Grand Slam Market Sizing: Six Opportunities
Prepared 2026-04-27. Synthesis is a human activity — this document surfaces data and estimates. Conclusions belong to Bliss and Dustin.
Section 1: Per-Opportunity Market Sizing
1. Compliance Wedge (UFLPA Screening, Export Controls, Supplier Onboarding SaaS)
TAM: $4.5–6.5B — The global trade compliance software market reached $1.95B in 2025 and is projected to hit $3.45B by 2030 at 12% CAGR (GlobeNewsWire/Research & Markets, Jan 2026). The adjacent supply chain risk management software market is estimated at $4.5B in 2025, growing to $9.2B by 2030 at 15.3% CAGR (Mordor Intelligence, 2025). Gartner projects that supply chain management software with agentic AI will grow from under $2B in 2025 to $53B by 2030, though this broader category includes non-compliance functions.
SAM: $500M–$1.5B — The semiconductor-specific slice. The global semiconductor industry hit $796B in 2025 revenue (SIA, Feb 2026), but the addressable customer base for compliance tooling is narrower: approximately 200–400 firms with $100M–$10B in semiconductor revenue, plus IDMs and OSATs. Compliance spend at these mid-tier firms runs $1–3M annually, with only 10–15% going to software today — the rest flows to outside counsel at $500–$1,200/hour and in-house FTEs. This suggests the current software wallet is $200–600M across ~300 firms, but the displacement opportunity (automating work currently done by outside counsel and manual labor) expands the addressable spend to $1B+. The key insight for product builders: you’re not competing for the existing software budget, you’re competing for the outside counsel budget.
SOM (Yr 1–3): $3–10M — 15–40 mid-tier fabless customers at $200–500K ACV. This is grounded in the enforcement pain: Applied Materials paid $252.5M to settle export violations in Feb 2026 (Supply Chain Dive); Cadence paid $140M+ for unlawful EDA exports in July 2025 (DOJ Press Release). UFLPA enforcement is surging: CBP stopped ~7,325 shipments in FY2025, a 50%+ YoY increase, with a single detention costing an average of ~$810K (CBP UFLPA Statistics; Sourcing Journal/Oritain). For a compliance officer at a mid-tier fabless firm, the ROI math on a $200–500K compliance platform against the backdrop of $100M+ penalties and $800K+ detention costs is straightforward.
Key comparables:
| Company | Valuation/Revenue | Relevance |
|---|---|---|
| Altana AI | $1B val, $37.5M rev (2024), $322M raised (Crunchbase; Latka) | Closest competitor. Multi-industry SC intelligence; serves CBP directly. Broad focus is both strength (data) and weakness (not semi-specialized) |
| Interos | ~$1B val, $310M+ raised, $39.7M rev (2024) (TechCrunch; Latka) | Government/defense-focused SC risk. 500 customers |
| Descartes Visual Compliance | Acquired by Descartes Systems Group | Market leader in restricted party screening. Launched AI-assisted screening Aug 2025 |
| Exiger | Private, defense SCRM platform | Used by DoD for NDAA compliance. McLean, VA-based |
Confidence: High. TAM is backed by multiple converging analyst reports. Enforcement data (penalty amounts, detention volumes) is public record from DOJ, BIS, and CBP. The main uncertainty is SAM — how many of those 200–400 mid-tier firms will actually buy, and at what ACV. That requires primary research.
Gap flag: No confirmed willingness-to-pay data from semiconductor compliance officers. The $200–500K ACV assumption needs validation through 10–15 buyer interviews.
2. Insurance (Parametric Supply Chain Policies, MGA/Reinsurance Layer)
TAM: $19–21B — The global parametric insurance market reached approximately $19–21B in 2025 and is projected to grow to $48–64B by 2035 at 10–12% CAGR (GM Insights; Market Research Future). The broader supply chain insurance market (CBI, cargo, political risk) adds substantially more. IUMI reported global cargo insurance premiums of $22.1B in 2023.
