Toward a data-driven supply chain

Supply chain operations are complex, fast-moving, and critically important to business performance in the manufacturing sector. Decisions need to be made every day under pressure, often with incomplete information and tight production profit margins. While many organizations already have plenty of data, the real challenge is whether people can trust the data, access it in time, and use it to improve the way work gets done.

One organization was facing a situation of this kind. Teams across internal processes worked with mountains of data every day, however that data lived in separate systems, was outdated, or had to be manually compiled in excel spreadsheets before it could be used for decision-making. Over the years, people had created clever workarounds (often close to magic Excel sheets) to keep operations running. The problem was that those workarounds only hid deeper challenges, that were fragmented processes, inconsistent master data, and a lack of real-time visibility into what was actually happening on the ground. The strategic question hence became: How do we become truly data-driven in supply chain operations in a way that improves efficiency, supports automation, and brings our people along?

Technology is not the main blocker

The organization wanted to evaluate and enhance workflow efficiency across its supply chain by leveraging data-driven insights, automation, and ultimately AI. The focus was, besides identifying optimization opportunities, also about understanding how to best support its employees through the change.

In the organization's environment, supply chain work often depends on experience, tacit knowledge, and manual routines that have been refined over years. Introducing automation or AI is, besides a technical upgrade, really a shift in how decisions are made and how daily work is structured. If that shift isn’t handled carefully, it will create resistance.

Many employees naturally worry that automation and AI are designed to reduce headcount. That perception alone can stop adoption before it even begins, regardless of how good a solution is. That’s why the organization also wanted to understand how change should be managed so employees can adapt and trust the new ways of working.

Our team were given the task to analyze the organization's situation from multiple angles. It was necessary to examine the organization's current processes to uncover automation potential, assess employee perspectives on automated workflows, evaluate existing tools and their ability to support data-driven decision-making, and discover how to best support adaptation when new technologies are introduced.

Starting with real work

The project took root in the organization's day-to-day reality. Over a few months, we worked alongside its teams through workshops, observation days on site, and targeted discussions that explored end-to-end workflows in planning, fulfillment, and product management.

In these sessions, unspoken friction points that slow people down were uncovered. The biggest blockers we identified were missing trust in data and missing confidence in decisions based on that data.

Across functions, people described the same patterns. Master data was maintained in multiple systems with inconsistent definitions. Reports were slow or outdated, real-time access to critical information was limited, and Excel had become the glue between systems that weren’t properly connected. These workarounds did work, but they also robbed teams of time, introduced risk, and made automation difficult.

co-creation and structured discovery

Co-creation and data design methods were used with the organization's process managers and key stakeholders to define the problems, map the processes, and identify improvement opportunities based on the actual operational needs. Workshops and observation days were held on site, supported by targeted interviews.

Alongside this, the identified opportunities were analyzed through mind mapping and case evaluation. Relationships between use cases were identified, estimated their business potential estimated, the data currently available was assessed, and the findings combined with knowledge in data structures, data platforms, automation, and AI/ML. This ensured the roadmap would reflect the business needs, technical feasibility and long-term scalability.

Turning insights into a roadmap for impact

The result of this work was a shared understanding of where the organization stands today, and a jointly developed roadmap that outlines practical steps toward a data-driven supply chain.

Three themes ran through the findings:

First, the foundation had to be strengthened. Real-time data availability and consistent master data are essential before any advanced automation or AI can deliver value. Without them, decision-making reverts to guesswork and manual validation. A solid data foundation is the prerequisite for scalable automation.

Second, reporting and analytics needed to become more reliable and intuitive. This meant improving real-time reporting capabilities, reducing manual rework, and increasing adoption of existing analytical tools so that teams can trust what they see.

Third, automation should remove friction where it matters most. The most valuable use cases were the ones that addressed everyday inefficiencies: workflows that require manual confirmation, capacity planning that lacks visibility, and supply chain processes that depend on spreadsheets because systems don’t integrate properly.

Change management is the real success factor

Understanding how to help employees trust and adapt to new technologies cannot be emphasized enough. In supply chain operations, people’s trust is everything. If employees don’t trust the data, they won’t trust automation. And if they don’t trust automation, they will work around it, even in cases where the solution is technically correct.

By involving process owners and key stakeholders throughout the work, the roadmap became something the organization could recognize and own. The use cases were expressed in the language of the people doing the work and they already know how to use their time better when cases are implemented.

Adoption begins by designing solutions that support employees. AI should be introduced as something that removes unnecessary manual work and helps experts focus their energy towards making better decisions.

a shared direction

By the time the project wrapped, the organization had a clear, realistic path toward a supply chain that uses its data as an asset.

The work identified the highest-impact improvement opportunities and defined practical next steps across forecasting, analytics, automation, and AI, starting with quick wins that can demonstrate business value early. Just as importantly, the organization now has an understanding of the actions needed to make its supply chain operations data-driven from an organizational perspective.

Do you recognize these challenges?

We are always happy to help you turn scattered, manual, and siloed information into a connected, well-governed, business- and AI-ready data foundation. Read more here!