The Seduction of the Moonshot
Every operations leader has sat through the pitch. A consultant, a software vendor, or an internal champion presents a bold vision: we will transform the operation. Reengineer the workflow. Replace the legacy system. Automate everything. The projected improvement is dramatic. 50% cost reduction. 3x throughput. A complete reimagining of how the work gets done.
The pitch is compelling because the pain is real. Operations teams know they are inefficient. They see the manual workarounds, the duplicate data entry, the tribal knowledge that should be documented, the processes that evolved by accident rather than design. The promise of sweeping it all away and building something better is seductive.
It is also, statistically, a losing bet.
According to McKinsey's research on operational transformations, 70% of large-scale transformation programs fail to achieve their stated objectives. BCG's analysis puts the number at 75%. The failure rate has been remarkably stable across industries and decades, despite improvements in technology, methodology, and change management practice.
Meanwhile, a less glamorous approach produces dramatically better outcomes. Companies that pursue systematic, incremental improvement across multiple processes simultaneously achieve cumulative gains that match or exceed transformation targets, with success rates above 80%. The math explains why.
Expected value = improvement * probability of success. Incremental approach delivers 3x higher expected value.
Why Transformations Fail
The 70% failure rate is not random. Transformations fail for structural reasons that are well-documented but persistently ignored because the transformation narrative is too appealing to abandon.
Scope creep is built into the model. A transformation project starts with a defined scope. Within three months, the scope has expanded by 30 to 50% because the team discovers interconnections and dependencies that were invisible during planning. Each expansion adds cost, extends the timeline, and increases the probability of failure. A Standish Group analysis of large IT-enabled transformation projects found that the average project delivers 56% of the originally specified functionality at 189% of the original budget.
The organization cannot absorb the change. Large transformations require people to change how they work across multiple dimensions simultaneously. New systems, new processes, new roles, new metrics, new reporting relationships. Research in organizational psychology consistently shows that humans can absorb 1 to 2 significant workflow changes at a time. Transformations typically require 5 to 10 simultaneous changes, which overwhelms the organization's capacity to adapt. The result is workarounds, resistance, and eventual reversion to old methods.
The business does not stop. Unlike a software company that can freeze features for a release cycle, an operations team must keep operating while it transforms. This creates a dual-operating reality where people run the current process while implementing the new one. The cognitive load and time pressure of running two parallel systems degrades both the current operation and the transformation effort.
ROI takes too long to materialize. Large transformations typically require 18 to 36 months before the full benefit is realized. During that window, leadership changes, priorities shift, budgets get pressured, and organizational patience expires. Projects that deliver incremental value along the way survive these disruptions. Projects that require full implementation before any value appears are vulnerable to cancellation at every quarterly review.
The Incremental Alternative
The alternative is not sexy. It does not produce dramatic before-and-after stories or impressive conference presentations. But it works, and the math explains why it works better than the alternative.
A 10% improvement in a single process is almost always achievable. It requires understanding the current state, identifying the largest friction point, and implementing a targeted fix. The risk is low because the scope is small. The time to value is short because the change is contained. The organizational disruption is minimal because only one thing is changing at a time.
Now consider what happens when you do this across multiple processes in parallel. An operations team that achieves 10% improvement in order processing, 10% in inventory accuracy, 10% in shipping efficiency, 10% in returns handling, 10% in customer communication, 10% in vendor management, 10% in quality control, and 10% in reporting does not get an 80% improvement. They get a 57% improvement, because the gains compound multiplicatively. But with an 80 to 90% success rate on each initiative, the expected value is 46 to 51%, compared to 13 to 15% expected value for a single 50% transformation with a 25 to 30% success rate.
The incremental approach delivers 3 to 4 times higher expected value than the transformation approach. This is not opinion. It is arithmetic.
The Toyota Lesson Nobody Learns
The most successful operations company in modern history did not build its advantage through transformation. Toyota built the Toyota Production System through decades of incremental improvement, what they call kaizen. The system that eventually became the global standard for manufacturing excellence was not designed in a boardroom and implemented in a 24-month program. It was built one small improvement at a time, over 40 years.
