and UNSPSC solves it reasonably well. It provides a structured, hierarchical taxonomy that covers a broad range of goods and services, is maintained over time, and is free to use. For spend classification, turning raw transaction data into something that can be aggregated and compared, it serves a legitimate purpose.

What it was never designed to do is tell a procurement team how to organise their categories, which suppliers to engage, how a supply market is structured, or where the savings opportunities lie. Those are procurement questions, and UNSPSC does not answer them.

Where UNSPSC Breaks Down for Procurement

In practice, organisations that use UNSPSC as their primary category structure encounter the same set of problems repeatedly. These are not edge cases, but structural limitations of the taxonomy itself.

It reflects product logic, not market logic

The UNSPSC hierarchy is built around what something is, not how it is bought or who sells it. That distinction matters enormously in procurement.

Supply markets are not organised the way UNSPSC is organised. A supplier providing facilities management services might span cleaning, security, catering and maintenance, activities that sit across multiple UNSPSC segments. A technology vendor might provide hardware, software licences and professional services, again, spread across different parts of the taxonomy. If your category structure mirrors UNSPSC, you will find yourself running separate sourcing exercises for things that the market treats as a single proposition, or grouping together things that have entirely different supply chains and competitive dynamics.

Effective category management requires categories that reflect how markets work, not how a classification system is structured. Those two things are often quite different.

It is too broad for strategic sourcing, too narrow for nuanced analysis

UNSPSC operates at a level of granularity that rarely aligns with what a category manager actually needs. At the top of the hierarchy, segments are so broad as to be strategically meaningless; “Information Technology Broadcasting and Telecommunications” is not a category anyone can meaningfully source. At the commodity level, the codes become highly specific in ways that fragment spend that should be managed together.

This creates a consistent problem: category managers either end up with spend data that is too aggregated to act on, or too fragmented to see the full picture. Neither is useful for building a sourcing strategy, conducting a market review, or developing a supplier relationship plan.

Digital and technology services are poorly served

This is one of the most acute limitations for modern procurement functions. The way organisations buy technology has changed fundamentally over the past decade, away from discrete hardware and software purchases, and toward cloud services, SaaS platforms, managed services, and complex multi-vendor arrangements. The commercial models are different, the suppliers are different, and the procurement approach required is different.

UNSPSC has not kept pace with this shift. Digital services are often lumped under broad IT codes that make no meaningful distinction between a SaaS productivity tool, a bespoke software development engagement, and a managed security operations service. These are not the same category. They have different markets, different pricing models, different risk profiles, and different negotiation levers. Treating them as equivalent because they share a four-digit UNSPSC family code produces category data that is structurally misleading.

It does not translate to supplier strategy

A coherent supplier strategy requires understanding which suppliers are critical, which are interchangeable, which present concentration risk, and which represent partnership opportunities. That understanding has to be built around commercial relationships and supply market realities, not around product codes.

An organisation might have a single strategic supplier that appears across a dozen different UNSPSC codes. A UNSPSC-based view of that relationship will fragment it across categories and obscure the true level of dependency. Conversely, two suppliers might share the same UNSPSC code but operate in entirely different niches with no competitive overlap. The taxonomy provides no way to distinguish between them.

The Trap Organisations Fall Into

The issue is rarely that organisations make a deliberate decision to run their procurement off UNSPSC codes. It happens gradually, and it happens because UNSPSC is the path of least resistance.

A spend analysis is commissioned. The platform classifies transactions to UNSPSC. The output is presented as a category view of spend. Category managers start using it because it is what exists. Over time, the UNSPSC structure gets embedded in reporting, in budget structures, and sometimes in the operating model itself. By the time anyone questions whether the taxonomy is fit for purpose, it has become the de facto category framework, even though it was never designed to be one.

The consequences show up in a few predictable ways:

• Sourcing exercises are scoped incorrectly: Too narrow because spend is fragmented across codes, or too broad because a segment contains categories with nothing commercially in common.
• Market engagement is misaligned: Suppliers are approached using category definitions that do not reflect how they organise their own businesses or capabilities.
• Savings opportunities are missed: Consolidation potential is hidden because related spend sits in different parts of the taxonomy and is never viewed together.
• Category strategies lack depth: Because the underlying data is structured around product descriptions rather than supply market dynamics, strategic analysis is harder to build and harder to defend.
• Digital spend is effectively unmanaged: Because the taxonomy cannot represent it accurately, technology categories become a catch-all where spend accumulates without meaningful structure.

What a Procurement-Led Taxonomy Looks Like

The alternative is not to abandon spend classification. It is to be clear about the difference between a classification taxonomy and a procurement taxonomy, and to build both, deliberately, and for different purposes.

A classification taxonomy, like UNSPSC, exists to standardise data. It answers the question: what was bought? It is the right tool for spend analysis, benchmarking, and system interoperability.

A procurement taxonomy exists to organise work. It answers the question: how should we manage this spend? It is the right tool for category strategy, sourcing planning, supplier management and procurement operating models.

In practice, building a procurement taxonomy means:

1. Defining categories around supply markets, not product descriptions. A category should correspond to a coherent group of suppliers competing for the same business. If a sourcing exercise would naturally go to the same shortlist of suppliers, it is probably one category. If it would go to entirely different markets, it should be split, regardless of how UNSPSC codes them.

2. Mapping UNSPSC codes to internal categories, not the other way around. UNSPSC classification remains useful for data standardisation and external benchmarking. The right approach is to run UNSPSC classification on transactional data, and then map those codes to your internal procurement taxonomy, so that spend data can be viewed through a procurement lens without losing the benefits of the standard.

In one engagement with a major transport and logistics operator, UNSPSC classification had been applied across the spend base and the codes were technically correct, but when the procurement team tried to use that data to plan sourcing activity, it did not map to anything they recognised. Operational spend was distributed across segments in ways that reflected individual transactions rather than category relationships, obscuring the supplier consolidation and contract planning opportunities that mattered most. The solution was to build a parallel internal category taxonomy structured around how the organisation actually sourced and managed spend, and map the UNSPSC codes to it. The underlying classification remained intact for reporting purposes; procurement decisions could be driven by a structure that reflected supply market realities.

Source: Procurato (2026)

3. Reflecting how the organisation actually buys. A large enterprise buying complex IT services has different category needs to a public sector body buying construction works. The taxonomy should reflect the organisation’s spend profile, supplier relationships and strategic priorities, not a universal standard designed for every organisation simultaneously.

4. Building in commercial and strategic context. Categories should carry information about strategic importance, supply market risk, current contract coverage, and savings potential. This is the layer that turns classification into decision support.

5. Treating the taxonomy as a live asset. Markets change, business priorities shift, and new categories of spend emerge, particularly in technology. A procurement taxonomy needs to be reviewed and updated regularly, not set once and left to calcify.

Summary

For organisations that have already built their category reporting and operating model around UNSPSC, change does not need to be disruptive. The starting point is diagnostic rather than structural, understanding where the taxonomy is genuinely serving procurement and where it is constraining it.

UNSPSC codes have a legitimate and useful role in procurement data management, providing a standardised classification layer that supports spend visibility, benchmarking, and system interoperability. For those purposes, they work. Where they fall short is when organisations ask them to do something they were never designed for: driving category strategy, structuring sourcing activity, informing supplier decisions, or representing the nuances of modern digital and services procurement.

The procurement teams that get the most value from their spend data are those that understand the difference between classifying spend and managing it, and that have built the infrastructure to support both, rather than conflating them into one.