SAM: $1–3B — The semiconductor-specific insurance opportunity. This estimate is built from three data points. First, the demand signal: Lloyd’s and WTW surveyed 100+ semiconductor risk professionals and found that 88% viewed supply chain insurance as “mission-critical or necessary,” yet 81% cited a lack of access to insurance solutions as a top challenge (Lloyd’s Futureset, “Loose Connections,” March 2023). Second, the loss exposure: the 2020–21 chip shortage cost automakers alone an estimated $210B in lost revenue and 9.5M units of lost production (AlixPartners/CNBC, Sept 2021; S&P Global Mobility). Third, the protection gap: new fab construction takes 3–4 years (SEMI/SIA; UltraFacility), while typical business interruption policies indemnify for only ~2 years — a persistent, monetizable gap that Lloyd’s specifically calls out. The Lloyd’s report also documents at least one semiconductor company already buying a parametric earthquake policy triggered by magnitude and distance from a key supplier fab (Lloyd’s Part 3: Insurance Innovation Opportunities).
SOM (Yr 1–3): $5–20M GWP — Initial parametric policies tied to measurable triggers (earthquake magnitude near TSMC fabs, government export control designations, port closure durations). The MGA model — where Project TBD underwrites using proprietary supply chain topology data and lays off risk to reinsurers — is proven in adjacent markets. Coalition, the cyber MGA, built to $3.5B valuation and $800M raised by using proprietary scanning data to underwrite risks that traditional insurers couldn’t price (CB Insights). In batteries, Munich Re is already providing 15-year warranty reinsurance for Hithium based on TWAICE monitoring data (Munich Re, via Hithium press).
The critical question is whether semiconductor supply chain risk can be structured into insurable events. Idiosyncratic events (a specific fab fire, an earthquake, a single entity being added to the BIS Entity List) are clearly insurable — they’re diversifiable and trigger-able. Systemic shortages (the 2021 chip crunch hitting all participants simultaneously) are harder because they violate insurance’s diversification requirement. The honest framing: the parametric opportunity is real for event-based triggers, but “semiconductor shortage insurance” as a category probably doesn’t work. The product must be designed around specific, measurable, diversifiable triggers — not broad industry conditions.
Key comparables:
| Company | Valuation/Revenue | Relevance |
|---|---|---|
| Coalition | $3.5B val, $800M raised (CB Insights) | Best analog for “data-advantaged MGA.” Used proprietary cyber scanning data to underwrite risks incumbents couldn’t price |
| Descartes Underwriting | Private, parametric specialist | Climate/NatCat parametric insurance. Relevant trigger design experience |
| FloodFlash | $15M raised | Parametric flood. Small but demonstrates trigger-based payout model |
| Munich Re / TWAICE | Partnership | Delivered performance warranty insurance for Li-ion batteries using monitoring/analytics — closest analog to “digital twin enables risk pricing” |
Confidence: Medium. The Lloyd’s/WTW survey is authoritative and semiconductor-specific. The protection gap (fab rebuild time vs. BI indemnity period) is structurally real. But no semiconductor-specific parametric product exists at scale beyond the single case Lloyd’s documents. The SAM is an estimate derived from loss exposure and survey data, not observed premium volume.
Gap flag: Zero conversations with reinsurers or specialty insurers. The structural insurability question (systemic vs. idiosyncratic risk) needs direct validation with 3+ specialty insurers (Munich Re, Zurich, Descartes Underwriting).
3. Trading/Benchmarks (Semiconductor Price Indices, Hedging Instruments)
TAM: $3–5B — The global price reporting agency (PRA) and commodity benchmark market. S&P Global Commodity Insights (formerly Platts), the market leader, generates an estimated ~$1B+ in annual revenue from its commodity pricing and intelligence business (Steel-Eye PRA analysis). Argus Media was valued at ~$1.4B when General Atlantic acquired a majority stake in 2016 and has grown substantially since. The broader financial data and indices market is $40B+ globally. PRAs enjoy high operating margins and premium valuations because benchmark pricing becomes embedded in contracts and derivatives — creating structural lock-in.
