Spend CubeVisibility of Procurement Volume by Supplier, Category and Time

Distinction from the classic spend report

A conventional spend report usually delivers static lists, such as the top 20 suppliers by revenue. The spend cube goes further by allowing arbitrary slices through the data: for example, it can be determined which category generated which volume with which suppliers in which quarter. This flexibility is a prerequisite for well-founded category strategies and effective bottleneck analysis, in which critical procurement items are identified.

Legal and data protection considerations

Since the spend cube can contain personal data such as requisitioner or buyer assignments, the requirements of the GDPR apply. In practice, it is advisable to anonymise or pseudonymise personal fields as long as they are not required for the analysis. Whether scorecard data or supplier evaluations may be integrated into the cube should also be clarified contractually with suppliers; this also concerns regulations such as the German Supply Chain Due Diligence Act when sustainability data is incorporated. Legal review is advisable in individual cases.

The three dimensions in detail

The supplier dimension ideally uses a cleansed supplier master list in which corporate group relationships are visible via parent company information. The category dimension is based on a commodity grouping, frequently along standardised classification systems, supplemented by company-specific category definitions. The time dimension is typically structured by months or quarters in order to capture seasonality and purchase price escalation rates.

• Supplier dimension: master data, corporate group assignment, region, supplier portfolio membership
• Category dimension: commodity groups, material numbers, classification codes
• Time dimension: month, quarter, financial year, comparison periods

Data quality and duplicate detection

In practice, raw data frequently contains duplicates, inconsistent spellings of supplier names, missing material assignments and empty category fields. Systematic duplicate detection and standardisation of supplier master data is therefore usually the first project step. Typical metrics are the share of categorised spend and the hit rate of supplier assignment. Initial classification rates are often below 70 per cent, which can be raised to well over 90 per cent through cleansing.

Match-and-merge rules and data enrichment

To consolidate and evaluate order and invoice data, match-and-merge rules are applied, as known from invoice processing. Using these rules, line items are merged, duplicates removed and open orders linked with goods receipt and invoice data. In addition, external data sources such as customs tariff numbers or supplier self-disclosure data can be used for enrichment.

Five-step procedure model

First, all relevant data sources are identified and extracted – alongside ERP systems, frequently also card payments, freight costs or data from bill-of-lading-based import settlements. Cleansing follows, including duplicate detection, and then assignment to consistent categories. The fourth stage is modelling as a cube, and the last is establishing regular reporting.

• 1. Inventory and extract data sources
• 2. Cleanse: duplicates, spellings, empty fields
• 3. Categorise: automated assignment plus expert review
• 4. Model: cube structure with supplier, category, time
• 5. Operationalise: roll-out, training, regular updates

Organisational anchoring: category governance

A spend cube only has an impact if its results are embedded in decision-making processes. As part of effective category governance, owners are defined for each category who derive category strategies based on cube data, prepare tendering processes and monitor contract renewal dates. Without this governance, the cube risks becoming a mere reporting artefact.

Integrated budgeting

The spend cube links sensibly with budgeting: actual volumes from the cube can be compared with planned values, enabling variance analyses at category and supplier level. In this way, procurement recognises early on whether price increases, volume changes or substitute purchases – for example via an emergency purchase – are straining the budget.

Classic spend metrics

Standard analyses include spend by category and supplier, supplier concentration (often expressed as the share of the top 10 suppliers of total volume), price developments as purchase price escalation rates and the development of maverick buying, i.e. spend outside agreed processes. In addition, contract coverage rates can be determined: what share of volume is handled under framework agreements.

• Spend per category and supplier (current vs. prior period)
• Supplier concentration: top-10 share of total volume
• Purchase price escalation rate per commodity group
• Contract coverage: contractually covered volume share
• Share of maverick buying and emergency purchases

Analytical extensions

Beyond description, the cube enables supply chain analytics: merging spend data with supplier risk scores, quality metrics or availability data from the available-to-promise (ATP) check. This allows categories to be identified where high volume meets few suppliers and safety stock planning is critical at the same time. The preparation of tendering processes also benefits: supplier shortlists and volume aggregates can be derived directly from the cube.

Typical pitfalls

Common sources of error are inconsistent supplier master data, multiple category logics running in parallel and incomplete recording of additional costs such as freight or packaging. An overly detailed category scheme can also be problematic: with hundreds of categories, analyses lose their explanatory power, while with too few they remain superficial. In practice, hierarchies with two to three levels have proven effective.

• Inconsistent supplier identification (corporate group vs. individual site)
• Lack of cost transparency for ancillary services
• Outdated data without a defined update cycle
• Low user acceptance due to complex tools

Controls and quality management

Effective controls include defined data quality metrics, sample checks of categorisation and a clearing process for structural changes. Regular reconciliations with financial accounting ensure that spend analyses and income statement positions are consistent. A documented update process with clear ownership prevents the cube from falling asleep after the initial build.

Risks of misuse and misinterpretation

Spend data can be misinterpreted if volume, price and currency effects are mixed. A sound variance analysis separates these drivers before conclusions are drawn. Cubes should also not be used as a surveillance instrument against individual buyers, as this permanently damages acceptance.

Automation and AI

Modern approaches automate categorisation using machine learning and detect supplier duplicates almost fully automatically. Language-based analytics enable buyers to ask ad-hoc questions of the spend data without having to build cube queries themselves. This significantly lowers the entry barrier and broadens the user base across the buying centre.

Extension to sustainability and risk

Increasingly, emissions data, supply chain data and information from supplier self-disclosures are linked with spend data, for example to estimate the CO2 footprint per category or to support compliance requirements such as those arising from the German Supply Chain Due Diligence Act from a data perspective. Certification documents such as the Certificate of Conformity can also be evaluated by supplier. Uniform frameworks such as the GRI Standards serve as orientation for sustainability reporting.

Integration into buying centres and marketplaces

In B2B marketplaces and digital procurement platforms, spend data is increasingly flowing directly into operational processes: from demand identification through the tendering process to supplier development and supplier innovation. The cube thus becomes a steering instrument for the entire procurement life cycle rather than just an analysis tool.