Most organisations know they spend too much. Fewer know exactly where, with whom, and why. Without clear visibility into spending patterns, procurement teams operate in the dark -- unable to negotiate effectively, identify waste, or build category strategies grounded in evidence.

This is where spend cube analysis comes in. A spend cube is a structured method for visualising procurement data across multiple dimensions, revealing the patterns, anomalies, and opportunities that spreadsheets and instinct alone cannot uncover. Organisations that invest in comprehensive spend analysis typically deliver savings of 5-15% across analysed categories, and those with mature spend analytics programmes save two to three times more than those without.

Whether you are a procurement professional inheriting a new portfolio, a finance director questioning rising costs, or a business leader seeking better control over supplier expenditure, this guide will show you how spend cubes work, how to build one, and how to use the insights to drive measurable commercial improvement.

‍

What Is a Spend Cube?

A spend cube is a way to look at spend data projected as a multidimensional cube, typically across three core dimensions:

- Category (What are we buying?): The type of goods or services purchased, classified using a structured taxonomy

- Supplier (Who are we buying from?): The vendors providing those goods or services

- Business Unit / Cost Centre (Who is buying it?): The departments, teams, or locations making the purchases

These three dimensions create a matrix of data points representing your organisation's spending patterns. The term "cube" comes from the three-dimensional nature of the analysis, though modern spend analysis often extends to four or more dimensions -- adding time periods, contract status, payment terms, or geographic location.

The cube answers three fundamental questions simultaneously. Category analysis tells you what specific types of goods and services you are buying, per supplier and per business unit. Supplier analysis reveals the range of suppliers per category and per business unit. Business unit analysis shows which cost centres within the organisation are buying what and from where.

As GEP's Knowledge Bank puts it: "The aim is to show who bought what from where."

‍

Why Spend Cubes Matter: The Business Case

‍

Quantified Savings

Spend analysis typically delivers savings of 5-15% across analysed categories, depending on the maturity of the procurement function and the quality of existing supplier relationships. First-time analyses often reveal the most dramatic opportunities because they expose previously invisible spending patterns.

Maverick Spend Identification

Maverick spend -- purchases made outside of agreed contracts, preferred suppliers, or procurement policies -- is one of the most persistent sources of value leakage. Fragmented spend data blinds procurement teams to wasteful purchasing patterns, allowing maverick spending to proliferate unchecked.

A spend cube reveals: purchases from non-approved suppliers in categories where contracts already exist; price variations for identical items across different business units; purchases that bypass approved procurement channels; and tail spend that accumulates across many small, unmanaged transactions.

Tail Spend Visibility

Tail spend typically represents 20% of total procurement spend but involves 80% of suppliers. These small, frequent, unmanaged transactions collectively represent a significant source of savings. Spend cube analysis provides comprehensive visibility into indirect spending, effectively curbing rogue purchasing and enabling consolidation.

Strategic Foundation

Spend cube analysis serves as the diagnostic foundation of category management. It provides the data needed to prioritise categories for strategic sourcing, understand supplier market dynamics, identify consolidation opportunities, set savings targets based on actual spend data, and monitor category performance over time.

‍

How to Build a Spend Cube: A Five-Stage Process

Building a spend cube is not a technology exercise -- it is a data discipline. The process involves five key stages, and the quality of your output depends entirely on the rigour you apply to each.

‍

Stage 1: Data Collection

Gather spend data from all sources across the organisation. This typically includes ERP systems, accounts payable records, purchase orders, invoices, corporate purchasing cards, expense reports, and contract databases.

The challenge is that data often resides in multiple systems with inconsistent formats. A manufacturing firm may use SAP for procurement, Sage for accounts, and spreadsheets for ad hoc purchasing. Your spend cube is only as complete as your data collection.

Practical tip: Map every source of expenditure before you begin. Include off-system purchases, P-card transactions, and department-level spending that may not flow through central procurement.

Stage 2: Data Cleansing

Remove duplicates, correct errors, standardise formats, and reconcile discrepancies. This is typically the most time-consuming stage, often accounting for 60-70% of the total effort in a spend analysis project.

Common data quality issues include: inconsistent supplier names (the same supplier recorded under multiple variations), missing category codes, duplicate entries from different systems, incorrect currency conversions, and unclassified transactions.

Practical tip: Invest the time here. Attempting to analyse dirty data produces unreliable insights and undermines confidence in the entire exercise.

Stage 3: Data Classification

Assign each spend item to the appropriate category within your taxonomy. This is where structure becomes critical.

A spend taxonomy is the system of classification, from general to specific, of spending within your organisation. Taxonomies typically have three to four levels:

- Level 1 (Group): Broadest classification (e.g., "IT Services")

- Level 2 (Family): Sub-category (e.g., "Software")

- Level 3 (Category): More specific (e.g., "Enterprise Resource Planning")

- Level 4 (Commodity): Most granular (e.g., "SAP Licensing")

Common taxonomy standards include UNSPSC (United Nations Standard Products and Services Code) and eClass, which is widely used in manufacturing. Many organisations develop custom taxonomies aligned to their industry and procurement needs.