SAM: $200M–$800M — This is the most speculative estimate in this document. No semiconductor price index or hedging product exists today. The SAM is derived by analogy: if the oil PRA market (~$3–5B) serves an industry with ~$3T in crude revenue, and semiconductors generate ~$800B in revenue (~25% of oil), a proportional semiconductor PRA market would be ~$750M–$1.25B. However, this analogy assumes comparable levels of price standardization and financial intermediation — assumptions that may not hold. The more conservative approach: size only the lead-time and broker-premium benchmark market for commodity-like chip segments (DRAM, NAND, trailing-edge MCUs/analog), which are the only segments with enough homogeneity to benchmark. DRAM prices have historically surged +88% or more in up-cycles and collapsed 50%+ in down-cycles (The Register, Dec 2025; Sourceability), and analog/MCU lead times exploded from ~12 weeks to 50+ weeks during the 2021 shortage (IEEE Spectrum; Bain & Company), creating enormous value at risk for buyers who had no way to hedge.
SOM (Yr 1–3): $1–5M — Subscription revenue from lead-time volatility indices and broker-premium benchmarks for trailing-edge chips. This is explicitly a benchmark-authority play, not financial instruments. The sequencing matters: BMI (Benchmark Mineral Intelligence) took ~5 years from founding (2014) to IOSCO price assurance (2019) and ~11 years to ICE futures contracts (BMI/Wikipedia). S&P Global Platts was founded in 1909; NYMEX oil futures launched in 1983. Benchmark authority must precede derivative products. The Day-1 product is a reference index designed to be embedded in supply contracts, not a trading platform.
The structural challenge is non-fungibility. Oil supports price indexes, futures, and insurance because a barrel of WTI crude is interchangeable with another barrel of WTI crude. A 5nm TSMC logic die is not interchangeable with a 28nm GlobalFoundries MCU. Prior attempts to financialize semiconductors have failed: Enron launched DRAM forward contracts in 2001 using its own capital, offering guaranteed prices to OEMs and suppliers, but the program died with Enron’s collapse before its viability could be tested (EE Times, 2001; EDN). Earlier DRAM futures attempts in the late 1980s–90s also failed due to the same fungibility barrier. The question is whether a benchmark can be designed for narrow, commodity-like segments (trailing-edge MCUs, commodity DRAM) rather than “semiconductors” broadly — a narrower but potentially viable approach.
Key comparables:
| Company | Valuation/Revenue | Relevance |
|---|---|---|
| BMI (Benchmark Mineral Intelligence) | $611M raised, Spectrum Equity invested (PitchBook; Wikipedia) | Closest analog: built PRA for battery minerals from zero. IOSCO-assured prices, ICE futures. But batteries are more commoditized than most chips |
| S&P Global Commodity Insights (Platts) | ~$1B+ revenue, part of $155B market cap S&P Global | Gold standard PRA. 100+ year history. Shows endpoint value |
| Supplyframe | Acquired by Siemens for $700M, ~$70M revenue (Siemens Press, 2021) | Electronic component intelligence platform. 10x revenue acquisition multiple validates the data layer, though Supplyframe serves design engineers, not financial buyers |
Confidence: Low. This is the most speculative opportunity. The TAM for PRAs is real but the semiconductor-specific SAM rests on an analogy that may not translate. Every prior attempt at semiconductor price indexing has failed. BMI’s trajectory is instructive but lithium/cobalt/nickel are significantly more commoditized than logic chips.
Gap flag: No validated demand from any buyer (commodity desks, hedge funds, corporate treasury, procurement). No research on why prior semiconductor index attempts specifically failed. Need: interview 5+ auto-OEM procurement officers, research all historical attempts, consult BMI’s Simon Moores on minimum conditions for PRA viability.
4. Government Intelligence (Critical Infrastructure Intelligence, CHIPS Act Compliance, Defense SC Mapping)
TAM: $10–20B — The government/defense intelligence and supply chain analytics market. The broader critical infrastructure protection market is estimated at $138–154B in 2025 (MarketsandMarkets), but the relevant slice is the software intelligence and analytics portion serving defense and national security customers. Palantir generated $4.48B in total revenue in FY2025, with approximately $2B from U.S. government contracts alone (Semafor, Aug 2025; CNBC). Mastercard acquired Recorded Future, the world’s largest threat intelligence company (1,900+ clients across 75 countries, including 45 governments), for $2.65B in Dec 2024 (Mastercard Investor Relations). These exits demonstrate that government intelligence platforms command premium valuations when they become embedded in decision-making workflows.