The relevant insight is not that Toyota was patient. It is that incremental improvement creates a compounding knowledge advantage that transformations cannot replicate. Every small improvement teaches the organization something about its own processes. That learning informs the next improvement, which is slightly more sophisticated. Over time, the organization develops a deep understanding of its operations that makes each subsequent improvement faster and more effective.
Transformation projects, by contrast, import understanding from outside. Consultants analyze the operation, design a new model, and hand it to the organization to implement. The organization gains a new system but not the understanding that produced it. When conditions change and the system needs to evolve, the organization does not have the muscle to adapt it. They need another transformation.
This is why companies that pursue serial transformations every 3 to 5 years never achieve operational excellence. They are perpetually starting over instead of building on what they learned. Companies that pursue continuous improvement get better at getting better. The gap between the two approaches widens every year.
The Execution Framework
Implementing incremental improvement at scale requires more structure than most people assume. Without a framework, incremental improvement devolves into random tinkering that produces inconsistent results and eventually loses organizational support.
The framework that produces the most consistent results has four elements.
Process mapping with pain-point scoring. Every key process is mapped at a level of detail that reveals where time, errors, and frustration concentrate. Each pain point is scored on two dimensions: impact (how much does this cost in time, money, or quality?) and tractability (how likely is a targeted intervention to fix this?). The scoring produces a prioritized list of improvement targets that maximizes return on effort.
Two-week improvement cycles. Each improvement initiative is scoped to deliver measurable results within two weeks. This is a hard constraint, not a guideline. If an improvement cannot be delivered in two weeks, it is either too large (break it into smaller pieces) or too complex (it needs more analysis before it is ready for implementation). The two-week cycle creates urgency, limits risk, and produces a steady stream of visible wins that maintain organizational momentum.
Measurement before and after. Every improvement is measured against a specific baseline. Not a general sense that things are better, but a number: cycle time went from 4.2 days to 3.1 days. Error rate went from 6.3% to 2.8%. Touches per transaction went from 7 to 4. Without measurement, improvements are invisible, and invisible improvements do not build organizational commitment to the process.
Portfolio management. Running 6 to 10 improvement initiatives simultaneously across different parts of the operation requires coordination. A lightweight portfolio view tracks what is in progress, what has been completed, what the cumulative impact is, and what is next. This portfolio becomes the operational improvement roadmap, replacing the transformation project plan with something more adaptive and more resilient.
When Transformation Is Actually Necessary
Incremental improvement is not always the answer. There are legitimate scenarios where a fundamental change is required.
The current system cannot be improved incrementally. If the core technology platform is at end of life, cannot integrate with modern systems, or has a fundamental architectural limitation that blocks improvement, replacement may be the only option. But even in this case, the replacement should be phased rather than all-at-once, and the process improvements should happen in parallel rather than waiting for the new system.
The market has shifted fundamentally. If a company's operating model is built for a market that no longer exists, incremental improvement to that model is optimizing the wrong thing. But this scenario is rarer than companies think. Most of the time, the core operating model is sound; it is the execution that needs work.
Regulatory or compliance requirements mandate change. When a new regulation requires a fundamentally different approach to data handling, reporting, or process documentation, incremental adaptation may not meet the compliance deadline. Even here, though, the best implementations are phased rather than monolithic.
The honest assessment is that fewer than 20% of the transformation projects currently underway in mid-market operations are genuinely necessary as transformations. The rest would produce better outcomes as structured incremental improvement programs.
Starting Tomorrow
The most practical thing an operations leader can do is pick the three processes that cause the most pain and commit to improving each one by 10% within 30 days. Not 10x. Not a transformation. A measurable, specific, achievable 10% improvement.
Then do it again. And again. Within six months, the cumulative effect will exceed what most transformation projects achieve in two years, at a fraction of the cost, with none of the organizational trauma, and with a team that has learned how to improve itself rather than how to survive a transformation.
The 10x myth persists because it makes for better stories than 10% improvements. But operations is not a story. It is a system. And systems improve through iteration, not revolution.
Not sure which processes to start with? Run a free diagnostic to identify the three highest-impact improvement targets in your operation.