2026 CEO Diaries: Stop calling it “soft.” Price execution-risk like insurers do.
The problem we keep mislabeling
If you’re accountable for industrial performance and risk, you already know what actually hurts: shift-to-shift misalignment, fuzzy handoffs, supervisor span overload, onboarding debt, and knowledge quietly walking out the door. None of it sits neatly on a balance sheet — yet all of it erodes throughput, increases rework, and slows decisions.
We keep calling this “soft.”
It isn’t. It’s execution risk — and in 2026, it’s increasingly being treated as margin-at-risk.
I refer to this exposure as MAR-T: Margin-At-Risk from Talent & Execution.
The goal is not precision theater. The goal is to create a decision-useful range that leadership teams can govern — the same way they govern other material risks.
Definition
MAR-T (%)
The expected portion of gross margin exposed to talent- and execution-side coordination failures
(handoffs, alignment, role load, onboarding debt, cross-functional latency, ownership drift).
MAR-T ($)
Gross Margin ($) × MAR-T (%) at a chosen decision percentile.
MAR-T is not a new KPI.
It’s a way of pricing people-side execution risk so boards, CFOs, and operators can treat it with the same discipline as credit, inventory, or capacity risk.
Why range-based pricing matters
Execution risk does not behave like a single number.
It is:
Domain-specific (some friction costs far more than others)
Uneven over time (peaks during change, growth, or leadership transition)
Quiet until it compounds
That’s why MAR-T is expressed as a range (floor / most-likely / upper bound), not a point estimate.
Once leaders see execution this way, conversations shift:
From anecdotes → exposure
From blame → tradeoffs
From sprawling programs → targeted moves
That is the real unlock.
A CFO- and board-friendly way to price MAR-T
You already know how to price risk. This simply applies the same discipline to execution.
1) Map where value stalls
Choose one high-value value stream (product family, plant, work center).
At each handoff, capture three signals:
Wait time
Rework
Decision latency
2) Quantify frequency × severity
For 7–10 days, log how often stalls occur and what they cost:
Lost throughput
Changeover delay
Order aging / expediting
Scrap and rework
Translate impact into dollars using your gross margin %, conservatively.
3) Apply credibility weighting
Blend observations with practical priors (for example, high-mix environments tend to experience higher handoff loss than stable lines). This prevents one noisy week from driving the narrative.
4) Simulate, then bound
Using frequency and severity distributions, simulate multiple quarters to produce a P10–P90 MAR-T range, bounded with floors and ceilings.
You’re not trying to be perfect.
You’re trying to be honest, bounded, and usable.
5) Choose a decision percentile
Many teams plan against P50–P60: material enough to matter, credible enough to act on.
Illustration
Revenue: $200M
Gross margin: 25% → $50M
MAR-T range: 8%–22% of gross margin
$4M–$11M of margin exposure
At this point, two questions matter:
Is this exposure acceptable — or clarifying?
If you can’t fix it all this quarter, where will you deliberately move the needle?
The 90-day play (no headcount)
This is where execution pricing becomes execution progress.
Weeks 1–2 — Instrument lightly and align
Walk the value stream; mark red-dot handoffs
Add a 15-minute daily huddle with a single “constraints” line
Log frequency × severity on 3–5 hotspots
Weeks 3–4 — Choose three moves
Score candidate actions by Value / Certainty / Effort.
Typical moves include:
Standard work on a high-variance step
Shift-change SOPs with a five-minute checklist
Knowledge capture before role transitions
Weeks 5–12 — Execute and show the basis points
One owner and date per move
Track 2–3 dollar-tied KPIs
Publish a weekly mini P&L: basis points gained + control plan
Week 13 — Lock in
Compare MAR-T signals to baseline
Document controls
Select the next value stream
This cadence often recovers tens of basis points per quarter — small individually, meaningful in aggregate, and confidence-building for the next cycle.
Where agent functionality fits (and where it doesn’t)
The hardest part of execution improvement isn’t identifying issues.
It’s preventing drift.
After the baseline and first 90-day cycle, organizations tend to slide back because:
Ownership changes
Priorities collide
Handoffs decay quietly
This is where agentic support layers add value — not by replacing leaders, but by keeping the execution radar current.
Properly designed agents:
Monitor a small set of MAR-T signals
Flag drift (“what changed?”)
Trigger lightweight check-ins or micro-pulses
Preserve continuity across leadership transitions
They don’t run the business.
They protect the gains and reduce the cognitive load of vigilance.
What not to do
Don’t boil the ocean — one clean dataset beats a massive spreadsheet
Don’t chase tech before governance — tools accelerate clarity, not accountability
Don’t accept slides without dollar-tied signals
Don’t run this by committee — appoint one owner per cycle
Where the numbers come from (and their limits)
MAR-T ranges are derived from simulation bounded by observed execution patterns in mid-market, capital-intensive environments with mix complexity and limited managerial bandwidth.
Your range will depend on:
Product and service mix
Asset intensity
Team design and turnover
Operating-system maturity
Treat the envelope as a starting hypothesis, not a verdict.
This approach does not replace safety, quality, or financial risk models.
It adds a missing layer: priced execution risk.
Where Navetra™ fits
All of this can be done manually — but it’s slow and hard to sustain.
Navetra supports teams by:
Establishing a credible MAR-T range using structured organizational intelligence plus financial anchors
Ranking priority execution domains and surfacing three moves for the next 90 days
Activating a lightweight agent layer that monitors drift and preserves gains
No enterprise rollout to start.
No added headcount.
Just priced execution risk, visible deltas, and governance that holds.
Share this article:
