The Rescue

What We Inherit

You're 14 months into a $30M implementation. The original vendor deployed 200 consultants. You have a prototype that doesn't pass compliance review, a burn rate that keeps you awake, and a board asking hard questions. The vendor's response is a change order and another 6 months. You need someone who can inherit the wreckage and ship a working system.

The technical wreckage assessment happens in week one. What we find is almost never surprising: a data model that commingles regulated and non-regulated data in ways that make compliance architecturally impossible without a rebuild. Access controls implemented as application-level role checks rather than database-level constraints — meaning any application bug can bypass them entirely. Audit logging that captures what the developer thought the auditor would want, not what the regulation requires. Infrastructure provisioned in cloud regions without the required BAAs. The assessment is demoralizing for the client. It is a normal Tuesday for us.

The failed vendor is usually still present. Sometimes still billing. The team that built the non-compliant prototype is often the team being asked to remediate it — a structural conflict of interest that produces slow remediation and perpetuates the sunk cost logic that kept the engagement going. Our process is to work alongside your existing team where it makes sense and replace the vendor's team where it doesn't. We are not interested in blame attribution. We are interested in production systems.

Switching vendors mid-engagement is the decision most organizations make too late. The sunk cost psychology is predictable: two more months of burn feels more manageable than acknowledging that the previous eighteen months produced nothing. But the math is consistent — every additional month with a vendor who cannot deliver is a month of burn that produces no production output, while reducing the time available before the deadline. We have never encountered an organization that wished they had switched vendors later.

Failed implementations aren't accidents. They're the predictable output of a business model that optimizes for engagement duration over delivery speed. The vendor sold a 24-month transformation. The team deployed on your project is measured on utilization, not outcomes. Discovery extends because discovery is safe billable time. Architecture decisions are deferred because decisions create accountability. Phase gates exist to manage the vendor's risk of scope reduction, not your risk of failed delivery. By month twelve, the sunk cost is too high to switch, and the vendor knows it. By month eighteen, the change orders have doubled the original budget. By month twenty-four, you have a prototype that doesn't pass compliance review and a new 12-month remediation roadmap.

The compliance failure that makes most failed implementations unshippable is not a surprise to the delivery team. The engineers who built the prototype know that the audit logging is insufficient. They know the access controls are not granular enough. They built it anyway because no one with authority to slow the build had the compliance depth to identify the gap in real time. The compliance workstream that was supposed to catch these issues was a separate team that joined in month six, after the architecture was locked. Retrofitting compliance onto a locked architecture is more expensive than building it compliant from the start — and the vendor knew this before month one.

The procurement process that produced the failed engagement was designed to find the most credible proposal, not the most capable delivery team. The proposal was written by senior partners who assembled a reference list from across the firm's portfolio. The delivery team was assembled from whoever was available after the proposal won. This is not fraud — it is the structural gap between consulting firms' proposal capability and their delivery capability, operating exactly as designed. The reference implementations were real. The team that delivered them was not the team on your project.

Failed Vendor Recovery Guide

How to assess inherited wreckage, triage a non-compliant system, and execute a mid-engagement vendor switch without losing the deadline.