Use case

Purchase requisitions and PO processing.

From the field, AI native workflow redesign of purchase order processing process within Procure-to-pay Procurement function.

Get the playbook
Convolving expertise

A senior Convolving delivery team partnered with the procure-to-pay function for one sprint. Operators from our expert network – with forty combined years inside enterprise procurement organisations – scored the redesign at each checkpoint. Forward-deployed engineers built on top of Coupa's guided buying capability and the team's internal procurement platform. One flat fee, artifact out, no retainer creep.

Situation

Today every requisition is touched by procurement. Standard requests, exceptions, and edge cases all enter the same queue.

Employees submit purchase requests through Coupa or the internal procurement platform. Procurement reviews each request for policy compliance, preferred-supplier alignment, and budget availability. Approved requests are converted to POs and routed to suppliers. The volume is rules-based but the touchpoints are manual – a procurement specialist reviews the same request shape thousands of times.

Touchpoints 100% Of requests reviewed manually
Cycle time 2–5 days Request to PO sent to supplier
Maverick spend 12–18% Of indirect spend, off-catalogue
Exception ratio Unknown Standard vs exception not separated

Click any node to see the activities and tools behind it. Open the canvas in fullscreen for the horizontal view.

Complication

Largest obstacles and inefficiencies.

Every request is touched by procurement.

Specialists review the same standard shape thousands of times. The judgement on edge cases gets crowded out.

Two to five days from request to PO.

Business partners learn to submit early – or work around procurement entirely on time-sensitive buys.

Twelve to eighteen percent of spend goes off-catalogue.

Maverick spend is structural when the path of least resistance for business partners is the one that bypasses the system.

Resolution

The AI-native cycle.

Same six steps. Click any node to see what the redesign does in that step.

Touchpoints 30–40% ▼ 60–70% vs today
Cycle time Same day ▼ 80% vs today
Maverick spend 5–8% ▼ 50% vs today
Exception ratio Surfaced Standard vs exception split
Key changes

What the redesign actually shifts.

Touchpoint reduction

  • Sixty to seventy percent of standard requests pass through without procurement touch.
  • Specialists work the exception queue rather than the standard queue.
  • PO creation is automatic for compliant, in-policy requests.

Cycle time

  • Two to five days drops to same-day for standard requisitions.
  • Approval chain runs in parallel where policy permits.
  • Business partners stop padding their timelines around procurement.

Maverick spend

  • Off-catalogue spend halves as the path of least resistance becomes the in-system one.
  • Preferred suppliers surface automatically at intake.
  • Anomaly detection flags new-vendor patterns before they normalise.

Exception discipline

  • Standard versus exception is now visible as a metric, not a feeling.
  • Every exception carries a structured recommendation.
  • Reviewer edits feed the policy library and improve auto-handling.

Deploy this in your team.

The redesign above ships as a step-by-step playbook. Intake conversation flow, policy rule schema, exception taxonomy, and the rollout cadence we use on engagements.

Get the playbook Or book a coffee