diff --git a/README.md b/README.md
index 02f8714..22fc7a0 100644
--- a/README.md
+++ b/README.md
@@ -95,6 +95,9 @@ The detail lives in dedicated documents:
 - [Total investment](docs/finance/total-investment.md): the full-vision
   roll-up (~4M to 7.7M EUR) and the ~1.5M to 2.5M EUR phased entry, with
   the basis of every block
+- [Revenue model](docs/finance/revenue-model.md): rough full-capacity
+  revenue (~3.2M to 5.9M EUR/yr), direct cost and margin per record, and
+  the levers and caveats behind the numbers
 
 ## Repository Layout
 
diff --git a/docs/finance/revenue-model.md b/docs/finance/revenue-model.md
new file mode 100644
index 0000000..949135e
--- /dev/null
+++ b/docs/finance/revenue-model.md
@@ -0,0 +1,101 @@
+# Revenue and Profit per Record: Rough Model
+
+A first-pass economic model at mature capacity. Every number here is an
+estimate from industry pricing norms and this repo's own cost research;
+nothing is market-validated yet. The point is the shape of the economics
+and which levers matter, not precision. FX: ~38 THB/EUR. Compiled July
+2026.
+
+## Pricing assumptions (what a pressing plant charges)
+
+Typical Western pressing-service prices run roughly 2.50 to 5.00+ EUR per
+12" record at volume for standard black, with premiums for color/splatter
+(+0.50 to 1.50), 180 g, and small runs; full-service with printed jacket
+lands around 4 to 8 EUR per unit. Per-title service fees come on top:
+mastering, lacquer cutting, stamper sets, test pressings, setup.
+
+Working assumptions for CNX at mature volume, blended across run sizes and
+standard/color mix:
+
+- **Blended pressing price: ~3.0 to 4.0 EUR per record** (pressing,
+  labels, inner sleeve, assembly into customer-supplied or locally-made
+  jacket; deluxe packaging passed through at cost plus margin)
+- **Per-title services: ~500 to 1,500 EUR per title** (mastering, cutting,
+  stampers, test pressings), at roughly 300 to 600 titles/yr at full
+  capacity
+- Dropshipping/fulfillment fees per parcel on top, not modeled here
+
+## Revenue at full capacity
+
+| Scenario | Records/yr | Pressing revenue | Title services | Total |
+|---|---|---|---|---|
+| Mature, low | 1.0M | ~3.0M EUR | ~0.15M EUR | **~3.2M EUR/yr** |
+| Mature, high | 1.5M | ~5.3M EUR | ~0.6M EUR | **~5.9M EUR/yr** |
+
+(Low: 1.0M x 3.0 EUR + 300 titles x 500 EUR. High: 1.5M x 3.5 EUR + 600
+titles x 1,000 EUR. A pure-volume plant runs at the low per-title end; a
+boutique-heavy mix pushes the title line up and the volume down.)
+
+## Direct cost per record (estimates)
+
+| Cost element | EUR/record | Basis |
+|---|---|---|
+| PVC compound | 0.25 to 0.40 | ~150 to 160 g effective virgin per record at 1.5 to 2.5 EUR/kg record-grade; regrind loop covers scrap |
+| Labels | 0.03 to 0.08 | printed pairs, volume |
+| Inner sleeve | 0.08 to 0.15 | poly-lined |
+| Jacket (standard) | 0.35 to 0.80 | local carton partners; pass-through-ish |
+| Shrink wrap | 0.02 to 0.03 | film |
+| Energy | 0.10 to 0.20 | ~1.5 to 2 kWh blended per record at grid/solar mix |
+| Production labor | 0.20 to 0.35 | 15 to 25 staff across shifts at Thai wage levels over 1.0 to 1.5M records |
+| Stamper amortization | 0.03 to 0.06 | in-house stamper over 1,000 to 2,000 pressings |
+| Maintenance, spares, consumables | 0.10 to 0.20 | estimate |
+| Reject allowance | 0.10 to 0.15 | 5 to 10% reject rate carried by good units |
+| **Total direct cost** | **~1.3 to 2.4** | |
+
+## Margin per record
+
+- Blended price ~3.0 to 4.0 EUR minus direct cost ~1.3 to 2.4 EUR gives a
+  **gross margin of roughly 1.2 to 1.8 EUR per record (~40 to 55%)**
+- Overheads on top: admin/sales, insurance, facility running costs, and
+  depreciation (~300k to 600k EUR/yr on the 4M to 7.7M full-vision capex
+  over 10 to 15 years), together roughly 0.4 to 0.7 EUR per record at full
+  volume
+
+Rough operating picture at full capacity: **EBITDA ~0.8 to 1.5M EUR/yr;
+operating profit after depreciation ~0.5 to 1.0M EUR/yr**, before tax.
+The BOI CIT exemptions (solar, automation, and any merit measures) shield a
+meaningful part of early-year profits; see [boi.md](../boi.md).
+
+Payback on the full-vision capex at these margins: **roughly 5 to 8
+years at mid assumptions** (the adverse corner, high capex meeting low
+profit, stretches past 15 years), materially better if the phased entry
+proves the market before the big blocks are bought.
+
+## What moves the model
+
+- **Utilization is everything**: the cost structure is mostly fixed, so
+  profit swings almost linearly with volume sold; 60% utilization at
+  mature fleet size roughly halves operating profit
+- **Mix**: color/splatter and boutique runs carry 20 to 50% price premiums
+  and suit the semi-automatic line; the title-services line (mastering,
+  cutting, stampers) has software-like margins once the rooms exist
+- **Regional position**: an Asia-Pacific full-service plant competes on
+  turnaround and freight against EU/US plants for regional labels; pricing
+  power there is unproven and is the single biggest assumption in this
+  model
+- **In-house galvanics** removes 150 to 400 EUR per title of external
+  stamper cost and days of turnaround, feeding both margin and the sales
+  story
+- **Energy**: solar plus heat recovery pushes the energy line toward the
+  low end and hedges Thai tariff rises
+
+## Honest caveats
+
+- No market validation yet: no signed customers, no confirmed regional
+  price points; the Making Vinyl trip and early conversations should test
+  the 3.0 to 4.0 EUR blended assumption before it anchors any financing
+  case
+- Demand risk is real: vinyl demand is cyclical and capacity has been
+  added globally; the model assumes the plant can sell what it presses
+- The reject allowance assumes the QC investment holds rejects at or below
+  industry norms; early production will run worse before it runs better