# Craig Cooper

Craig Cooper writes about Lean and Six Sigma, operational improvement, and the governance needed to make digital and AI transformations work. The posts focus on practical questions such as who owns decisions, whether data can be trusted, and how organizations can tell if a transformation is delivering results.

- Profile: https://b2binfluencer.marketing/creators/craig-cooper/
- LinkedIn: https://www.linkedin.com/in/craigmrcooper
- LinkedIn followers: 2,636
- Topics: artificial-intelligence, supply-chain, ai-governance, digital-transformation, data-governance, operational-excellence, continuous-improvement, lean-manufacturing
- Based on 17 dated posts, latest 2026-10-08

## Summary

Craig Cooper is most useful for practical governance tests and operational examples about digital, AI, and transformation decisions.

- Best for: A manufacturing consultancy or governance-led transformation thought leadership campaign.
- Not for: A cloud or AI vendor wanting high-volume sponsored video, demos, or hard reach targets.
- Check first: Confirm terms, sponsorship slots, and recent LinkedIn performance and audience data before contacting them.

### Takeaways

- Failed transformations are governance problems wearing an execution costume.
- Dashboards cannot settle a disagreement about which number is true.
- Autonomous AI raises one question: who decides when the human disagrees?

## Main ideas

- **Shared technology is not an advantage; the organisation around it is**: Cooper argues that when everyone can buy the same machines or models, the differentiator is how people are organised around them. He makes the case through manufacturing history: in the 1980s, Western manufacturers attributed Japan's dominance to better robots and automation, then bought the same flexible manufacturing systems. But a 1986 Harvard study by Jaikumar found that US plants using identical equipment produced 10 different parts while Japanese plants produced 93. The MIT study published in 1990 as "The Machine That Changed the World" traced the gap to management philosophy, not machinery. Cooper applies the same pattern to AI: organisations pulling ahead are not winning with better algorithms or more data, but with governance structures built for complex adaptive systems. His refrain is "governance before systems, systems before tools."
- **Failed transformations are governance problems wearing an execution costume**: Cooper treats most failed transformation projects not as methodology failures but as unclear decision rights, and offers tests rather than slogans. In a steering committee Cooper suggests asking three questions: who decides when we stop this project, what specific decision rights are changing, and which behaviours are being rewarded differently. Confused or circular answers, in Cooper's reading, mean the project is already on a trajectory that ends in lessons-learned presentations nobody reads. A second test crosses functions: ask a production supervisor what happens if they stop the line to fix a quality issue, then ask finance which metrics determine that supervisor's bonus, and compare the answers. Cooper calls the gap between the org chart and who actually decides a "transformation tax," and sets a 90-day rule for getting clear answers.
- **Dashboards cannot settle a disagreement about which number is true**: Cooper distinguishes data availability from data trust. Better pipelines, real-time analytics and master data management do not answer who has authority to decide which version of a metric wins, or what happens when systems disagree. Without that authority, organisations automate their data chaos and get faster chaos, while executives fall back on gut feel when the numbers conflict. The hard part is often political: departments that have reported a metric for years must admit that the previous strategy relied on figures they cannot defend.
- **Vendor selection is a balance-sheet and delivery-risk decision, not just a capability comparison**: Cooper extends his governance argument to suppliers and AI deployments. Organisations announce transformational commitments before establishing how to fund and execute them, while procurement processes assess what a vendor can do more readily than whether it can financially deliver. He contrasts two companies cutting roughly 30,000 people for different reasons: Amazon, which he says built mechanisms to scale decision-making and trims when layers accumulate, and Oracle, which committed to infrastructure at a scale that raises financing questions. His summary: cloud strategy is no longer only an IT decision but a balance-sheet governance decision. Chief information officers with cloud contracts should treat the buildout as an infrastructure delivery risk, not just a service agreement.
- **Autonomous AI raises one question before any others: who decides when the human disagrees**: When AI agents are given real authority inside regulated operations, Cooper argues the interesting question is not technical capability but conflict resolution. Cooper frames it narrowly and concretely: if an agent recommends an action and a human reviewer flags it, whose judgment prevails, and where is that rule written down? Cooper calls this a governance question rather than a technical one, and notes that deployments can proceed at enormous scale without that rule being answered publicly. The post is short and works as a prompt for a risk discussion rather than a developed argument.
- **Pull systems expose the truth about process capability; buffers hide it**: Cooper argues for demand-driven flow over forecast-driven building, and says a pull system's value is clarity, not efficiency. When material moves only against actual demand, stockouts and excess inventory both signal process capability; buffer stock conceals those signals. One failed implementation involved a supplier running Kanban while keeping an MRP-driven schedule: pulling from one system while pushing with another. Cooper also cautions against treating inventory cuts as automatically safe: the post asks whether halving inventory would expose capability gaps or cause customer delivery failures, and treats that breaking point as where the real improvement work starts.
- **Test improvement claims by what is still running two years later**: Cooper's stated priority for continuous improvement is durability rather than pilot wins, and goes outside the sector to pursue it. Writing from 25 years of Lean and Six Sigma work in automotive, aerospace, pharmaceuticals and energy, Cooper asks National Health Service practitioners for their field findings and repeats the same standard in Cooper's questions: what patterns show up repeatedly, what fails predictably, and what is genuinely embedded and still functioning 18–24 months later without constant intervention. The prompt came from personal experience as a patient, where the same hospital, department and staff produced two very different experiences within two weeks. Cooper frames that as a gap between capability and consistency, not competence. This post is a request for practitioner input rather than a finished position.
- **Opting out of AI is not a strategy**: Cooper relays Geoffrey Hinton's argument that AI development cannot be stopped because it is too useful and too internationally competitive, and that governments should push companies to invest heavily in safety research so superintelligence remains under human control. He translates that for manufacturing executives: competitors, suppliers and customers are all moving toward AI, so declining to participate is not a viable position. Hinton's view is his own; Cooper's contribution is the business framing.

