Every procurement team knows they should be doing spend analysis. Far fewer are doing it effectively. The gap between "we have a spend cube" and "we're making better sourcing decisions because of spend data" is where most organizations get stuck.
This guide walks through how to actually execute spend analysis within a procurement function — the process, the common pitfalls, and the practices that separate organizations getting real value from those generating reports nobody reads. If you're looking for a foundational overview, start with our complete guide to spend analysis. This article is for teams ready to put it into practice.
What is spend analysis in procurement?
Spend analysis in procurement is the process of collecting, cleansing, classifying, and analyzing an organization's purchasing data to understand where money is going, identify savings opportunities, and make better sourcing decisions. It turns raw transactional data — purchase orders, invoices, P-card transactions, contract records — into actionable intelligence that drives procurement strategy.
In practice, spend analysis in procurement answers questions like:
- How much are we spending with each supplier, and is that concentration intentional?
- Which categories have fragmented spend across too many vendors?
- Where are we buying off-contract, and what's the cost premium?
- Which supplier relationships should be renegotiated based on volume trends?
- Are we paying different prices for the same goods across business units?
The distinction between "spend analysis" as a concept and spend analysis as practiced within procurement is important. Finance teams do spend analysis too — but their lens is budget variance and cost accounting. Procurement spend analysis is specifically oriented toward sourcing decisions, supplier management, and category strategy. The same data gets interpreted through a fundamentally different lens.
Spend analysis vs. spend analytics vs. spend reporting
These terms get used interchangeably, but they represent different levels of analytical maturity — from backward-looking reporting to forward-looking automation:
Understanding where your organization sits on this spectrum helps set realistic expectations:
Spend reporting is backward-looking: "Here's what we spent last quarter." It summarizes historical data in dashboards and charts. Useful but passive — it tells you what happened without suggesting what to do about it.
Spend analysis adds interpretation: "Here's what we spent, and here's what it means for our sourcing strategy." It involves classification, benchmarking, and identification of opportunities. This is where most procurement teams should be operating.
Spend analytics adds prediction and automation: "Here's what's likely to happen next, and here's what you should do now." It uses machine learning and AI to identify patterns, forecast trends, and proactively surface recommendations. This is where AI-driven procurement platforms are pushing the frontier.
Why spend analysis matters for procurement teams
The business case for spend analysis in procurement is well established, but it's worth quantifying. Organizations that invest in structured spend analysis consistently report benefits across four dimensions:
Savings identification and realization. The most direct ROI. Spend analysis typically reveals 5-15% savings opportunities through supplier consolidation, contract compliance enforcement, demand management, and rate benchmarking. For a $500M spend organization, even the low end of that range represents $25M in addressable savings.
Supplier rationalization. Most enterprises have far more suppliers than they need. Spend analysis reveals the long tail — the hundreds of suppliers representing a small fraction of total spend but a disproportionate share of procurement workload, risk exposure, and missed volume leverage. Rationalizing the tail frees capacity and improves negotiating position with strategic suppliers.
Contract compliance. Spend analysis surfaces maverick buying — purchases made outside negotiated contracts, often at higher prices. Research consistently shows that 20-40% of enterprise spend occurs off-contract. Identifying and redirecting this spend to contracted suppliers is one of the fastest paths to procurement savings.
Category strategy development. Without spend analysis, category strategies are built on incomplete information and category manager intuition. With it, strategies are grounded in data: spend trends by supplier, price variance across business units, demand patterns over time, and market benchmarks. Better data produces better strategies, which produce better outcomes.
Risk visibility. Spend analysis reveals concentration risk that isn't visible at the category level. You might have a well-diversified IT services category — until you discover that three of your five vendors all subcontract to the same offshore provider. This kind of supply chain risk only becomes visible through deep spend data analysis.
The spend analysis process: 6 steps
Effective spend analysis follows a repeatable process. The steps are straightforward in concept but challenging in execution — particularly the data quality steps that most organizations underinvest in.
Step 1: Define scope and objectives
Before touching data, clarify what you're trying to achieve. A spend analysis for supplier rationalization requires different data granularity than one aimed at identifying maverick spend. A category-specific deep dive needs different source systems than an enterprise-wide baseline.
Define the time period (typically 12-24 months of history), the business units in scope, the categories to analyze, and the specific questions you want answered. This scoping step prevents the most common failure mode: boiling the ocean with data and producing analysis that's comprehensive but not actionable.
Step 2: Collect and consolidate data
Gather procurement data from all relevant source systems: ERP platforms (SAP, Oracle, Workday), P-card systems, accounts payable, contract management tools, and any departmental purchasing systems operating outside the main procurement platform.
This is typically the most painful step. Enterprise procurement data lives in multiple systems with different formats, different vendor naming conventions, and different levels of completeness. A single supplier might appear as "IBM," "IBM Corp," "IBM Corporation," "International Business Machines," and "IBM Global Services" across five systems.
The goal is a single consolidated dataset with every transaction mapped to: supplier, category, business unit, cost center, date, amount, and contract (if applicable).
Step 3: Cleanse and normalize
Raw procurement data is messy. Vendor names are inconsistent, categories are missing or wrong, and duplicate records are common. Data cleansing involves:
Vendor normalization — resolving the "IBM" problem described above. Every variation of a supplier name gets mapped to a single canonical entity. For organizations with thousands of suppliers, this step alone can take weeks if done manually.
Category classification — mapping every transaction to a standardized taxonomy (UNSPSC, eClass, or a custom taxonomy). This is where spend analysis in procurement gets its analytical power — you can't identify savings by category if transactions aren't consistently classified.
De-duplication and error correction — removing duplicate records, correcting obvious data entry errors, and flagging anomalous transactions for review.
This is also where automation provides the most leverage. Manual classification of millions of transactions is prohibitively slow and error-prone. AI-powered spend classification tools can process this data in hours rather than weeks, with higher consistency than manual approaches.
Step 4: Analyze and segment
With clean, classified data, the actual analysis begins. Key analytical approaches include:
Spend by supplier — rank suppliers by total spend, identify your top 20 (who typically represent 80% of total spend), and flag opportunities for consolidation in the long tail.
Spend by category — understand where money is going at the category and sub-category level. Identify categories with high fragmentation (many small suppliers) or high concentration (excessive dependence on one vendor).
Spend by business unit — compare purchasing patterns across divisions. Price variance for the same goods across business units is one of the most common — and most actionable — findings in procurement spend analysis.
Contract coverage analysis — map spend against active contracts to quantify off-contract (maverick) purchasing. Calculate the cost premium your organization is paying by buying outside negotiated terms.
Trend analysis — track spend trajectories over time by supplier, category, and business unit. Rising spend in a category might signal demand growth (requiring a new sourcing strategy) or price creep (requiring renegotiation).
Step 5: Identify opportunities and prioritize
Analysis produces a list of opportunities. The challenge is prioritization. Not every finding warrants action — the cost of pursuing a savings opportunity must be weighed against the expected benefit and the organizational effort required.
Build a prioritized action list based on estimated savings value, implementation complexity, and strategic alignment. Quick wins (contract compliance enforcement, obvious consolidation opportunities) should be executed first to build organizational momentum and credibility for the spend analysis program.
Step 6: Operationalize and repeat
Spend analysis is not a one-time project. The organizations that get the most value treat it as a continuous capability — refreshing data regularly, monitoring savings realization, and using updated analysis to inform each sourcing cycle.
Quarterly refresh cycles are a reasonable starting point. As your program matures and you invest in automated spend analytics, you can move toward monthly or even real-time analysis that continuously surfaces opportunities as spending patterns change.
Spend analysis best practices
Based on a decade of supporting enterprise procurement teams at Suplari, these are the practices that consistently separate high-performing spend analysis programs from those that underdeliver:
Start with good-enough data, not perfect data
The most common reason spend analysis programs stall is the pursuit of perfect data. Don't wait until every transaction is flawlessly classified. Start with 80% accuracy, generate insights, deliver value, and improve data quality iteratively. The insights from imperfect data are infinitely more valuable than the insights from a perfect dataset that never gets built.
