The $150,000 Weekly Burn Rate
The Board meeting lasted exactly 12 minutes. That’s how long it takes to tell a CIO that if the ERP system isn’t live by quarter-end, they should start updating their LinkedIn profile.
This wasn't a hypothetical scenario. It was a Tuesday morning in October with a client I’ll call "Tom," a VP of Engineering at a $2B manufacturing firm. The situation was a textbook disaster: The project was nine months behind schedule. The budget was $8M over the original $25M allocation. And the System Integrator (SI) had just requested another "change order" for $1.2M to fix data migration scripts they claimed were broken by legacy data issues.
The burn rate was $150,000 per week. That’s $30,000 a day, just to keep the lights on in a project room where no decisions were being made.
Tom was trapped in what I call the "Sunk Cost Death Spiral." He had spent so much political capital defending the delay that he couldn't admit the root cause. He thought it was a technology problem—specifically, that the middleware wasn't handling the transaction volume. He was wrong.
We ran a 5-Day Operational Assessment and found the truth in 48 hours. The middleware was fine. The problem was Decision Latency.
The Math of Indecision
We tracked the lifecycle of a critical decision: "Should we migrate 7 years of historical sales data or just 2 years plus open orders?"
- Day 1: Issue raised by the SI technical lead.
- Day 4: Discussed in the "Working Group." No consensus. Deferred to Steering Committee.
- Day 9: Presented to Steering Committee. The CFO wasn't there; decision deferred.
- Day 16: Steering Committee meets again. Requests a "risk analysis" from the SI.
- Day 23: Risk analysis presented. Decision made to migrate 2 years.
That is a 23-day cycle time for a decision that should take 23 minutes. In a project with 200+ open decision points, this latency guarantees failure. The "Latency Tax" wasn't just time; it was costing the company $1M per month in delay-related overhead.
The 45-Day Recovery Playbook
We didn't have time for a new roadmap. We had 45 days. We implemented a "War Room" operating model designed to eliminate decision latency and force a go-live.
1. The Governance Guillotine
The first thing we did was kill the Weekly Steering Committee. It was a theater of the absurd where slides were read to executives who didn't understand the technical blockers.
We replaced it with a Daily Executive Standup. 8:30 AM. 15 minutes. Standing room only. No slides.
The Rule: Only three people allowed—The CIO (Tom), the CFO (the money), and the VP of Operations (the user). If a decision was blocked, it had to be resolved in that room, right then. If they couldn't decide, the default answer was "No Change / Stick to Vanilla."
In the first week, we cleared a backlog of 42 stalled decisions. We reduced the average decision latency from 14 days to 4 hours. By removing the "committee safety blanket," we forced leadership to actually lead.
2. The Scope Machete
The original scope included "Phase 1" requirements that were ludicrous for a first release, including an automated AI-driven inventory forecasting module. This is a classic example of why digital transformations fail: trying to innovate before you can operate.
We applied a brutal filter: "Can we ship a box and send an invoice without this?"
If the answer was yes, the feature was cut. We slashed 40% of the Phase 1 scope, including the forecasting module, complex rebate management logic, and 15 custom reports that nobody had looked at in three years. We focused entirely on the "Order-to-Cash" and "Procure-to-Pay" value streams.
3. The Vendor "Show Me" Rule
The System Integrator was hiding behind RAG (Red-Amber-Green) status reports that were perpetually "Amber." They would report "90% complete" on development for weeks.
We instituted a new rule: No PowerPoint. Only Production.
In our daily reviews, the SI could not speak about a feature unless they could demo it in the UAT (User Acceptance Testing) environment. "Almost done" meant zero. This exposed the reality that the "90% complete" code was failing basic integration tests. We forced the SI to swap out their lead architect for someone who had actually deployed this specific ERP version before. Intervening with vendors is uncomfortable, but when you're 9 months late, politeness is a luxury you can't afford.
The Outcome: Ugly but Operational
On Day 45, we went live. It wasn't a celebration with champagne and balloons. It was a tense, quiet weekend of data validation.
Was it perfect? No. We had to use manual workarounds for rebate processing for two months. The warehouse team had to use scanners that looked like they belonged in 2010 until the new tablets arrived.
But the system worked. Orders flowed. Invoices went out. The $150,000 weekly bleed stopped.
The Lesson for the C-Suite
Recent data from 2025 shows that 70% of ERP initiatives fail to meet their original business goals. In discrete manufacturing, the failure rate is even higher, with budget overruns averaging 215%. These projects don't fail because the software is bad (usually). They fail because organizations try to pave cow paths—digitizing broken processes—and then govern the project like it's a casual Tuesday afternoon.
If you are a CIO or Operating Partner looking at a stalled project, stop looking at the code. Look at your calendar. If you see a "Monthly Steering Committee" and no daily decision log, you are already dead. You just haven't stopped moving yet.
Fix the governance, and the code will follow.
