Roughly 70% of industrial transformations fail not because technology is inadequate, but because organizations lack the fitness to implement it. The solution is to start with a single packaging line, set a bounded, measurable goal, and execute a 12-week reset using the Four Clocks framework to measure awareness, decision, action, and recovery—producing a banked result that proves transformation capability.
- 70% of industrial transformations fail, primarily due to organizational readiness issues rather than technology limitations.
- The Four Clocks framework measures time to awareness, decision, action, and recovery to diagnose and fix operational fitness.
- A successful packaging line reset takes 12 to 24 weeks, targets one bounded goal (like reducing changeovers from 45 to 20 minutes), and requires a named line owner with real authority.
- Most AI pilots on packaging lines fail because they violate three rules: data first (trained on unrealistic conditions), an owner named (unclear accountability), and a gate that kills (no baseline or exit criteria).
- Organizations must prepare for crises through simulation, inoculation with real scenarios, and institutionalization—making time to recovery a tracked KPI with regular drills.
ALTIX Consulting, Yannick Schilly
The book’s named examples come from industrial manufacturing rather than consumer goods. Its framework, however, is built for any plant network. A central tool is the Four Clocks, which measure an organization’s time to awareness, decision, action, and recovery. Schilly answered questions about how the approach applies to CPG packaging operations.
Packaging World: Our readers are packaging engineers and packaging operations managers. What can they do with this book if their leadership isn’t driving a transformation?
Yannick Schilly: You do not need permission to get fit. The book’s machinery is deliberately usable at line level. Read your own Four Clocks on the last three incidents your line had, and scope a reset you own: one line, one bounded ambition, 12 to 24 weeks, inside your existing authority. Bank one number your controller will sign, and you have done two things. You have improved your operation, and you have produced the most persuasive argument that exists inside a corporation, which is a result. In a mature CPG operation, the opportunity at your level is sharper, not smaller, because the daily management rhythm and trustworthy baseline data a reset needs are already on your wall. Leadership is a behavior, not a title.
What does one of those resets look like on a packaging line?
Weeks one and two, scope tight: one line, one bounded ambition with a number. Say, changeovers from 45 minutes to 20, or OEE from 55% to 65% on the constraint line. If the result will not be visible in the P&L inside the window, the scope is wrong. Weeks two and three, mobilize a small team with real authority, a line owner rather than a project office, and a daily drumbeat. Then execute and bank. Stabilize first, put standard work before software, and measure progress weekly against the baseline taken before anything changed. In the final weeks, lock in and reset. The gain is written into standard work and next year’s budget, the team is debriefed, and the next reset is scheduled. The banked result is a number the controller signs, not a dashboard and not a story.
Vision inspection, predictive maintenance, and OEE analytics are among the most common AI pilots on packaging lines. What do CPG pilots most often get wrong?
Rarely the technology. The models are ready; the organization often is not. Most failed pilots violate all three clauses of a simple rule: data first, an owner named, and a gate that kills. On data, the vision system is trained on curated day-shift images, then meets night-shift lighting, film gloss variation, and a changed label supplier. The pilot ran on data the line will never see again. On ownership, the pilot belongs to IT or a digital team, so when the model flags a failing bearing, no one on the line owns the response. Predictive maintenance without a changed maintenance calendar is a subscription, not a saving. And without a gate, pilots live indefinitely in “promising.” Nobody took a baseline before switching anything on, so no result can ever be proven. Artificial intelligence rewards organizations that were already intelligent.
How should a packaging operation prepare for a recall, a material shortage, or an input-cost spike?
With three doses, administered on your terms. First, simulate. Run a mock recall end to end on a real lot, with a stopwatch on the traceability query: hours to full lot genealogy, not the hours the system diagram promises. Tabletop a resin or film shortage the same way, along with the scenario every operation now owes itself, a cyber attack. Could you still run the factory and deliver product with the ERP and MES dark? For how long, and on what paper process? Second, inoculate with something real but bounded. Qualify an alternate material on one SKU against a hard deadline, or run one line for a week under a simulated 20% input-cost increase and make the margin math visible. Third, institutionalize. Time to recovery becomes a tracked KPI, drills go on the calendar like audits, and every exercise ends with the Four Clocks read against last time.
The Four Clocks at the bottom of the back cover measure how long an organization takes to become aware of a problem, decide, act, and recover. Readers can measure their own operation against the clocks with the free Superfit Score diagnostic at superfitcompany.com.ALTIX Consulting, Yannick Schilly
What should CPG companies expect from packaging equipment suppliers during a transformation, and what stays with the brand owner?
Expect honest readiness talk, including a willingness to tell you a line is not stable enough for what you are buying. Expect co-engineering presence on your floor rather than remote diagnosis, and machines that ship contextualized, open data you own. Training and spare-parts commitments belong in the contract, along with uptime partnerships where the supplier’s revenue is tied to your output.
Process stability, line ownership, standard work redesign, and decision rights over what runs on your floor never transfer.
Co-packers deserve special attention. They are often smaller, private businesses without a mature operating system behind them, and your network’s fitness is set by its least fit node. Audit their fundamentals, share your standard work generously, and write data and recovery expectations into the tolling agreement rather than assuming a maturity that is not there. You can outsource machinery, engineering, even production. You cannot outsource fitness.
What’s your advice for a packaging operations leader at a CPG company?
Your numbers can look excellent for years while your fitness quietly decays, and that is how most of the 70% begin. Pick one line, run one honest 12-week reset, and bank one number your controller will sign. Transformation is not a corporate program that happens to you. It is a capability you train while you are healthy.
The Superfit Company: Why Industrial Winners Never Stop Transforming by Yannick Schilly (ALTIX Press) is available in hardcover (ISBN 979-8-9970398-0-6), paperback (ISBN 979-8-9970398-1-3), and ebook through Amazon and major online booksellers. A free diagnostic, the Superfit Score, and a companion toolkit are at superfitcompany.com.