AI-powered classification engines can categorise 60-70% of data automatically on the first pass, with accuracy improving after human review and correction. This can reduce classification timescales from months to days.

Practical tip: Do not over-engineer your taxonomy. Levels beyond four tend to become repetitively granular and difficult to categorise consistently. Three to four levels is optimal for most organisations.

Stage 4: Data Analysis

With clean, classified data, you can now apply analytical methods to identify patterns, trends, anomalies, and opportunities. Key areas of analysis include:

- Supplier consolidation: How many suppliers are providing similar products or services? Can you consolidate to leverage volume discounts?

- Contract compliance: Are purchases being directed to contracted suppliers at agreed rates, or is off-contract purchasing eroding negotiated savings?

- Price benchmarking: What price variations exist for identical items across different business units or time periods?

- Demand management: Are all purchases necessary? Can specifications be standardised to reduce variety and cost?

- Spend trends: How is spending changing over time? Are categories growing faster than expected?

The Kraljic matrix can be overlaid onto spend cube data to classify categories by supply risk and profit impact, helping prioritise which categories warrant strategic sourcing initiatives versus routine purchasing approaches.

Stage 5: Reporting and Action

Present findings in actionable formats and develop action plans for each priority category. This stage transforms analysis into commercial outcomes.

Modern spend analysis platforms integrate with data visualisation tools like Power BI and Tableau, enabling interactive spend cube visualisations that allow users to drill down into specific dimensions, filter by time period, and explore patterns dynamically.

Practical tip: Avoid analysis paralysis. The purpose of a spend cube is to drive action, not to produce the perfect dataset. Identify quick wins early and pursue them while continuing to refine the analysis.

‍

Real-World Applications: What Spend Cubes Reveal

‍

Case Study: Safety Equipment Consolidation

A manufacturing firm used spend cube analysis and identified that five departments were buying the same safety equipment from different suppliers at different prices. By standardising to one vendor through a consolidated contract, the firm saved 15% on safety equipment spend and reduced lead times significantly.

Case Study: Software Licence Maverick Spend

A technology company used spend analysis to flag an increase in tail spend for software licences. Investigation revealed that engineers were bypassing the standard procurement process and purchasing licences independently. By enforcing the procurement policy and consolidating software purchasing, the company reduced per-licence costs by 22%.

Case Study: Supplier Rationalisation Across Sites

A multi-site organisation discovered through spend cube analysis that it was using 47 different suppliers for office supplies across 12 sites, with prices varying by up to 35% for identical items. Consolidating to three preferred suppliers reduced office supplies spend by 18% while improving service levels and simplifying administration.

‍

Common Pitfalls to Avoid

Spend analysis projects fail more often from poor execution than from poor intent. The most common pitfalls include:

- Incomplete data capture: Missing data sources or off-system purchases create blind spots that undermine the analysis.

- Poor data quality: Inconsistent supplier names, missing category codes, and duplicate entries produce unreliable results.

- Overly complex taxonomy: Too many levels make classification inconsistent and the analysis difficult to interpret.

- Treating it as a one-off exercise: Spend analysis should be ongoing, not a single project. Markets change, suppliers change, and spending patterns shift continuously.

- Lack of executive sponsorship: Findings need authority to drive change. Without senior support, even the most insightful analysis will gather dust.

- Analysis paralysis: Spending too long perfecting the data before taking action. Pursue quick wins while refining the analysis in parallel.

- Ignoring tail spend: Small transactions that collectively represent significant value. The 80/20 rule means your tail spend involves the vast majority of your suppliers.

‍

AI and Technology in Modern Spend Analysis

Technology has transformed spend analysis from a manual, spreadsheet-based exercise into a data-driven capability. AI-powered classification engines can process and categorise spend data in days rather than months, and machine learning algorithms improve accuracy with each iteration.

As Spend Matters notes in their 2026 category management insights: "AI helps category managers move beyond spend analysis into scenario planning, predictive sourcing, and supplier risk anticipation. However, successful organisations invest in continuously maintaining clean, harmonised data across their systems."

Modern platforms offer: automated data extraction from multiple source systems; AI-powered classification and categorisation; interactive dashboards and visualisations through Power BI or Tableau integration; anomaly detection and alerting; predictive analytics for spend forecasting; and integration with e-procurement and contract management systems.

Technology accelerates the process, but it does not eliminate the need for human expertise in interpreting results and designing category strategies.

‍

How Athena Commercial Can Help

Building a spend cube for the first time can be daunting, particularly when data is fragmented across multiple systems and there is no established taxonomy. Athena Commercial helps organisations consolidate their spend data into comprehensible analyses, identify savings opportunities, and develop category strategies grounded in evidence.

Whether you need a one-off spend analysis to establish a baseline or ongoing support to build a mature spend analytics function, our procurement consultants bring the expertise and objectivity that internal teams sometimes lack. A fresh pair of eyes on your spending data can reveal opportunities that have been hiding in plain sight.

To discuss how spend cube analysis could benefit your organisation, visit www.athena-commercial.co.uk (add link - https://www.athena-commercial.co.uk) or contact our team directly.

‍

‍