Professional Services · 32 Employees

Two Critical Risks Inside a Firm the Partners Thought They Knew

A 32-person CPA firm was growing on the strength of a few indispensable people. Confidential interviews across all 9 departments surfaced two existential risks, and 21 shadow systems holding daily operations together, that leadership had never heard about.

Schedule a Consultation See the Findings

Participation Rate

97%

31 of 32 employees spoke candidly, across all 9 departments and 5 role levels.

Critical risks surfaced

Shadow systems found

The Engagement

What we assessed.

Firm Type

CPA Firm

Professional services

Employees

Across all functions

Interviewed

97% participation rate

Departments

Every department covered

Role Levels

Individual contributor to executive

Friction Points

Occurrences surfaced in interviews

Reports Delivered

Structured diagnostic reports

Prioritized Actions

Sequenced across four phases

The Founder Blind Spot

The partners did not feel uninformed. They felt clear.

That feeling was the product, not the evidence. As the firm grew, the founding partners gradually lost access to ground-level operational truth, and the organization’s own habits kept it that way. Managers softened bad news. Employees stopped raising recurring problems. The leadership team avoided the hard conversation about its own decision-making. The result was a confident, unchallenged, and incomplete picture of the firm.

Thirty-one confidential interviews measured the distance between that picture and the reality the staff lived every day. The partners believed systems worked, decisions got made, and key people were assets. Employees described unreliable tools, decisions that stalled for over a year, and a handful of individuals whose departure would cost the firm “weeks, maybe months of pain.”

Underneath nearly every problem sat the same dynamic. The firm had grown through the dedication and expertise of a few people, and that same dependence had quietly become its largest structural risk. Their effort kept the firm running. It also masked the gaps leadership needed to see. The heroism was a symptom of missing infrastructure, not a failing of the people carrying it. A similar pattern surfaced in a behavioral health organizational discovery engagement, where individual sacrifice concealed systemic fragility behind a healthy-looking score.

What leadership believed

What employees reported

Leadership believedThe practice management system worked.

Employees reportedIt was always out of date and unreliable. 21 shadow systems had been built to compensate.

Leadership believedDecisions got made.

Employees reportedStrategic initiatives stalled for over a year. A three-year employee still did not know who to ask.

Leadership believedKey personnel were assets.

Employees reportedThey were existential dependencies. The firm faced “weeks, maybe months of pain” if two people left.

Leadership believedOnboarding brought new hires up to speed.

Employees reportedNew hires were “set up to fail.” One first complex return had to be almost entirely redone.

“It’s all up here... and that’s a problem.”
Operations Manager, on the knowledge that lived only in memory

Key Findings

Two critical risks and a parallel firm running in the background.

Critical Severity

A partner decision-making vacuum stalled the firm's strategic decisions.

Neither founding partner wanted to force the issue when they disagreed, and the third partner was hesitant to act as tiebreaker. The result was indefinite deferral on strategic decisions. Employees across every role level described unclear escalation paths and decisions that simply did not get made.

One executive reported that strategic initiatives stalled for over a year waiting for partner alignment. A purchase of a new practice management system had been delayed more than a year. A three-year employee said they still sometimes got it wrong about who to ask. Nearly every operational challenge in the firm traced back to this single governance gap.

Why this was invisible

Employees will not attribute systemic delays to partner dynamics on a written survey. Surfacing the root cause required conversational depth, follow-up questioning, and a confidentiality guarantee strong enough that staff felt safe describing leadership's own decision-making.

Critical Severity

Essential institutional knowledge lived in two people's heads, undocumented.

Critical operational knowledge existed primarily in the memories of two key individuals, with no documentation and no backup. Knowledge concentration created bottlenecks, prevented effective onboarding, and left the firm existentially dependent on a small number of people who could never afford to be unavailable.

One manager openly acknowledged the firm would face “weeks, maybe months of pain” if they departed. Another maintained a 47-tab spreadsheet as an unofficial tracking system because the official tools could not be trusted. A senior manager kept a personal cloud archive holding eight years of client notes that no one else knew existed, a resourceful adaptation that also meant the knowledge sat entirely outside the firm’s reach.

Why this was invisible

People do not self-report as single points of failure on a checkbox survey. They describe it as "how things work here." This level of candor required the confidentiality guarantee and conversational rapport that Dave, Privagent's interviewer, creates.

High Severity

System unreliability drove an entire parallel infrastructure of workarounds.

