How AI Turns Three Months of Procurement Spend Analysis Into Two Weeks

A procurement team might be sitting on millions of dollars in potential savings, but the analysis needed to find those savings takes so long that half the opportunities expire before anyone acts on them. That gap between knowing savings exist and actually capturing them has plagued procurement teams for years.

A recent Domo AI livestream tackled that exact problem head-on. Mark Boothe, Domo Chief Marketing Officer, and Chris Sweeney, Domo's Professional Services and Procurement Lead for Europe, the Middle East, and Africa, walked through a five-step process that compresses procurement spend analysis from months to weeks using AI. Chris has lived this transformation firsthand, noting that what "used to take us two to three months" is now "consolidated down to one to two weeks."

Here is how that process works, step by step, and what it means for procurement teams ready to move faster.

Start with clean, standardized supplier data

Every spend analysis project hits the same wall early: dirty data. Supplier names arrive in dozens of variations. "IBM," "International Business Machines," and "IBM Corp." all refer to the same vendor, but a spreadsheet treats them as three separate entities. Multiply that across thousands of suppliers, and the data becomes nearly impossible to analyze.

Instead of a team manually combing through supplier records for weeks, AI scans the entire dataset and suggests name normalizations. It groups likely duplicates, flags inconsistencies, and presents recommendations for a human reviewer to accept or decline. The human stays in control of every decision, but the heavy lifting of identifying patterns across massive datasets happens in minutes rather than days.

This step alone can save weeks of manual effort, and it sets a clean foundation for everything that follows.

Generate a spend taxonomy tailored to the business

Once supplier data is clean, the next challenge is organizing spend into meaningful categories. Most organizations either force-fit their data into a generic industry taxonomy or spend weeks building a custom one, and both paths slow down the analysis before the real work begins.

AI examines the company's actual spend data alongside its business profile and generates a custom category hierarchy. The taxonomy reflects how the organization actually spends money, not how a consulting framework assumes it should. And because a human can edit the taxonomy at any stage, the final structure combines AI speed with organizational knowledge.

Chris described this as one of the moments where the time savings become dramatic. Building a taxonomy manually requires deep institutional knowledge and careful iteration. AI delivers a strong first draft in a fraction of the time, and the team refines from there.

Categorize every spend line automatically

With a clean taxonomy in place, AI agents take over the categorization work. Every spend line item lands in the appropriate category in the hierarchy.

This is the step that would traditionally consume the most analyst hours. Categorizing thousands (or tens of thousands) of individual transactions requires painstaking review. AI handles the volume at scale, applying consistent logic across the entire dataset. The result is a fully categorized spend profile that a team can trust as a starting point for strategic analysis.

For data professionals, this is where the value of upstream data cleaning pays off. Clean supplier names and a well-structured taxonomy mean the AI categorization engine has high-quality inputs, which translates directly to higher-quality outputs.

Generate savings plays for each category

Categorized spend data is useful, but it only becomes valuable when someone acts on it. This is where AI shifts from organizing data to surfacing opportunities.

For each spend category, AI analyzes the specific spending patterns and generates three to four savings strategies. These are not generic suggestions pulled from a template. Each savings play comes with supporting evidence drawn from the actual data, along with estimated financial impact.

A procurement lead reviewing these plays can accept the ones that align with organizational priorities and reject the rest. The AI handles the analytical work of finding savings. The human decides which opportunities to pursue.

"The AI is not making the decision," Chris explained. "It is giving you the evidence and the options. The human is always the one who says yes or no."

That balance between AI capability and human judgment is what makes the process practical for enterprise procurement teams that need both speed and accountability.

Track savings progress with live dashboards

The final step closes the loop. Live dashboards show before-and-after snapshots of spend by category, with AI-generated analysis of whether savings targets are on track.

This turns procurement analysis from a one-time project into a continuous process. Instead of running a quarterly or annual spend review, teams can monitor progress in something close to real time. When a savings play is underperforming, the data surfaces the issue early enough to adjust course.

For business decision-makers, this is the step that transforms spend analysis from a reporting exercise into a management tool. The dashboard does not just show what happened. It shows whether the organization is capturing the value it planned to capture.

Key takeaways

The five-step process covers the full arc of procurement spend analysis.

• AI handles data cleaning, taxonomy building, and spend categorization in days rather than months, with humans reviewing and approving at each step.
• Human oversight remains central at every step, from approving supplier name matches to accepting or rejecting savings plays.
• Live dashboards turn one-time analysis into ongoing spend management, surfacing problems early enough to act on them.
• The combination of AI speed and human judgment makes the process practical for enterprise teams that need both velocity and accountability.

Watch the full session

The livestream covered more than the five steps outlined here. Mark and Chris also demonstrated the Contract Management AI App and walked through a guided buying application that extends procurement intelligence into day-to-day purchasing decisions.

Watch the full session to see each step in action and explore the additional tools discussed.

Domo is an AI and data products platform that gives procurement teams the speed to act on savings opportunities before they disappear. Get a demo to see how Domo can accelerate spend management for any organization.