Why most transformations stall In many enterprises, “AI” and “ERP modernization” are treated as parallel initiatives: one owned by innovation teams, the other by IT delivery. The result is predictable—proofs of concept that don’t scale, ERP roadmaps that don’t change outcomes, and supply chain teams that feel the change but don’t see the value. The alternative is to treat transformation as process reimagination with measurable business impact—using AI, SAP/ERP product thinking, and design thinking as a single system. A simple way to frame the work When I’m asked to help shape or rescue a transformation program, I start with three questions:

• Which decisions matter most? (e.g., allocation, replenishment, promise dates, inventory buffers)
• Where does the truth live? (master data, transactional signals, external signals, and who owns them)
• How will value be measured? (service, working capital, cost-to-serve, productivity, speed to market)

If these aren’t clear, teams default to shipping features instead of changing outcomes. The playbook: 5 steps that connect AI, SAP/ERP, and outcomes

1. Anchor on a value case with operational metrics. Pick 1–2 outcomes that executives care about (for example: forecast bias reduction, inventory turns, OTIF, expedited freight, or planner productivity). Define baseline, target, and measurement cadence.
2. Map the end-to-end process and decision points. Use design thinking to identify friction, handoffs, and “decision latency.” This is where AI can actually change the work—not just add dashboards.
3. Design the product: workflows first, models second. Specify how a planner, buyer, or customer service rep will act differently. Then decide what predictions, recommendations, or automations are needed—and what guardrails are required.
4. Modernize data and master data ownership. Most AI failures are data failures. Establish ownership, quality thresholds, and a pragmatic data product approach (signals, definitions, lineage) that fits your ERP reality.
5. Industrialize delivery inside the ERP ecosystem. Integrate into SAP/ERP workflows, security, and change management. Treat the solution as a product with releases, adoption metrics, and continuous improvement—not a one-off project.

Common pitfalls (and how to avoid them)

• “AI first” without process ownership: assign a business owner for each decision and metric.
• ERP upgrades that replicate old processes: redesign the workflow before configuring the system.
• Over-automation too early: start with decision support and clear exception handling, then automate where trust is earned.
• Value not tracked: build measurement into the operating rhythm—weekly for leading indicators, monthly for financial impact.

If you’re planning 2026 initiatives If you’re considering AI in supply chain, an SAP/ERP roadmap refresh, or a broader digital transformation program, I’m happy to compare notes. A short conversation can often clarify where to focus for measurable impact.

Transformation succeeds when AI, ERP, and operating model changes are designed as one system—measured by outcomes, not activity.

Contact me if you’d like to discuss an engagement, respond to this post, or explore what “measurable impact” could look like in your context.