## Work to read

- [How to Spot Transformation Failure Before It Costs ££££££££££££](https://www.linkedin.com/posts/craigmrcooper_how-to-spot-transformation-failure-before-activity-7420378526391967744-kgB2)
- [Three years. £47M. Zero ROI.](https://www.linkedin.com/posts/craigmrcooper_three-years-47m-zero-roi-thats-what-activity-7415305098345512960-nD9W)
- [We keep making the same mistake. For 40 years now...](https://www.linkedin.com/posts/craigmrcooper_aitransformation-leanthinking-operationalexcellence-activity-7422915238494244864-NxUw)
- [The Romans Already Knew: Pull vs Push in Supply Chains](https://www.linkedin.com/posts/craigmrcooper_leanmanufacturing-supplychain-operationalexcellence-activity-7414580346450493440-YCI8)
- [$156 billion in capex commitments.](https://www.linkedin.com/posts/craigmrcooper_156-billion-in-capex-commitments-major-activity-7424198729945063424-_VMv)
- [To Every Continuous Improvement Practitioner Working in the NHS](https://www.linkedin.com/posts/craigmrcooper_to-every-continuous-improvement-practitioner-activity-7417208673753563136-Me9r)
- [The news broke quietly last week.](https://www.linkedin.com/posts/craigmrcooper_the-news-broke-quietly-last-week-goldman-activity-7426539101832581120-FNkd)
- [Why "We Can't Stop This" Should Change Your Strategy](https://www.linkedin.com/posts/craigmrcooper_why-we-cant-stop-this-should-change-your-activity-7420524339021647873-Bf76)

## Brand relationships

- **Leancore Ltd** (owned_business)
- **Human Tech Lab** (owned_business)
- **Oracle** (mention)
- **Goldman Sachs** (mention)

## Partnerships

The eight posts provide no documented sponsorship offer, paid partnership, or way to commission work from Craig Cooper. They contain no sponsored-content markers, partner hashtags, or disclosures of paid arrangements. The posts do establish the subject matter a partnership would draw on. Cooper writes as a practitioner of Lean and Six Sigma and operational improvement, applying that lens to digital and AI programmes: who owns a decision, whether two systems report the same number, and what governs the distance between a board announcement and a delivered result. A manufacturing consultancy proposing a thought-leadership partnership on governance-led transformation would be working in their established territory. The posts are specific and example-led: a FTSE 250 manufacturer whose finance, operations and warehouse systems each reported different inventory turns; a £47M MES rollout nobody owned; the Roman supply depot as a pull-versus-push analogy; and banks pulling back from Oracle-linked data-centre lending. That is the kind of concrete drafting such a collaboration would produce. Before costing a campaign, ask Cooper directly for current commercial terms, available formats, disclosure policy, and a booking route; none is published in the posts.

## Alternatives

- Julius Lilly: Choose Lilly when the consultancy's campaign makes the industry-level case for modernising manufacturing rather than focusing on one client's steering committee. Lilly frames AI adoption in manufacturing as a national competitiveness problem: Japan's shrinking operator base and low generative-AI adoption versus China's dark factories; the argument that Japanese plants deploy AI beside the production lane while Chinese plants put it inside the lane; and a Western reindustrialisation sequence from AI implementation to productivity recovery to reshoring. That material suits a piece aimed at government-adjacent audiences, industrial policy readers, or a consultancy selling a sector thesis to boards deciding whether to build domestically. The same partnership need — governance-led transformation — would be argued from the outside in: why the productivity window is open, how cheap inference changes the reshoring maths, and what the missed driver of cognitive latency costs. Lilly's figures include US factory construction peaking at $235.6bn in 2024, 3.8M US manufacturing workers needed by 2033 with about 1.9M potentially unfilled, and roughly 40% average labour-productivity gains in a recent McKinsey lighthouse cohort.
- Amresh Kumar: Choose Kumar when the consultancy wants to sell the delivery side of transformation rather than the operating-model argument: finding Oracle Cloud, PeopleSoft, SAP, and Workday specialists who are not on job boards. Kumar writes as the founder of a recruiting firm, describing a passive network built through long-term relationships, an active ERP and IT talent database, and AI-driven sourcing layered with human qualification. The strongest overlap with a campaign on governance-led transformation is the observation that ERP programme success depends on consultants who understand delivery realities, stakeholder management, and business processes inside large organisations — not technical expertise alone. For a firm whose buyers keep failing transformation programmes on skills and continuity, that offers a credible alternative angle with a named population: public sector, federal, higher education, financial services, manufacturing, healthcare, and CPG environments.
- Kit Yu: Choose Yu only if the campaign is reframed around capital rather than operations. Yu's posts address investors: management meetings with a fund's chief financial officer and head of investor relations, confidence in 2028 financial objectives, fundraising and deployment trends in private equity and credit, valuation discounts, and the growth of Asia-Pacific and Middle East infrastructure private debt from $12.9bn in 2020 to $24.9bn in 2025, with a forecast of $49.5bn by 2030. This is the right partner if the consultancy wants to address the finance side of transformation — the people who decide whether a plant, transmission line, or portfolio company gets funded — and argue that governance and delivery risk belong in the investment case. That is the closest point of contact with Cooper's own vendor-financing post.

---
AI-assisted summary of public posts. Audience demographics are not available. Engagement appears only where we checked recent posts. Method: https://b2binfluencer.marketing/methodology/