Field notes · Enterprise AI Notes from the Jagged Frontier
People Operations PE Operating Cost Structure Framework 09

The HR & Org Design Diagnostic

Soujanya Madhurapantula  ·  May 2026

People cost is the largest line item on most operating budgets. It is also the least systematically managed. Companies track revenue per employee as a headline metric but rarely dig into where the talent lifecycle actually leaks — which stage drives the most cost, which lever moves the needle fastest, and where AI can accelerate the work versus where it will make things worse faster.

This diagnostic maps the five stages of the talent lifecycle — recruit, onboard, perform, retain, exit — and asks the same question at each stage: where is the system breaking, and what does fixing it do to EBITDA? It connects directly to the operating toolkit and the product and GTM diagnostics. A company cannot scale its product or its revenue motion on a people system that is leaking.

People cost → lifecycle stage → operational lever → EBITDA impact

Start with the triage questions. If multiple stages fail, start with retention. Backfill cost at 1.5 to 2 times annual salary per exit is the fastest EBITDA lever available without adding headcount.

Triage

Where to start

Five questions, one per lifecycle stage. Any yes is a starting point. Multiple failures: start with retention.

Recruit
Is cost per hire rising or time to fill extending without a clear explanation?
Yes → start with Recruit
Onboard
Are new hires taking significantly longer than expected to reach full productivity?
Yes → start with Onboard
Perform
Are certain teams consistently outperforming others and do you know why?
No → start with Perform
Retain
Is voluntary attrition running above benchmark or concentrated in specific functions or managers?
Yes → start with Retain
Exit
Are exit interview themes synthesized into patterns or filed and forgotten?
Filed → start with Exit
Org design
Are spans of control too wide or narrow, or is role clarity breaking down at functional handoffs?
Yes → start with Perform
The talent lifecycle

Five stages, same question sequence

For each stage: where is the process breaking? What does it cost? Which lever fixes it first?

Stage 1: Recruit

Acquisition efficiency
  • What is cost per hire by role type and how has it trended over the last four quarters?
  • Where is time to fill longest — is it a sourcing problem or an approval bottleneck?
  • What is offer acceptance rate and why do candidates decline?
  • What is quality of hire at 90 days — are the right people actually being hired?
EBITDA impact
Lower cost per hire and shorter time to fill reduces recruiting spend and the opportunity cost of open headcount directly. In revenue roles, every week a quota-carrying seat is open is measurable lost revenue.

Stage 2: Onboard

Time to productivity
  • How long before a new hire reaches full productivity by role — and is that tracked or estimated?
  • Are 30/60/90 day milestones defined and completion tracked?
  • Which managers consistently ramp people fastest, and what do they do differently?
  • What is early attrition rate — people leaving before six months?
EBITDA impact
Faster ramp time means revenue-generating capacity sooner. A 30-day improvement in ramp time for a quota-carrying role is directly measurable in quota contribution. Early attrition means full recruiting cost with zero return.

Stage 3: Perform

Output and org design
  • What is the performance distribution — calibrated across functions or just a default bell curve?
  • Which teams consistently outperform and what is the common factor?
  • Is promotion velocity healthy or are high performers stalling and leaving?
  • What are the actual spans of control — and are there management layers that add process without adding decision quality?
EBITDA impact
Concentrating high performers and improving calibration raises output without adding headcount. Removing unnecessary management layers reduces cost while improving decision speed — typically 15 to 25% of that layer's total compensation in overhead.

Stage 4: Retain

Flight risk and backfill cost
  • What is voluntary attrition rate by function, manager, and tenure band?
  • What is the eNPS trend over the last four quarters — is it a leading indicator of what is coming?
  • Are flight risk signals being tracked before people decide to leave?
  • What is the fully loaded backfill cost per role — recruiting, onboarding, and lost productivity?
EBITDA impact
Reducing voluntary attrition by 5% eliminates significant backfill cost — typically 1.5 to 2 times annual salary per exit. Engagement leading indicators catch flight risk 60 to 90 days before a resignation, creating a window to intervene.

