The proof

What a Profit Leak Audit actually looks like.

A section-by-section walkthrough of a real ROME audit on a sample restaurant — the same engine, the same screens, the same math your restaurant would get.

Sample Sample audit of a fictional restaurant built from realistic synthetic data. Your audit uses your real numbers.

The full plan, in one ladder

$11,961/mo operational fixes + $7,276/mo pricing & growth plays = $19,237/mo full plan

Westport flagship. The headline below is the operational rung; the ranked plan spans the whole ladder.

Section 1

The headline: what the audit found

Net recoverable from the core operational fixes

$11,961/mo

Westport flagship · operational fixes only · 92-day window scaled to 30 days

$143,532 per year · 5.8% of sales · the sum of 6 quantified leaks

One sample restaurant, one location, one recent window of its own POS data — and $11,961 a month quietly leaking through pricing, prep, pours, and scheduling. Every dollar of it is itemized below.

How we calculated this →

Data sources · Every input is a real row in the sample restaurant’s point-of-sale exports — checks, order lines, menu costs, labor shifts — over the recent 92-day reporting window. Nothing is modeled from industry averages.

Monthly basis · Leaks are measured over the window and scaled to a 30-day month (window total × 30 ÷ 92 days) — a monthly average, not a calendar-month actual. The headline is the sum of the quantified leaks.

Honesty · These are estimates of operational opportunity, not guaranteed recovery — what you capture depends on the actions you take. Every figure in the product carries its math one click away.

Section 2

The biggest leak

Fajita platters underpriced for protein cost

$8,214/mo

The fajita family runs a 54% margin — the worst-margin high-volume group on the menu at 8,397 sold. A $3 reposition across the family recovers margin on the highest-protein-cost plates you sell.

The math · $3.00 reposition × 2,738 fajita plates/mo = $8,214/mo

This is what a leak looks like in ROME: a named problem, the dollars it costs each month, and the exact arithmetic behind the figure — never a vague “you should tighten up.”

Section 3

Menu engineering: every item earns its place, or doesn't

Star — high volume, high margin

Enchiladas Verdes

1,906 sold this window · 75% margin

A genuine star — protect it and feature it; this is the plate the menu is built around.

Plowhorse — high volume, thin margin

Fajitas Steak

2,773 sold this window · 51% margin

A classic plowhorse — sells constantly but earns too little per plate — the classic candidate for a careful reprice or re-cost.

Puzzle — high margin, low volume

Mole Poblano

470 sold this window · 75% margin

A puzzle — earns well but nobody orders it — promote it, reposition it, or let the menu make the case.

Dog — low volume, low margin

Tableside Guacamole

931 sold this window · 51% margin

A dog — takes up prep, inventory, and menu space without paying rent — cut it or rework it.

ROME classifies the whole menu this way from real sales and real plate costs — with a concrete promote/reprice/rework call for each item, not just a chart.

Section 4

Labor: sales per labor hour, day by day

Mon
$31/hr
Tue
$32/hr
Wed
$39/hr
Thu
$45/hr
Fri
$51/hr
Sat
$53/hr
Sun
$41/hr

Sales generated per labor hour, by weekday, over the reporting window.

The mismatch is visible at a glance: Saturdays generate $53 of sales per labor hour while Mondays manage $31 — yet the schedule barely changes. That gap is where the staffing findings (and their dollars) come from.

Section 5

The ranked action plan — top 5 of 15 moves

Shown: the 5 highest-impact moves by dollar impact — these five: $15,606/mo of the $19,237/mo full plan (operational fixes + pricing & growth, all 15 moves). The rest are in the full plan below.

1

Fajita platters underpriced for protein cost

Moderate · High confidence · Menu

$8,214/mo
2

Nudge prices on your high-volume "Plowhorses" (Fajitas Steak, Fajitas Pollo, Taco al Pastor)

Quick win · Medium confidence · Pricing

$2,753/mo
3

Launch a win-back to your 559 dormant customers

Moderate · Medium confidence · Customer

$1,920/mo
4

Prompt Tableside Guacamole on Taco al Pastor orders (attach rate 10%)

Quick win · Medium confidence · Service

$1,406/mo
5

Perishable over-prep & slow-mover spoilage

Moderate · High confidence · Purchasing

$1,313/mo

Full plan: 15 moves worth $19,237/mo = $11,961/mo operational fixes + $7,276/mo pricing & growth plays Westport flagship · full plan

Not a dashboard to interpret — a to-do list ranked by payoff, each move carrying its effort, confidence, and the evidence behind it.

Section 6

The 30-day roadmap: quick wins first

Week 1 — quick wins

  • House margarita pour cost at 44%$848/mo
  • Tableside guacamole underpriced for labor$607/mo
  • Dinner overstaffed on Mon/Tue$504/mo
  • Avoidable weekend kitchen overtime$475/mo

Weeks 2–4 — bigger moves

  • Fajita platters underpriced for protein cost$8,214/mo
  • Perishable over-prep & slow-mover spoilage$1,313/mo

Sequenced so the easy money lands first — the Week 1 fixes fund the patience for the bigger moves.

Section 7

What it takes: four exports, nothing else

  • Sales / orders
  • Order line items
  • Menu & plate costs
  • Labor shifts

The four CSV exports your point-of-sale system already produces. Column-by-column guide: the SpotOn export guide.

No new hardware. No new staff logins. Nothing changes about how your restaurant runs.

Section 8

What you receive

Your Profit Leak Audit

Every leak named, dollarized, and backed by its math — like the excerpts above, on your real numbers.

Live dashboards

Explore every check, item, daypart, and location — every total traceable to the transactions behind it.

The daily briefing

A plain-language read on your restaurant — the handful of things worth your attention, before any chart.

Ask ROME

A conversational AI advisor grounded in your data — ask anything about your numbers and get the why behind it.

Sample Sample audit of a fictional restaurant built from realistic synthetic data. Your audit uses your real numbers.