Within six weeks, a B2B company lost two critical members of its revenue team: the VP of Sales and the RevOps Manager. Both had been with the company for over three years. Both had built and maintained the systems, reports, and processes that the revenue function ran on.
Nobody had documented anything. The VP's weekly forecast was a manual exercise she ran from memory and instinct. The RevOps Manager maintained a set of Salesforce reports that nobody else knew how to build or interpret. When both left, forecast accuracy dropped from 85% to 44% within six weeks. The pipeline number became unreliable. The board started asking questions leadership couldn't answer with confidence.
"We had a revenue system that worked. Right up until the people running it left. The system wasn't actually a system. It was two people's knowledge."
The Revenue Diagnostic identified System Resilience as the dominant failure layer. The audit revealed a complete dependency map. Every critical revenue function that lived in an individual's head rather than the system:
The first priority was stabilising the forecast. A weighted probability model was built in Salesforce based on historical close rates by stage. Replacing the VP's manual estimate with a system-generated number that updated automatically as deals moved. Forecast accuracy recovered to 71% within three weeks of implementation, without any human intervention.
The Salesforce report library was rebuilt from scratch - 14 reports covering pipeline health, stage velocity, rep performance, and forecast accuracy. Each report was documented: what it measures, what data it draws from, what decisions it informs, and how to read it. Any new hire could pick this up on day one.
Stage exit criteria were written into the CRM as required fields. A deal could not move forward without confirmation that the buyer had taken the relevant action. This eliminated the "hope stage". Deals that were moving based on optimism rather than evidence.
25% of the pipeline that had become uninterpretable without the two departed team members was recovered through a structured deal audit. Reviewing call recordings, email threads, and contact history to reconstruct deal status and next steps. The rest was written off cleanly, producing a smaller but accurate pipeline number.
A 47-process RevOps runbook was built covering every recurring task, cadence, and decision rule in the revenue function. Written to be executed by anyone, not just a RevOps specialist.
Forecast accuracy recovered from 44% to 89% within 60 days. The pipeline number became trustworthy again. Smaller than before, but accurate. The board meeting that quarter was the first in two months where leadership arrived with confidence in the numbers they were presenting.
When the company hired a new VP of Sales three months later, onboarding took four days instead of the estimated six weeks. The system documentation existed. The runbook existed. The reports existed. The new VP walked into a functioning infrastructure, not a blank slate.
A revenue system that only works because of specific people is not a system. It is a risk. This engagement turned the risk into infrastructure.
The Revenue Diagnostic identifies your resilience gaps in 90 seconds. Free, no CRM access required.
Run the Free Diagnostic →This case study describes the most expensive kind of revenue system failure: the kind that only becomes visible when it's too late. The company had a functioning revenue system. Deals were closing, forecasts were being submitted, reports were being produced. What it didn't have was a system that could run without the two people who had built it.
When both the VP of Sales and the RevOps Manager left within six weeks of each other, the operational knowledge they had accumulated over three years left with them. Forecast accuracy dropped from 85% to 44%. The pipeline number became unreliable. The board was asking questions that leadership couldn't answer with confidence.
The audit produced a complete dependency map: every critical revenue function that lived in an individual's head rather than the system. The forecast had never been documented. The Salesforce reports were built in ways only one person understood. The onboarding process lived in a Notion doc last updated 18 months earlier.
The rebuild took 60 days: documented forecast methodology, rebuilt Salesforce reports with annotation and owner fields, a structured onboarding playbook, and a system health dashboard the CEO could read without interpretation. Forecast accuracy recovered to 89%. Revenue continuity was maintained throughout the transition. Zero external disruption.
System resilience is the degree to which a revenue system can operate consistently when key individuals are unavailable. Through attrition, absence, or growth. A resilient system has documented processes, annotated reporting, structured handoffs, and onboarding infrastructure that allows any qualified person to maintain continuity. A non-resilient system is one where critical knowledge lives in people rather than the system.
Start by mapping every critical function to the person currently running it and ask: if that person left tomorrow, would this function continue? For each gap, build documentation: the decision logic for the forecast, the annotation for every report, the process steps for every recurring task. The goal is not to replace people. It is to make the system's operation independent of any one person's continued presence.
Typically: deal context is lost for every active opportunity the VP was involved in, forecast methodology disappears with them, and the team loses its primary point of escalation for stalled deals. If the CRM contains deal notes, next actions, and structured stage data, continuity is possible. If deal knowledge lived in the VP's head and email, the pipeline becomes unreliable within weeks. This engagement demonstrates both outcomes. Before and after the resilience rebuild.
Document the forecast methodology before the transition occurs. This means: the stage probability weightings and their basis, the adjustment factors applied to the weighted number, the deals excluded from forecast and why, and the standard for what counts as "commit" versus "best case." A documented methodology can be executed by anyone with access to the pipeline. An undocumented one disappears with its author.