SAM: $2–5B — Semiconductor-specific defense and government intelligence. Three demand vectors converge here. First, CHIPS Act compliance: $52.7B in subsidies flow to 20+ award recipients who face mandatory milestone reporting, expansion restrictions, and domestic content disclosures — 24 of 161 milestones reported as of July 2025 (GAO-26-107882; Commerce OIG). Second, defense supply chain mapping: the DoD needs to understand semiconductor dependencies across the defense industrial base, evidenced by Palantir’s $10B Army data contract (CNBC) and growing Project Maven deployment (20,000+ users, DefenseScoop). Third, export control intelligence: BIS designated 140 new entities to the Entity List in December 2024 alone — all involved with advanced semiconductor production or supporting China’s Military-Civil Fusion strategy (BIS/WilmerHale). The pace of regulatory change creates persistent demand for real-time intelligence.
SOM (Yr 1–3): $2–8M — CHIPS Act compliance contracts (20+ recipients need reporting tools) and 1–2 defense/intelligence agency pilot engagements. The government sales cycle is long (12–24 months) and requires infrastructure most startups lack — FedRAMP authorization for cloud products, potentially IL4+ certification for classified work, physical presence in DC, and personnel with active security clearances. Realistically, government revenue is more likely a Year 3–5 story unless Project TBD partners with a prime contractor (Booz Allen, SAIC, Leidos) that can white-label the technology under their existing vehicle contracts. The CHIPS Act compliance track has a shorter sales cycle because recipients are commercial entities, not government agencies — this is the likelier near-term entry point.
Key comparables:
| Company | Valuation/Revenue | Relevance |
|---|---|---|
| Palantir | $4.48B total rev FY2025, ~$2B gov (Semafor) | Dominant government data platform. $13.7B+ in multi-year contract ceilings. The incumbent to beat |
| Recorded Future | Acquired for $2.65B by Mastercard (Mastercard IR) | Threat intelligence. 1,900+ clients, 45 governments. Shows acquisition premium for government-embedded intelligence |
| Exiger | Private, defense SCRM | Supply chain risk management platform used by DoD. McLean, VA |
| Interos | ~$1B val, $310M+ raised (TechCrunch) | Operational relationship intelligence. Strong gov/defense presence. Kleiner Perkins-backed |
Confidence: Medium-high. Government demand is funded and mandated by law (CHIPS Act, export controls). Palantir and Recorded Future exits demonstrate market reality. But the semiconductor-specific government intelligence market is not broken out in any public reporting — the $2–5B SAM is inferred from comparable contract values, not directly measured. The execution barrier (clearances, FedRAMP, long cycles) is the main risk to SOM, not the underlying demand.
Gap flag: No conversations with government buyers. Need: CHIPS Act program office, 1–2 defense primes, congressional staff. FedRAMP timeline assessment is critical before committing resources.
5. Marketplace (Supplier/Buyer Matching, Excess Inventory, Capacity Intermediation)
TAM: $200–418B — The global electronic components distribution market was valued at $200–418B in 2025 depending on scope definitions (Mordor Intelligence; Research Nester). This is the largest TAM of any opportunity here, but also the most misleading — the vast majority of this value flows through incumbent distributors (Arrow Electronics ~$37B revenue, Avnet ~$24B) via established relationships that a startup cannot displace head-on. The relevant wedge is the long-tail: surplus/excess inventory, shortage-driven spot buying, and capacity pre-reservation — segments where incumbents are weakest and information asymmetry is highest.
SAM: $5–15B — The excess and surplus electronic component market, spot/broker transactions, and capacity intermediation. During the 2021 shortage, broker spot prices for analog and MCU chips ran 5–20x above contract prices (IEEE Spectrum), indicating enormous value in intermediation during supply disruption. The gray-market broker ecosystem is estimated in the low billions annually, though precise sizing is difficult because it is inherently fragmented and poorly tracked. Capacity reservation (forward allocation contracts for foundry time) is an entirely greenfield concept — no equivalent of oil forward contracts exists for semiconductor capacity. This is speculative but structurally interesting: TSMC’s CoWoS packaging capacity is the current binding constraint on AI chip supply, and companies would plausibly pay for guaranteed future allocation if the contract structure existed.
