Your Approval Stack Is Organizational Theater

The Autopsy of an Approval Stack That Was Killing More Innovations Than Any Competitor Ever Did

Quick Answer: The Refrigeration division required seventeen signatures to change an engineering bracket. Not a strategic pivot. Not a platform decision. A bracket. A decision that should have taken a single afternoon routinely took forty-seven days as the paperwork moved through seventeen separate approvers, most of whom had no operational stake in the outcome. When we stripped the approval stack down to four signatures — Engineering, Quality, Manufacturing, and the GM — cycle time dropped to six days. Quality incidents: zero. The thirteen eliminated “safeguards” had not been safeguards at all. They had been organizational theater, performing the appearance of control while producing no actual control value. The autopsy of that approval stack is the cleanest example I have ever documented of how structural calcification quietly kills a division one bracket at a time.

The Autopsy Begins: What the Seventeen-Signature Stack Was Actually Protecting

When I walked into the Refrigeration division and asked why a routine engineering change required seventeen signatures, the answer came back with complete confidence. “Quality control. Cost control. Making sure changes don’t create problems.” Every person on the approval chain could defend their signature. Engineering Lead needed to verify technical correctness. Mechanical Director needed to approve departmental impact. Quality Manager needed to check specifications. Quality Director needed to confirm the Quality Manager’s work. Manufacturing Manager needed to assess shop-floor feasibility. Operations Director needed to authorize the Manufacturing Manager’s assessment. And so on, all the way through seventeen separate review layers before the change could go live.

Every individual signature looked defensible. The stack as a whole was killing the division.

I asked the obvious follow-up question: “What happens if we skip approvals?”

The engineering director stared at me like I had suggested arson. “We can’t. Those approvals exist for a reason.”

“What reason?”

Long pause. “Quality control. Cost control. Making sure changes don’t create problems.”

“Has anyone calculated the cost of not changing? The opportunities lost while waiting for seventeen signatures?”

Longer pause. “That’s not how we think about it.”

That is the diagnostic sentence. Not defiance. Not ignorance. A complete, structural inability to weigh the cost of the approval stack itself against the risks the approval stack was supposedly protecting against. The division had built a governance apparatus designed to prevent one category of problem — bad engineering changes — while creating a much larger category of problem — the systematic inability to ship engineering changes at competitive speed. And the cost of the second category never appeared on any report, because nobody had ever been asked to measure it.

The Forensic Evidence: What the Thirteen Extra Signatures Were Actually Doing

An autopsy requires quantification. Before you eliminate a governance layer, you document what the layer is actually catching. Here is what the forensic audit revealed across sixty consecutive engineering changes that had moved through the seventeen-signature stack:

Signatures That Caught Real Issues. Engineering Lead: caught technical errors in roughly 8% of changes. Quality Manager: caught specification issues in roughly 4%. Manufacturing Manager: caught shop-floor feasibility problems in roughly 6%. GM: caught strategic misalignment in roughly 2%. These four signatures, combined, caught approximately 20% of issues before they shipped — and the issues they caught were material enough to justify the review time.

Signatures That Caught Nothing. The other thirteen signatures caught essentially nothing across the sixty-change sample. Mechanical Director: no catches. Quality Director: no catches (the Quality Manager’s review had already caught everything). Operations Director: no catches (the Manufacturing Manager’s review had already caught everything). Supply Chain Manager: no catches. Procurement Director: no catches. Finance Analyst: no catches. Finance Director: no catches. Legal Review: no catches. Compliance Officer: no catches. Regulatory Affairs: no catches. Program Manager: no catches. Product Director: no catches. Executive Signoff: no catches.

Thirteen signatures. Sixty engineering changes. Zero caught issues across the entire sample.

The thirteen extra signatures had not been preventing problems. They had been adding calendar time. Each approver averaged 2.7 days of queue time before their signature arrived, and because the signatures were sequential rather than parallel, the full stack added roughly thirty-five days of pure process latency to every engineering change — all in exchange for zero incremental defect prevention.

The Deep Framework: Why “Approval Stacking” Is a Structural Failure Mode, Not a Governance Problem

The seventeen-signature stack is the canonical example of what I call approval stacking — the structural pattern where governance layers accumulate over years without ever being audited for whether they are still producing value. Approval stacks grow through a predictable sequence that every mature organization eventually falls into.

It starts with a specific incident. Fifteen years earlier, a bad engineering change shipped without adequate Quality review. Leadership responds by adding a Quality Director signature to the approval chain. Reasonable response to a specific incident. Ten years earlier, a different change created unexpected manufacturing disruption. Leadership adds an Operations Director signature. Again, reasonable. Seven years earlier, an engineering change had downstream cost implications nobody predicted. Leadership adds a Finance Director signature. Still reasonable, in isolation.

Each individual addition is defensible. None of them ever get removed. And after fifteen years, you are running a seventeen-signature approval chain where the original incidents that justified the original signatures are not even remembered by the people executing the process. The signatures persist because they exist. The process persists because nobody has ever been tasked with auditing whether it is still doing what it was installed to do.

This is why approval stacking is a structural failure mode rather than a governance problem. It is not caused by bad judgment. It is caused by the absence of any routine mechanism to sunset governance layers that have outlived their usefulness. Most organizations have mechanisms to add approval requirements (any incident can trigger one). Very few organizations have mechanisms to remove approval requirements (it takes a forensic audit). The accumulated result, over decades, is a governance apparatus that looks defensible layer by layer but is catastrophically dysfunctional as a whole.

The Sacred Terms: “Organizational Theater” Versus Actual Control

In the theology of this framework, the sacred distinction is between organizational theater and actual control. Both look identical from inside the organization. A signature requirement that catches real issues and a signature requirement that catches nothing are both, visually, signatures on an approval chain. The governance documentation cannot distinguish between them. The audit trail cannot distinguish between them. The people executing the process cannot distinguish between them, because everyone is performing their role correctly — the Quality Director is reviewing, the Finance Analyst is approving, the Compliance Officer is signing off.

The distinction only becomes visible when you run the forensic audit that asks the uncomfortable question: across the last sixty decisions that moved through this approval layer, what did this layer actually catch?

If the answer is “caught real issues in 5% or more of decisions,” the layer is actual control. Keep it. If the answer is “caught nothing, or caught issues that upstream layers had already caught,” the layer is organizational theater. Eliminate it.

Most mature organizations discover, when they run this audit, that 60-80% of their approval stack is theater. Not because the organization is poorly run, but because theater is the default steady state of any governance apparatus that has not been forensically audited in the last five years. Entropy does the work. Layers accumulate. Nobody ever removes them until someone comes in with a stopwatch and a clipboard and starts asking what each layer is actually doing.

The Result: Six Days, Zero Quality Incidents, Millions in Unlocked Velocity

We eliminated thirteen of seventeen signatures over the course of four weeks. Engineering Lead stayed. Quality Manager stayed. Manufacturing Manager stayed. GM stayed. The other thirteen were removed, documented as “eliminated due to forensic audit showing zero catches across sixty-change sample.”

Cycle time dropped from 47 days to 6 days. Engineering change throughput increased by roughly 7x. Quality incidents in the twelve months following the change: zero.

This is the outcome every organization that runs the audit eventually produces. Faster cycle times. Higher throughput. No increase in defects. And a clearer understanding of which governance layers are actually earning their existence and which layers have been riding on organizational inertia for a decade or more.

The thirteen eliminated approvers did not disappear. They were redeployed to work that actually required their expertise. The Legal Review function still existed — it just stopped being inserted into every engineering change regardless of whether legal considerations were relevant. The Finance Director still approved financial decisions — just not routine engineering changes with no financial implications. The Compliance Officer still enforced compliance — just not as a checkbox on bracket modifications that had no compliance dimension.

The Uncomfortable Truth

Seventeen signatures protected nothing. The division had been operating under the illusion that each signature was catching potential problems when the forensic audit revealed that thirteen of them were catching nothing at all. The stack persisted because every signature was defensible in isolation, and because no one had ever audited the stack as a whole against the actual outcomes it produced. Every mature organization carries a version of this stack — an accumulated governance apparatus built layer by layer over decades, where each individual layer can be justified but the combined effect is a calcified decision-making system that loses to faster competitors on every meaningful timeline. The only question that matters is whether you run the audit before or after the market has already moved around you.

About the Author

Todd Hagopian is the architect of the Hypomanic Operational Turnaround (HOT) System and the author of Stagnation Assassin: The Anti-Consultant Manifesto. He has led five Fortune 500 and Fortune 1000 transformations, including turnarounds at Berkshire Hathaway, Illinois Tool Works, and Whirlpool Corporation, generating over $3 billion in documented shareholder value. His frameworks — including the 80/20 Matrix, the Karelin Method, the 3-A Method, the 52-Project Pipeline, the 48-Hour Decision Guarantee, the Orthodoxy Evaluation Matrix, the Four-Dimension Capacity Assessment, the Three Integration Points, and the Exploit-Subordinate-Elevate execution protocol — have been featured across Forbes, Fox Business, NPR, and The Washington Post. He holds an MBA from Michigan State University and writes from his desk in Solon, Ohio.

Join the Stagnation Assassin Circle

If this framework resonated, the full arsenal — including the Corporate Implementation Guide, the HOT System video course, the framework summaries library, and monthly office hours with me — is waiting inside the Stagnation Assassin Circle. It is a private community for leaders actively waging war on stagnation, and your membership is free.

Claim your membership and continue the conversation at toddhagopian.com.