Align analysis to sourcing calendar
Time your deep-dive category analyses to feed directly into upcoming sourcing events. Analysis that lands on a category manager's desk six months before their contract renews gets used. Analysis that arrives two months after they've already signed an extension gets filed.
Invest disproportionately in data classification
The quality of your spend analysis is bounded by the quality of your classification. Organizations that invest in robust taxonomies, automated classification tools, and ongoing data governance consistently produce more actionable analysis than those that treat classification as a one-time cleanup exercise.
Make insights accessible, not just accurate
A brilliant analysis that lives in a spreadsheet on an analyst's laptop has zero organizational impact. Build dashboards that category managers, sourcing leads, and procurement leadership can self-serve. Package insights with clear recommended actions — not just "here's what the data says" but "here's what you should do about it."
Measure savings realization, not just identification
Identifying a $2M savings opportunity in your MRO category is step one. Capturing that savings through a sourcing event, negotiating new terms, and ensuring compliance with those terms is where value is actually created. Track the full funnel: identified → validated → in progress → realized.
Build feedback loops with category managers
Category managers are the ultimate consumers of spend analysis. Their feedback — "this category split doesn't match how we actually buy" or "these two suppliers are actually the same parent company" — is essential for improving data quality and analytical relevance. Build structured feedback mechanisms, not just ad hoc corrections.
Common spend analysis mistakes to avoid
Analyzing spend in isolation from market context. Your spend data tells you what you're paying. Market data tells you what you should be paying. Without benchmarking against market rates, you can't distinguish competitive pricing from overpayment. Always contextualize internal spend data with external benchmarks.
Ignoring indirect and tail spend. Organizations tend to focus spend analysis on their top categories and strategic suppliers. But tail spend — the thousands of small transactions with hundreds of minor suppliers — often represents 20-30% of total spend and is where procurement has the least control. This is where the biggest per-unit savings opportunities often hide.
Treating spend analysis as an IT project. Spend analysis is a procurement capability that uses technology, not a technology project that procurement sponsors. When IT leads the initiative, the result is often technically impressive infrastructure that doesn't produce the specific insights procurement teams need to make better sourcing decisions.
Running analysis too infrequently. Annual spend analysis is better than nothing but misses dynamic opportunities. Supplier pricing creep, shifting demand patterns, and market volatility all happen continuously. Match your analysis cadence to your sourcing cadence — at minimum, refresh quarterly.
Failing to act on findings. The most expensive mistake. Spend analysis has zero value until someone changes a sourcing decision based on the insights. If your organization produces analysis reports that nobody acts on, the problem isn't the analysis — it's the connection between insight and action.
Spend analysis examples: How Suplari customers put it into practice
The process above is straightforward in theory. Here's how three Suplari customers applied spend analysis in procurement to drive real results:
BT Group: From analytics to action at telecom scale
BT Sourced — the procurement arm of British Telecom and one of the largest telecom procurement operations globally — needed spend analysis that went beyond static dashboards. As Diarmuid O'Donoghue, Head of Digital Procurement at BT Sourced, put it: "We didn't want just a general spend platform that relayed spend information back to us. We really wanted something different."
The challenge was typical of large enterprises: thousands of users across the organization needed access to spend insights, but traditional tools required analyst expertise to generate reports. Category managers couldn't self-serve.
Using Suplari, BT Sourced transformed their spend analysis process. "In essentially two clicks, your category manager can identify high-growth suppliers in a category or pinpoint POs raised against suppliers without a contract," O'Donoghue explains. The tool became embedded in daily operations rather than reserved for periodic analysis cycles: "It's not just a tool. It's basically part of our procurement team's daily life."
Key takeaway: Spend analysis in procurement delivers the most value when it's accessible to category managers and sourcing leads — not locked behind analyst-only tools. Read the full BT case study →
Nordstrom: Building spend visibility from scratch
When Karoline Dygas, a Global Supply Chain Executive with over 20 years of retail procurement experience (including Walgreens and Starbucks), joined Nordstrom, she faced a foundational challenge: "In order to make change, you actually need to see what is going on."
