08 · How we work · The method

Our method — how we actually work.

Five steps, shown as a swimlane: what you see, and what stays under the IP line. We put the shape and the discipline of the method in daylight — the engine itself stays under the hood. There is one step where we hold nothing back: step three, how every figure is labelled.

IP line — Below this line runs the engine. We show you its shape and its discipline — never its internals.
01
Step 1 · What you see

Source

We start at authority sources — SEC filings, CNAE sector data, public financial records. Nothing begins with a guess; every input has a documented origin before we touch it.

  • Public filings and official sector registers
  • A documented origin for every input
  • No black box at the front door
02
Step 2 · What you see

Quantify

We work the sector economics all the way down to EBITDA level — from broad sector figures to what actually lands on the margin line.

  • Sector economics carried down to EBITDA
  • Each figure tied back to its source
  • No leap you can’t retrace
04
Step 4 · What you see

Impact

We translate the sector picture into what it means for your margin — your baseline, your gap, your exposure — not a generic benchmark.

  • Sector figures translated to your P&L
  • Baseline and potential, side by side
  • What each gap is quietly costing you
05
Step 5 · What you see

Recommendation

You get one thing to fix first — the single binding constraint — and the route toward it. Not a forty-point list; the one move that unlocks the rest.

  • One binding constraint, named
  • The path from today to the fix
  • No lock-in, no dependency built in
Inside the quantify step · Monte Carlo

We don’t guess a number — we simulate the uncertainty.

Every uncertain driver is a range, not a single guess. We run thousands of scenarios across those ranges and report the percentiles: p50, the median — the honest expected outcome — and a ceiling, a high percentile. That is why our figures read "+153 bps (p50), ceiling +202 bps" — never "you will win X%".

p50 · medianThe honest expected outcome — half the simulated scenarios land above it, half below.
Ceiling · high percentileA favourable-but-plausible outcome, not a promise. Marked as a model result, never a guarantee.

Applied across: EBITDA uplift · Corridor / emigration · Asset safety · Deal scoring

We show the shape of the method — simulate, then report percentiles. The input distributions and driver rates stay under the IP line.

Want to see the method run on your numbers? A 20-minute conversation shows you where you sit — measured, labelled, no black box. No pitch. No obligation.

Talk to Gino about the method →