Behavioral Health · 56 Employees

A 200-Point eNPS Spread Hidden Behind a “Good” Score

A 56-employee behavioral health and recovery center carried an eNPS of +36.5. Confidential interviews across all 23 departments revealed two organizations living inside one: departments scoring +100 sitting beside departments scoring -100.

Participation Rate

92.9%

52 of 56 employees spoke candidly, across every department and all five role levels.

92

Friction points surfaced

23

Departments covered

The Engagement

What we assessed.

Industry

Behavioral Health

Addiction recovery, regulated environment

Employees

56

Across all functions

Interviewed

52

92.9% participation rate

Departments

23

Every department covered

Role Levels

5

Individual contributor to executive

Friction Points

92

Occurrences surfaced in interviews

Reports Delivered

8

Structured diagnostic reports

Prioritized Actions

18

Sequenced across four phases

The Compounding Cycle

Three friction patterns feeding each other in a loop no one could see from the top.

The most significant finding was not any single friction pattern. It was the way the top three patterns reinforced one another in a self-sustaining loop. Capacity overload left staff with no time to document processes or maintain data quality. Undocumented processes created data gaps and single-point-of-failure risk. Data unreliability pushed people toward personal spreadsheets and manual workarounds. Those workarounds consumed still more capacity, feeding the overload that started the cycle.

This is the Workaround Spiral in action. The organization was not failing at any one thing. It was being slowly ground down by the accumulation of moderate-severity friction across nearly every function. Of the interviews assessed, the large majority registered at moderate severity, only one reached high severity at the interview level, and the rest were low. This was not a crisis. It was a slow grind that would prevent scaling and eventually exhaust the people holding the organization together.

The pattern is invisible from the top. The eNPS looks healthy. The work gets done. Clients receive care. But the mechanism producing those outcomes is individual heroics, not scalable systems. Confidential organizational discovery across all 23 departments made that invisible mechanism visible.

Key Findings

Three high-severity findings that aggregate data would never surface.

High Severity

Institutional knowledge dangerously concentrated in individuals.

Essential operational knowledge lived exclusively in individual memories, with no documentation behind it: all facility system knowledge (HVAC, electrical, plumbing, septic), IT system administration, clinical program templates, and executive-level process knowledge. Employees across ten departments described their work in ways that revealed how much of the organization’s continuity depended on specific people never being unavailable.

One team member acknowledged that all facility knowledge lived in their head with virtually no written record. An IT supervisor described having almost no visibility into a colleague’s functions. Clinical staff maintained personal repositories of referral lists, templates, and program materials that existed nowhere else. The departure of any one of several key individuals would have halted essential operations with no documented fallback, and cross-training was absent.

Why this was invisible

Employees do not volunteer single-point-of-failure risk on a survey. They describe it as "how things work here." It took conversational depth and adaptive follow-up to surface how much operational continuity rested on individuals who could never afford to be out.

High Severity

The EMR and data infrastructure could not support the organization.

The electronic medical records system lacked needed functionality, and the CRM module was described at the executive level as non-functional. External contractors were required to pull basic reports. Staff across 14 departments built and maintained personal spreadsheets because the official system could not give them data they trusted, a smart adaptation to a tool that was failing them.

One executive described the system’s reporting limitations and the reliance on outside contractors. A marketing team member kept personal spreadsheets she trusted more than the platform. A front office team member tracked compliance data such as OSHA, EEOC, and training certificates in personal spreadsheets. The same pattern appeared 26 times across 14 departments. Data fragmentation undermined billing accuracy, created rework loops in Finance, and left leadership unable to manage by reliable metrics.

Why this was invisible

A survey question about "data systems" returns a satisfaction rating. It does not reveal that 14 departments independently built shadow systems because they could not rely on the official platform. That specificity required adaptive conversation that followed each frustration to its structural root cause.

High Severity

Capacity overload was systemic, not situational.

Workloads across nearly every department left no margin for the unexpected, for documentation, or for process improvement. This was not seasonal or temporary. It was the organization’s operating norm, surfaced as the single most widespread pattern across 15 departments.

One team member worked part-time yet carried 24/7/365 on-call responsibility as the sole maintenance person for a facility that had gone five years without dedicated maintenance. Clinicians described two minutes between assessments and group sessions. Clinical staff took laptops home at night to finish notes and avoided sick days because no coverage existed. This sustained overload directly drove burnout and hero behavior, which in turn prevented documentation, training, and process improvement, feeding the compounding cycle.

Why this was invisible

Employees under capacity overload will rate "workload" as a concern on a survey. A survey cannot distinguish "busy season" overload from "the organization structurally depends on my personal sacrifice to function" overload. That distinction changes the entire intervention, and it only emerged when people described their actual daily experience.

eNPS Deep Dive

A +36.5 score that concealed four departments in crisis.

Company-Wide eNPS

+36.5

Rated “Good”

Departmental Spread

200 pts

From -100 to +100

Promoters

57.7%

Passives

21.2%

Detractors

21.2%

Highest Dept

+100

Nursing

Lowest Dept

-100

Customer Support

The aggregate concealed two fundamentally different employee experiences. In some departments, staff felt supported, aligned with the mission, and satisfied. In others, staff were deeply disengaged. The same organization, the same leadership, the same mission, and a 200-point gap between the best and worst departmental experience.

A founder reading the dashboard sees +36.5 and a 57.7% Promoter rate, a number worth celebrating. The department-level view shows something else: four departments in the significant-concern range (Client Services at -66.7, Customer Support at -100, Marketing at -50, and Medical at -33.3) and a variance pattern that demands department-specific intervention rather than organization-wide messaging.

Deliverables

Eight structured reports. One sequenced action plan.

Executive Summary

Headline findings, the top three issues, and immediate actions for the leadership team.

Scaling Leadership Report

Decision fog, escalation dependency, role ambiguity, shadow systems, and governance gaps constraining leadership.

Change Readiness Assessment

Readiness scoring, resistance signals, change fatigue indicators, and leadership bandwidth analysis.

AI & Automation Readiness

Automation opportunities, candidates, risk assessment, and a phased roadmap aligned to data-integrity prerequisites.

Consolidated Action Plan

The authoritative reference for all 18 actions with owners, deliverables, dependencies, and success metrics.

Follow-Up Agenda

Structured discussion guide with clarifying questions, a data gap analysis, and implementation sequencing.

Trust Architecture

Employee Summary

A plain-language document written for the workforce, closing the feedback loop by showing employees what was heard and what comes next.

Solutions Roadmap

Discovery findings mapped to specific operational tools designed to address the friction patterns staff surfaced.

Why This Matters

What surveys, town halls, and traditional consulting would have missed.

The compounding cycle is invisible from any single vantage point.

No one employee sees the full loop of overload driving documentation gaps driving data unreliability driving workarounds driving more overload. The AI-powered analysis surfaced it because the conversation reached 52 people across 23 departments and the pattern emerged across the full data set.

The eNPS spread requires granular, confidential data.

An anonymous survey can produce a +36.5 aggregate. It cannot produce the department-level analysis showing a 200-point spread, and it will not reveal the qualitative reasons behind the divergence.

Healthcare-specific friction requires adaptive conversation.

A survey cannot follow a clinician's description of rushing between sessions to the realization that the pace prevents adequate documentation, which creates compliance risk, which compounds burnout. Dave's interviews follow those threads because the conversation adapts in real time.

Knowledge concentration is invisible until someone asks the right questions.

No one reports "I am a single point of failure" on a survey. They describe how their work gets done. It took adaptive follow-up to surface that the process lived entirely in their head with no documentation. This pattern appeared 14 times, none of which a traditional mechanism would have caught.

The Path Forward

18 priorities across four phases, sequenced by dependency.

Discovery did not end with a pile of findings. It produced a Stabilize-Then-Transform plan: fix the foundation first, then build on it, with prerequisite actions that downstream work depends on.

Phase 1 · Foundation

Stabilize the foundation

0-30 days · 6 actions

Knowledge documentation, data integrity, capacity assessment, engagement interventions, clinical documentation time, and communication cadences.

Phase 2 · Process

Repair the workflows

30-90 days · 5 actions

Admissions workflow, revenue cycle, handoff protocols, role clarity, and facility maintenance.

Phase 3 · Systems

Build durable systems

90-180 days · 5 actions

Training program, automated reporting, AI policy framework, growth strategy, and compliance audit.

Phase 4 · Innovation

Extend with automation

180+ days · 2 actions

AI-assisted clinical documentation and process automation for high-volume workflows, once the foundation can support them.

Methodology

This assessment is based on 52 confidential interviews, a 92.9% participation rate across all five organizational levels and 23 departments. The breadth of participation provides high confidence in the patterns identified. Severity ratings reflect interview-level assessments, with the large majority registering at moderate severity.

Findings are based on self-reported employee data. Where specific time estimates are referenced (hours spent on workarounds, days to complete processes), these come from interviewee estimates. Interview-based methodology carries inherent limitations, including potential response bias and self-selection effects. Where possible, triangulation across multiple interviews was used to strengthen confidence in the findings. Interviews were conducted by Dave, Privagent’s AI-powered conversational interviewer.

Next Steps

See what discovery would surface in your organization.

This behavioral health organization thought its eNPS told the full story. It did not. If you are making decisions based on aggregate metrics, surveys, or gut feel, there is a version of this case study waiting inside your own organization.

Schedule a Consultation

Tell us a little about your organization and we will be in touch.

Your information is confidential and is never shared. Findings are always anonymized. privagent.com/privacy