Nordstrom needed a comprehensive baseline of their procurement spend — who the suppliers were, what was being spent with them, and which business units were engaging which vendors. This is the classic Step 1-3 work of spend analysis (map, collect, classify), but at enterprise retail scale with fragmented data across multiple systems.
"Suplari gave us the tools to truly understand our spend — who the suppliers are, what we're spending with them, and who within the company is engaging with them," Dygas says. The data quality focus paid off: "You have to make sure your data is clean, readily available, consistent with what you see. And I think that Suplari does a really good job with how the data is represented."
From that visibility foundation, Nordstrom could build sourcing strategies grounded in data rather than institutional memory. "Building relationships is all about trust. And to have that trust, you have to have factual, repeatable, succinct, and meaningful information that you can share — and even spark discussion."
Key takeaway: You can't run effective procurement spend analysis without clean, reliable data. Invest in the data foundation first — the strategic insights follow. Read the full Nordstrom case study →
MediaNews Group: Uncovering cross-property price discrepancies
MediaNews Group — publisher of the Chicago Tribune, NY Daily News, Denver Post, and San Jose Mercury News — faced a spend analysis challenge unique to multi-property organizations: inconsistent pricing for identical goods and services across locations.
Jeff Ball, Head of Procurement, uses Suplari as the foundation for his sourcing approach: "When you think about building a comprehensive baseline of spend in your organization and being able to slice it and dice it very easily and very quickly, Suplari is the first phase of procurement's do, check, act cycle."
The most powerful insight came from cross-property spend comparisons — exactly the kind of Step 4 segmentation analysis described above: "All of this comes from having the data at your disposal, at your fingertips, to be able to say, 'Why is this cluster of titles or business units paying 5X, 10X more for the same product and service that these clusters of titles are paying for?' You can't do that without Suplari."
Ball also co-leads zero-based budgeting with MediaNews Group's CFO — demonstrating how procurement spend analysis bridges into finance when the data is credible: "One key to success in my role is the fact that we do zero-based budgeting. I co-lead that effort with our CFO."
Key takeaway: The highest-value findings in procurement spend analysis often come from comparing spend patterns across business units, locations, or properties — surfacing price variance that isn't visible at the category level. Read the full MediaNews Group case study →
How AI is changing spend analysis in procurement
The most labor-intensive parts of spend analysis — data collection, cleansing, classification, and pattern recognition — are exactly the tasks where AI delivers the most value. Here's how:
Automated classification at scale. AI models can classify millions of transactions against procurement taxonomies in hours, with accuracy rates that match or exceed manual classification. This eliminates the biggest bottleneck in the spend analysis process and makes continuous (rather than periodic) analysis feasible.
Anomaly detection. Machine learning identifies unusual patterns — sudden price increases, duplicate payments, transactions that don't match historical patterns — that human analysts would miss in large datasets. These anomalies often point to errors, fraud, or savings opportunities.
Natural language querying. Rather than building complex reports or waiting for analyst support, procurement professionals can ask questions in plain language: "Which IT services suppliers increased their rates more than 10% year over year?" or "Show me our top 5 categories by off-contract spend percentage." This democratizes access to spend insights across the procurement organization.
Predictive insights. AI doesn't just analyze what happened — it forecasts what's likely to happen. Predicting supplier price increases based on market signals, forecasting demand shifts that affect sourcing strategies, and identifying contracts approaching unfavorable renewal terms before category managers are caught off-guard.
Suplari's AI Procurement Agent embeds these capabilities directly into the spend analysis workflow. Rather than requiring procurement teams to master data science tools, the Agent delivers classified data, surfaces anomalies, answers natural language questions, and proactively recommends actions — transforming spend analysis from a periodic project into a continuous intelligence function. See how it works.
Get started with Suplari
If your spend analysis program is stuck in spreadsheets — or if you're generating reports that don't translate into sourcing action — Suplari can help. Our AI-powered platform automates the data-heavy steps of spend analysis and delivers the actionable insights that drive better procurement decisions.
Start with a spend analysis assessment to understand where your data stands and what quick-win opportunities exist in your procurement spend.