SOM (Yr 1–3): $1–5M — Take-rate (5–15%) on matched surplus inventory transactions, starting with trailing-edge components where the compliance wedge provides a data advantage (knowing which suppliers are screened, which components are origin-verified). The compliance-to-marketplace pathway is: compliance data tells you who the verified suppliers are and what they make → marketplace connects buyers to those verified suppliers during shortages. The data advantage is real but the marketplace execution challenge is classic chicken-and-egg: you need both buyers and sellers to create liquidity. Supplyframe solved this by building a design-engineer audience first (10M+ users), then monetizing supplier access — a content-first strategy that took 18 years to reach $70M revenue and a $700M exit.
Key comparables:
| Company | Valuation/Revenue | Relevance |
|---|---|---|
| Supplyframe | $700M acq by Siemens, ~$70M rev (Siemens Press) | Design-to-source intelligence. 10M+ engineering users. Content-first marketplace model |
| Octopart | Acquired by Altium | Component search engine. Aggregates distributor inventory |
| Arrow Electronics | $37B revenue (public) | Incumbent distributor. Shows the size of the distribution value chain but also the difficulty of displacing it |
| Z2Data | Private, component intelligence | Electronics-specific lifecycle, compliance, and risk data |
Confidence: Medium. Distribution TAM is well-documented from public company revenues. Supplyframe’s exit validates the component intelligence category. But the surplus/broker market is poorly measured, capacity reservation is speculative, and the marketplace chicken-and-egg problem is a real execution risk. The compliance-to-marketplace pathway is logical but untested.
Gap flag: No conversations with distributors, brokers, or procurement teams. Need: 5+ broker/distributor interviews to understand surplus market mechanics and willingness to transact on a new platform.
6. Prediction Markets (Geopolitical/Supply Chain Risk Middleware)
TAM: $44B+ notional volume — Prediction market trading volume exploded to over $44B in 2025, with February 2026 run rates suggesting $200B+ annualized ($7B Polymarket + $9.8B Kalshi monthly, TRM Labs). ICE/NYSE invested up to $2B in Polymarket at an $8B valuation in Oct 2025 (TradingView). However, TAM measured in notional trading volume is misleading for Project TBD’s purposes — platform revenue (fees on trading volume) is a fraction, likely $200–500M combined for Polymarket and Kalshi, and Project TBD would not operate a prediction market.
SAM: $50–200M — Geopolitical and supply chain risk contract markets. More than $1B has been traded on geopolitical prediction markets since 2022 (CFR), with a single Iran-related contract attracting $73M in Feb 2026 on Polymarket. Supply chain-specific contracts (Taiwan disruption, BIS entity list changes, fab closure durations) do not exist yet but are a natural extension. Project TBD’s role would be as a data supplier, not a market operator — providing supply chain risk probabilities, event likelihood scores, or structured triggers that prediction markets could use to create new contract types.
SOM (Yr 1–3): $0–1M — Middleware/data feed revenue. This is the lowest-conviction opportunity in the portfolio. The theoretical path: Project TBD’s digital twin generates probabilistic risk assessments (e.g., “probability of >30-day disruption at TSMC Fab 18 in the next 6 months”) that prediction market platforms or enterprise risk teams would purchase as a data feed. This requires (a) a functioning digital twin with sufficient data to generate credible probabilities, (b) prediction market platforms that want external risk feeds (unproven), and (c) regulatory clarity for event contracts on supply chain disruptions (evolving). Realistically, this is a Year 5+ opportunity that may emerge organically as the digital twin matures, not something to actively build toward.
Key comparables:
| Company | Valuation/Revenue | Relevance |
|---|---|---|
| Polymarket | $8B val (ICE/NYSE investment, TradingView) | Dominant crypto-native prediction market. Geopolitical contracts are fastest-growing category |
| Kalshi | CFTC-regulated exchange | U.S.-regulated. Expanding contract types aggressively |
| Metaculus | Forecasting platform, private | Community-based probabilistic forecasting. More analytical, less trading-oriented |
Confidence: Low. The prediction market TAM is real and well-documented from on-chain/public data. But supply chain-specific contracts are entirely theoretical. The “middleware” business model (selling risk feeds to prediction markets) is unproven — no analog exists. This should be tracked as an optionality play, not sized as a near-term revenue opportunity.
