What Gigabite sells, to whom, at what price, with what economics, gated by what proof. The umbrella over the marketing, sales, operations, and financial plans.
Private-label foodservice sourcing run on evidence-backed truth: demand intake → attribute-level equivalency → landed-cost quote → order → payment, all on an append-only ledger.
Distributors and multi-unit operators pay a take rate on sourced GMV — aligned with provable savings — plus fintech and software lines as they mature.
$2M → $8M → $22M → $48M → $100M, gated at every stage. Seed of $6M funds Y1–Y2 to the trust and leverage gates.
The wedge is not "foodservice" — it is a specific buyer with a specific pain, where the savings report lands hardest.
| ICP ATTRIBUTE | TARGET PROFILE | WHY IT MATTERS |
|---|---|---|
| Annual disposables purchasing | $5M–$100M | Large enough for real savings, small enough to move fast |
| Private-label share of purchases | ≥20% | Equivalency and landed cost are live problems, not theory |
| Active disposable SKUs | 300+ | Enough surface for the extraction & matching machine |
| Sourcing tooling today | Email · XLSX · broker | No incumbent system of record to rip out |
| Decision maker | VP Supply Chain / Owner | One or two people can say yes to a pilot |
Landed-cost quotes with itemized components and evidence links — savings the buyer's CFO can audit, not a vendor's claim.
Attribute-level equivalency with explanations. "Is this the same cup?" answered with data, not thumbnails.
Net terms underwritten from actual payment behavior on the platform — better terms than a bank guessing from a bureau score.
| LINE | MECHANIC | LAUNCH | Y5 TARGET |
|---|---|---|---|
| Transaction | Take rate on sourced GMV, compressing 4.4% → 3.5% by design | Y1 | $70M |
| Embedded finance | Net terms, ~1.5% blended fee per cycle on financed volume (0% → 60% of GMV) | Y3 | $18M |
| Software & data | Ledger seats for buyer ops teams; supplier intelligence; category benchmarks | Y2 pilots | $12M |
Self-serve (Y2) means buyers staff their own review queues — the seat is the queue, so attach follows platform adoption. ~1.5 paid seats per active buyer at maturity → ~$9M by Y5.
Category benchmarks and supplier intelligence sold as subscriptions once gold tables are deep enough to price — ~$3M by Y5.
~$1M in Y2 as self-serve launches. The line grows with buyer count and data depth — no separate sales motion.
A Gigabite price is never re-guessed. The same figure hardens from a held estimate into a settled fact — like a renovation quote — and every step downstream reads it instead of recalculating. That mechanic is what makes the revenue lines above defensible: the cost is passed through honestly, and our margin sits on the receipt, in the open.
The quoted price is held 14 days; factory time is reserved against it.
Order placed; the price is re-checked against the market at that moment, then locked.
Goods arrive; estimate and actual are reconciled line by line; savings are rebated.
Every plan has load-bearing assumptions. These are ours, written down so the gates can test them instead of everyone discovering them later.
| # | ASSUMPTION | TESTED AT |
|---|---|---|
| A1 | Buyers will trust AI-proposed matches when every field links to evidence | G1 · Y1 |
| A2 | Avg buyer sustains $2.0–2.4M GMV/yr through the ledger | G1–G2 |
| A3 | Savings report → pilot converts ≥40%; pilot → paid ≥70% | G2 · Y2 |
| A4 | Auto-approval reaches 60% without quality-gate breaches | G2 · Y2 |
| A5 | Contribution margin ~60% holds as review labor shifts to agents | G3 · Y3 |
| A6 | Ledger payment history underwrites credit at losses below bank benchmarks | G3 · Y3 |
| A7 | A new product family onboards in <30 days by Y4 | G4 · Y4 |
| A8 | Recognized revenue tracks ~½ of (entry + exit) ARR each year; the burn plan absorbs the ARR-to-revenue lag | Financial plan |
| A9 | Logo retention ≥90% (→95% by Y5); NRR ≥110% from Y3 via category and terms attach | G2–G3 |
| A10 | New buyers ramp to steady-state GMV within ~2 quarters of landing | G1–G2 |