The firm’s practice management system was unreliable enough that staff had built a parallel infrastructure of personal spreadsheets, email folders, and undocumented databases to get their work done. These were smart adaptations to tools that were failing them, but together they meant the firm effectively ran on unofficial systems. The pattern appeared as 21 independent instances across all 9 departments.

One manager maintained a vendor contact database of more than 200 entries entirely outside official systems. HR kept personal spreadsheets for recruiting, onboarding, and benefits because the payroll system did not provide the functionality they needed. Essential data was fragmented across personal computers with no synchronization, no audit trail, and no backup.

Why this was invisible

Employees view their workarounds as normal, not as organizational risk, so they never flag them on a survey. Only interviews that explored how people actually do their work, rather than how they are supposed to, could surface a parallel infrastructure this large.

Deliverables

Seven structured reports. One sequenced action plan.

Executive Summary

Top three issues, five immediate actions, and a confidence assessment for leadership consumption.

Leadership Report

Cross-functional organizational assessment, scaling challenges, and 11 prioritized actions.

Operations Deep Dive

Process efficiency audit, handoff analysis, capacity assessment, and throughput bottleneck identification.

Change Readiness Assessment

Readiness score of 5.5 out of 10, resistance signals, fatigue assessment, and leadership bandwidth analysis.

AI & Automation Readiness

Automation opportunity map, candidate analysis, risk assessment, and a phased implementation roadmap.

Consolidated Action Plan

18 sequenced actions with dependency mapping, success metrics, and owner assignments.

Follow-Up Agenda

12 clarifying questions, a data gap analysis, and a 90-day critical path visualization for leadership debrief.

Why This Matters

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

Shadow operations only surface when you ask how the work actually gets done.

A consultant asks about processes. The AI-powered interview asked employees how they really do their work, and a parallel infrastructure emerged: a 47-tab spreadsheet, a 200-entry vendor database, an eight-year personal cloud archive. 21 instances across all 9 departments, none of which a process review would have found.

Employees will not name partner dynamics as the root cause on a survey.

The decision vacuum traced nearly every operational problem back to a governance gap. That attribution only emerged through adaptive follow-up under a confidentiality guarantee strong enough that staff would describe leadership itself.

Single points of failure are described as "how things work," never as risk.

Two people held knowledge that would take weeks or months to rebuild. That candor surfaced because employees spoke to Dave, not to a person who might carry their words back into the firm.

The numbers came from the people living inside the dysfunction.

35 to 44 hours a month lost to workarounds, 30 to 45 minutes of manual entry per client, 15 to 20 hours a month of integration savings: employees calculated their own inefficiencies. Sample-based consulting produces directional estimates. Full-organization discovery produces specifics.

The Path Forward

18 priorities across four phases, sequenced by dependency.

Discovery did not end with a pile of findings. It produced a sequenced plan with four prerequisite actions: governance clarity preceded every strategic initiative, knowledge capture preceded cross-training and succession, and a shadow-system inventory preceded any technology decision.

Phase 1 · Immediate

Close the governance and knowledge gaps

0-30 days · 5 actions

Partner decision-making framework, emergency knowledge capture, tiered review authority, shadow-system inventory, and critical-role backups.

Phase 2 · Short-Term

Build the operational foundation

30-90 days · 6 actions

Structured onboarding, data quality remediation, role clarification, HR capacity support, handoff protocols, and a technical training curriculum.

Phase 3 · Medium-Term

Rationalize systems and capacity

90-180 days · 5 actions

Technology rationalization roadmap, workload visibility and capacity planning, AI usage policy, client communication standards, and rework tracking.

Phase 4 · Long-Term

Institutionalize resilience

180+ days · 2 actions

Leadership succession framework and a continuous improvement process, once the foundation can sustain them.

Methodology

This assessment is based on 31 confidential interviews, a 97% participation rate (31 of 32) across all 9 departments and 5 role levels, with tenure ranging from under two years to over five years. Core themes repeated independently across departments and tenure bands, with employees who had never discussed issues with one another reporting identical patterns. That consistency indicates structural rather than perception-based issues and provides high confidence in the findings.

Findings are based on self-reported employee data. Quantified impacts, such as hours lost to workarounds or minutes of manual entry per client, come from interviewees’ own 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. Interviews were conducted by Dave, Privagent’s AI-powered conversational interviewer.

Next Steps

See what discovery would surface in your firm.

These partners thought they had a clear view of their own business. Thirty-one confidential interviews showed otherwise. If you are running a growing firm on the expertise of a few indispensable people, there is a version of this case study waiting inside your organization.

Schedule a Consultation

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