Spend Analysis: A Practical How-To Guide
Most procurement organisations say they understand their spend. Few actually do. Spend analysis — the systematic collection, cleansing, classification, and analysis of expenditure data — is the discipline that closes that gap. Done well, it reveals consolidation opportunities, unmasks maverick spend, quantifies supplier risk, and gives procurement leaders the evidence they need to make the case for strategic sourcing initiatives. This guide walks you through the four steps and the common pitfalls at each one.
Why Spend Analysis Matters
Procurement strategy without spend data is guesswork. Organisations that do not know who they are buying from, what they are buying, or how much they are spending in each category cannot negotiate intelligently, cannot consolidate their supplier base, and cannot identify where maverick purchasing is undermining contracts already in place. Spend analysis is not a one-time exercise — it is a standing discipline that should be refreshed at least annually and ideally quarterly.
The business case is straightforward. A typical organisation can identify cost-reduction opportunities worth two to five per cent of addressable spend in the first analysis cycle alone. For a public-sector body with $50 million in annual procurement, that is $1–2.5 million in potential savings — before any sourcing work begins.
The Four Steps of Spend Analysis
Collect the data. Pull expenditure data from every source that records a payment: ERP systems, procurement cards, accounts payable ledgers, purchase order systems, and — in the public sector — grant and transfer payment records. Completeness matters more than cleanliness at this stage. Aim to capture at least 80 per cent of total organisational spend. Gaps in data are findings in themselves.
Cleanse and normalise. Raw spend data is almost always messy. The same supplier appears under a dozen different names. Amounts are recorded in different currencies, with and without taxes, across different fiscal periods. Cleansing means standardising supplier names (often using a supplier master or a matching algorithm), normalising currencies and periods, removing inter-company transactions, and flagging duplicate payments. This step is time-consuming but non-negotiable — dirty data produces misleading analysis.
Classify the spend. Categorisation is where spend analysis moves from data management to strategic insight. Assign each line of spend to a category hierarchy — typically two to three levels deep. At the top level, most organisations separate direct spend (materials and services that go into the product or service being delivered) from indirect spend (everything else: facilities, IT, professional services, travel). Within indirect, further classification — by commodity, by business unit, by supplier — enables the strategic questions to be asked.
Analyse and act. With clean, classified data in hand, the analysis can begin. Look for consolidation opportunities: categories where spend is fragmented across many suppliers but could be aggregated with one or two preferred suppliers. Identify preferred supplier leverage: where you already have a contract but purchasing is still happening outside it. Quantify maverick spend: the proportion of category spend that bypasses established contracts and purchasing channels. Prioritise categories by spend volume, supplier risk, and ease of consolidation, and sequence your sourcing activities accordingly.
Spend Categories and Classification
The direct vs. indirect distinction is a starting point, not an end point. Within direct spend, categories might include raw materials, sub-assemblies, and contract manufacturing. Within indirect, facilities management, IT hardware and software, temporary labour, marketing services, and professional services each warrant their own category strategies.
Strategic vs. tactical is a useful secondary lens. Strategic spend categories are those where supply risk is high, switching costs are significant, or the category directly affects service quality. Tactical categories are lower-risk, more commoditised, and amenable to reverse auctions or catalogue procurement. Category classification drives sourcing strategy: strategic categories need relationship management and long-term contracts; tactical categories need efficiency and price competition.
Data Quality Challenges and How to Address Them
The most common data quality challenges in spend analysis are: incomplete coverage (some payment systems are not connected to the analysis), inconsistent supplier naming (the matching problem), missing or incorrect category codes, and currency/period mismatches. Practical approaches include: building a supplier master file and updating it as part of the AP process; adopting a standard commodity code taxonomy (UNSPSC is widely used in the public sector); running duplicate-payment detection as part of the cleansing routine; and establishing a data governance owner for spend analytics.
Do not let perfect be the enemy of useful. An 80-per-cent-complete spend cube with good category classification will surface more actionable insight than a 100-per-cent-complete dataset that has never been analysed. Start with what you have, document the gaps, and improve coverage incrementally.
XNM Consulting helps public-sector and Indigenous organisations build spend visibility and sourcing strategies that deliver measurable value. Learn more about our procurement and sourcing services.