Research methodology

How we underwrite product capital

A transparent process for turning Amazon product ideas into investment decisions— with explicit hurdles, weighted factors, and kill switches.

Philosophy

Inventory is capital. Shipping a container is a capital call. Advertising is a tax on discovery. Big Sky ROI treats product selection as capital allocation under uncertainty, not as content marketing dressed up as research.

We optimize for risk-adjusted free cash generation from inventory—not vanity metrics, not “#1 BSR screenshots,” and not category hype.

The Product Conviction Score (0–100)

Each idea receives a weighted composite. Weights are desk defaults; vertical pages may rebalance when structure differs (e.g., consumables vs. durables).

  1. Demand durability (18%) — Structural vs. fad demand; seasonality; replacement cycle; category growth base rate.
  2. Competitive structure (18%) — Concentration, review barriers, brand power, private-label headroom, paid-only traffic dependency.
  3. Margin structure (16%) — Contribution after COGS, FBA, referral, returns, and realistic advertising intensity.
  4. Capital intensity (14%) — MOQ, cash conversion, storage exposure, tooling, and time capital is trapped.
  5. Operational risk (12%) — Supply concentration, QC, compliance, size-tier and hazmat traps.
  6. Brand & moat optionality (12%) — Path to differentiation, IP, bundles, consumables attach, pricing power.
  7. Exit liquidity (10%) — Ability to sell brand equity, liquidate, or pivot without catastrophic loss.

Action mapping

  • BUY — Score ≥78, base case clears hurdles, no red-flag kill switches unmitigated.
  • WATCH — Score 68–77, or clears hurdles only under improvable conditions (ACOS path, COGS negotiation).
  • HOLD — Score 55–67; only maintain or pilot-size; do not scale.
  • AVOID — Score <55 or structural failure (e.g., negative contribution at realistic ads).

Capital hurdles

Base case must clear both gates before full deployment:

  • Cycle ROI ≥ 45% on inventory capital (simplified as COGS × units for first order).
  • Net margin ≥ 18% after FBA, referral, ads, and other costs.

We also inspect annualized ROI against hold period. A 45% cycle return over 12 months is not the same trade as 45% over 3 months. Storage peak seasons and tariff scenarios are stress cases, not footnotes.

Scenario book

Every serious idea gets at least three cases:

  • Bull — Faster organic mix, better COGS, mild CPC.
  • Base — Realistic ACOS path and fee stack; our decision case.
  • Bear — Ad inflation +10pts, COGS +8–12%, slower sell-through.

If the bear case destroys capital and there is no operational hedge, the idea is WATCH at best—regardless of a pretty base sheet.

Position sizing

Rules of thumb (not rigid mandates):

  • High conviction (BUY ≥85) — Up to a defined share of deployable inventory capital after cash reserve.
  • Standard BUY — Pilot to proof of contribution, then scale in tranches.
  • WATCH — Research size only; no warehouse-filling POs.
  • Single-name risk — No single product should threaten firm survival if the bear case hits.

What we refuse to count as research

  • Unverified “revenue estimates” without fee and ad reality.
  • Category heat maps with no unit economics.
  • Competitor counts without structure (moat, reviews, brand).
  • Advice that ignores working capital and storage.

Tools on this site

The capital model is a simplified interactive underwriting sheet. The conviction board and research book show how scores and actions look in practice. Numbers on illustrative theses are educational—not live trade recommendations.

Research disclaimer: Methodology and examples are educational. Big Sky ROI does not guarantee results and is not a registered investment adviser for securities. Big Sky ROI, LLC · 30 N Gould St, Suite R, Sheridan, WY 82801.

Apply the framework to your next PO.

Run the capital model, then read the research book for worked examples.