Stage 5: Exit

Signal capture and learning
  • Are exit interview themes synthesized across functions, managers, and tenure bands — or filed and never reviewed?
  • Where is attrition highest by tenure band — first year, second year, five-plus years?
  • Are there manager-level attrition patterns that have not been surfaced to leadership?
  • Do employees believe their feedback leads to visible action?
EBITDA impact
Exit data is the most honest signal in the organization. Synthesized properly, it surfaces systemic issues — management problems, compensation gaps, role clarity failures — before they compound into broader attrition waves.

Org design overlay

Structure and clarity
  • Does the org structure reflect the current strategy or historical decisions that were never revisited?
  • Is role clarity breaking down at the handoffs between functions?
  • Are there teams with spans of control so wide that managers cannot actually develop their people?
  • Are the right decisions being made at the right levels, or is everything escalating up?
EBITDA impact
Org design problems show up as slow decisions, high coordination cost, and frustrated high performers leaving. Fixing structure before adding headcount is almost always the higher-leverage move.
From diagnostic to action

Choosing the right lever

Each stage surfaces a problem. The fix is either a process or people intervention, or an AI application — but only once the process is defined.

Process, people, or tooling fix

  • Redesign the recruiting funnel and approval process
  • Build structured onboarding with milestone tracking by role
  • Establish a regular exit interview synthesis cadence
  • Define spans of control and role clarity frameworks explicitly
  • Set manager accountability for ramp time and retention metrics
  • Automation and AI applications
When the problem is a process or a people issue, fix the process first. An AI app built on an undefined onboarding model or inconsistent exit data produces broken outputs faster.

AI app candidate

  • Process and people fixes
  • Recruiting funnel redesign
  • Onboarding framework definition
  • Exit synthesis cadence
  • Automation and AI applications
When the right lever is automation, apply the Go/No-Go filter before building. Business value, data quality gate, and org readiness all need to pass. People data is sensitive — misuse of flight risk scores destroys trust faster than any process problem.
AI applications

Four apps across the talent lifecycle

These are the workflows where AI automation passes the Go/No-Go filter, assuming the underlying process and data are in place.

App Stage Goal Data required and gate
Resume screening and fit scorer Recruit Score inbound candidates against role requirements. Surface top fits and flag mismatches before recruiter review. Reduce time spent on manual screening. Requires a defined job requirements framework and historical hire quality data to calibrate against. Biased training data produces biased outputs — this gate is not optional.
Onboarding journey automator Onboard Personalize onboarding paths by role. Track milestone completion. Flag early risk signals — disengagement or stalled ramp — before the 90-day mark. Cannot automate a process that has not been defined. Requires a structured onboarding framework with defined milestones to exist first.
Exit interview synthesizer Exit Aggregate exit themes across functions, managers, and tenure bands. Surface systemic issues — management patterns, compensation gaps, role clarity failures — before they compound. Requires consistent exit interview capture. If exit data is incomplete, biased, or inconsistently collected, synthesis reflects those gaps and produces misleading signals.
Flight risk monitor Retain Score employee retention risk based on engagement signals, performance trends, manager feedback patterns, and tenure. Surface at-risk employees 60 to 90 days before a resignation. Requires connected HR, performance, and engagement data. Requires explicit organizational policy on how scores are used and who sees them. Misuse destroys trust faster than any attrition problem.
"Backfill cost at 1.5 to 2 times annual salary per exit means a 5% reduction in voluntary attrition often has a larger EBITDA impact than the next sales hire."
People cost → lifecycle stage → operational lever → EBITDA impact
Framework at a glance
HR & Org Design Diagnostic: one-page visual with lifecycle flow, cost bars, and lever routing
View the framework
Related
The Operating Toolkit: five functions, three tools, one playbook
Read the playbook