Efficiency Without Resilience Is Just a Faster Way to Stop

Insights from the Scaling Business Summit 2026, Ho Chi Minh City.

Moderating was Alyssa Nguyen, SAP Sales & Marketing Specialist at Gradion. On stage: Christoph Dirr, General Manager at Würth Industries Vietnam, and Christian Boos, VP Global - Head of Sustainability Innovation at SAP. The session moved through five concrete questions: inventory, AI, supplier gaps, sustainability ROI, and what one investment in 2026 would matter most.

What they agreed on was not what most factory managers expected to hear.

1. The Right Amount of Stock Is Not Zero

Zero inventory sounds like the goal. It is not. At least not yet.

Christoph Dirr was clear: "Invisible inventory is still something for the future. The goal is not to have nothing but to have the exact right amount." That amount depends on how resilient the factory is, what price tag is put on safety, and how much visibility exists into what can go wrong.

Christian Boos offered a different framing: buffer with data rather than with inventory. End-to-end visibility makes safety stock a calculated decision, not a gut call. Once the data is there, stock levels can be reduced with evidence, not optimism.

The practical implication: the KPI is not the size of the buffer. It is the quality of the data behind the decision.

Lesson 1: Stop optimizing for zero. Start optimizing for visibility.

2. Switzerland vs. Germany - Resilience Wins the Long Game

Christoph Dirr told a story about two railway systems. Switzerland and Germany. Same era of infrastructure, both dating to the 1800s. Same mix of traffic. Every train type running on the same rail. One system is famously on time. The other is not.

The difference is not the workforce. It is not the technology. "Switzerland optimizes for resilience and sustainable efficiency. Germany tries to put as much volume as possible on the rail every single day. Throughput, throughput, throughput and complexity punishes that."

Switzerland spends nearly double per inhabitant on its rail infrastructure compared to Germany. The network can absorb pressure. Germany's cannot and when one thing breaks, everything breaks.

The manufacturing parallel is direct. Christian Boos framed it as a question of what you are asking AI to do. Maximize efficiency or prevent failure? The design question matters more than the technology. A smart factory is an investment in the right architecture. Not just more throughput.

Lesson 2: The factories that will scale are not the most efficient. They are the most resilient.

3. Do Not Force Suppliers to Change. Route Their Data Instead

The supply chain gap is real. Local suppliers using Excel. Buyers demanding clean data and transparent reports. Most solutions involve convincing small vendors to adopt new systems. That rarely works.

Christian Boos's answer was direct: let them stay on Excel. Treat it as a data source. Use AI agents to make the spreadsheet data readable, structured, and integrated into end-to-end planning. "Rather than spending time pushing four or five employee vendors into ERP solutions, take the data and argue on it."

The result is the same information, at a fraction of the cost and none of the change management. The data exists. The integration architecture is the missing piece.

Christoph Dirr confirmed the outcome from the manufacturer side. Using SAP combined with AI agents, Würth Industries Vietnam shifted sourcing decisions and is now manufacturing more in Vietnam than before. The calculation came from the system. Not from intuition.

Lesson 3: The data already exists. Connect it. Do not try to reformat the people behind it.

4. Sustainability Is a USP If You Have the Data to Prove It

The factories that treat sustainability as a compliance exercise will always see it as a cost. That is the wrong frame.

Christian Boos drew the line clearly: "If you do sustainability only to fulfill reports and be compliant, that is a cost driver. If you do it to gain competitive advantage, it becomes a USP."

The difference is product-level data. An emission factor per unit. Energy and water consumption per run. Numbers that can be compared against a competitor's equivalent.

Christoph Dirr added a live example. The EU's new carbon border adjustment mechanism assigns Vietnam a relatively low carbon emission rate. For Würth Industries Vietnam, fasteners produced here that means their products are becoming more attractive to German buyers. Sustainability is already influencing sourcing decisions in Europe.

The broader principle: sustainability done right has no final price tag. "It should always return otherwise it is not sustainable by definition." The investment is in making the data measurable, not in the optics.

Lesson 4: Sustainability without data is a cost. Sustainability with data is a competitive edge.

5. The One Investment That Makes Everything Else Possible

The closing question was precise: if a factory can only make one investment for its sustainability goals in 2026, what should it be?

Both speakers answered without hesitation. Data points.

"The only thing we can be sure of right now is that you will need data in the future," said Christoph Dirr. "And the first thing to have data are data points." Clean.

Automated. Continuous. Not a new machine. Not another ERP module. The infrastructure that makes every other decision arguable.

Christian Boos extended the frame. Connect what is already in the factory before buying anything new. Then go further, bring in external signals too. Weather patterns. Climate disruptions. Anything that affects supplier delivery. "In the first wave, connect what you know and what is there. Then go beyond."

Predictive forecasting improves when the data chain reaches past the factory wall. The factories that will lead are not the ones that automate fastest. They are the ones that know what is happening before anyone else does.

Lesson 5: Connect before you buy. Clean the data before you ask AI to use it.

The CEO Execution Playbook: What to Do Tomorrow

1. 1. Map your buffer decisions to data, not instinct. Audit one inventory category this week. Identify what data would let you reduce it without increasing risk. Start there.
2. 2. Pressure-test for resilience, not throughput. Ask your operations team: if one critical supplier fails tomorrow, what stops? The answer tells you whether you have built for volume or for stability.
3. 3. Stop forcing supplier standardization. Identify two or three Excel-based suppliers in your chain. Explore whether AI integration can route their data into your planning layer before investing in a change management project.
4. 4. Build a sustainability fact sheet for your top three products. Emission factor, energy use, water consumption per unit. Even rough estimates. This is the start of a competitive data layer and increasingly a precondition for European buyers.
5. 5. Audit your data infrastructure before your next capital expenditure. Before any new machine purchase, ask: can we collect data from what we already have? If the answer is no, fix that first.

→ Watch the full conversation on YouTube