Preventing process debt 2.1 Understanding process debt All Chief Information Officers (CIOs) know of technical debt, which accumulates when many disjointed solutions are deployed over time, compounding the costs to ultimately fix and integrate them. Eventually a “tax” must be paid to improve system-wide performance – a costly barrier to scaling-up endeavours.22 In today’s tech-enabled world, all Chief Operating Officers (COOs) should think similarly about their process debt. The value-add of a new technology is rarely in the technology itself, but in the process improvements it enables. Counsel on how to introduce new technology comes in adages such as: “Don’t automate a bad process”, or “Don’t optimize that which shouldn’t exist”. The more new technologies are deployed at a production line without corresponding process changes, the more process debt is accrued and the more investment may be needed in process changes, SOP updates23 and training to capture the full potential value of already-deployed technologies. Process debt therefore refers to the “tax” a company pays as it redresses accumulated waste and suboptimal practices of any operation – paid in the form of inefficiencies that eat into the value- adding potential of new technologies.24 Optimizing information flows across people, material and data Sources of variability and waste naturally accumulate across systems, raising the total cost of ownership (TCO).25 Lighthouses invest to fully understand flows of information across people, material and data before redesigning them – going directly to the frontline to learn all the human elements of a process, conducting process mining exercises to map true material flows and tracing data pipelines to ensure accuracy and scalability of digital solutions. Haitian in China mapped out its craftsmen-led production processes to identify that their AI-enabled smart “nose” solution, a machine learning-based olfactory analysis platform leveraging aroma sensor data, left gaps in communicating results to necessary team members. To address the process debt, they invested in features to auto-alert the right people to act on non-compliant batches of soy sauce. To supplement this, given that bad batches could be delivered to the lab on automated guided vehicles (AGVs) and that failure inspection results were now digitalized, their QC engineers and responsible operators could spend more time on the production shopfloor. This solution improved inspection capacity by over 4x.26 Lighthouses such as Haitian leverage data to streamline interactions between people and material flows. They do this by not only integrating tools that present trustworthy information and recommendations, but also by redesigning the processes around those tools to capture total value. EXAMPLE Haitian Global Lighthouse Network: The Mindset Shifts Driving Impact and Scale in Digital Transformation 13

Global Lighthouse Network 2025