Gap flag: No conversations with prediction market operators. The entire premise requires validation: do Kalshi/Polymarket want external data feeds, and would they pay for them?
Section 2: Consolidated Opportunity Map
| Opportunity | TAM | SAM | SOM (Yr 3) | Confidence | Key Risk |
|---|---|---|---|---|---|
| Compliance wedge | $4.5–6.5B | $500M–$1.5B | $3–10M | High | Altana incumbency; customer data reluctance; stuck-as-SaaS risk |
| Insurance | $19–21B | $1–3B | $5–20M GWP | Medium | Systemic risk uninsurability; no reinsurer validation; actuarial data gap |
| Trading/Benchmarks | $3–5B | $200–800M | $1–5M | Low | Non-fungibility barrier; historical failures; 10+ year buildout |
| Government | $10–20B | $2–5B | $2–8M | Medium-High | Long sales cycles; clearance requirements; Palantir incumbency |
| Marketplace | $200–418B | $5–15B | $1–5M | Medium | Liquidity chicken-and-egg; distributor incumbency; different buyer persona |
| Prediction markets | $44B+ volume | $50–200M | $0–1M | Low | Theoretical; unproven middleware model; regulatory uncertainty |
Combined realistic Yr-3 SOM range: $12–49M across all opportunities the compliance wedge unlocks.
Section 3: Honest Assessment
What’s backed by real data vs. rough analogy?
| Category | Data Quality |
|---|---|
| Compliance TAM/SAM | Strongest. Multiple analyst reports converge. Enforcement data (penalty amounts, detention volumes) is public record from DOJ, BIS, CBP |
| Insurance demand signal | Moderate. Lloyd’s/WTW survey of 100+ semi risk professionals is authoritative and semiconductor-specific. One confirmed parametric policy. But no observed premium volume for semi-specific products |
| Trading/Benchmark TAM | Weak. PRA revenue (S&P Global, Argus) is real. But scaling by semiconductor market size assumes comparable standardization — the very thing that doesn’t exist. BMI trajectory is real but batteries ≠ logic chips |
| Government demand | Moderate-strong. CHIPS Act mandates are law. Palantir and Recorded Future contract values and acquisition prices are public. But semi-specific government SC intelligence spending is not broken out anywhere |
| Marketplace TAM | Strong for distribution, weak for capacity reservation. Arrow/Avnet revenues are public company data. Supplyframe exit validates component intelligence. Capacity intermediation is speculative with no precedent |
| Prediction market TAM | Strong for overall market, very weak for SC-specific. Polymarket/Kalshi volumes are on-chain/public. SC-specific contracts are entirely conceptual |
Where are we most likely wrong?
-
Insurance SAM assumes demand converts to purchasable product. 88% of semi firms calling SC insurance “mission-critical” (Lloyd’s/WTW) does not mean 88% will buy. Systemic semiconductor shortage risk (the 2021 type) hits all participants simultaneously, violating insurance’s diversification requirement. Parametric insurance works for idiosyncratic events (earthquakes, single-entity sanctions) but may not be extendable to broad shortage risk. The protection gap may persist because it can’t be closed by insurance — only by operational hedges (buffer stock, dual-sourcing).
-
Trading/Benchmark SAM assumes the fungibility problem is solvable. Every prior attempt at semiconductor financial instruments has failed on this exact barrier. The oil analogy assumes standardization that does not exist: a barrel of WTI crude is fungible, a 5nm TSMC logic die is not. Even the narrow approach (trailing-edge MCU indices) requires sufficient homogeneity and market depth to create a meaningful benchmark. This needs direct confrontation, not analogy.
-
Government SOM assumes we can sell to government in Yr 1–3. FedRAMP authorization alone takes 12–18 months. Security clearances for personnel take 6–12 months. DC presence and prime contractor relationships take years to build. Realistic government SOM is Year 3–5 unless entering through CHIPS Act (commercial recipients) or prime contractor partnership.
-
Marketplace SOM conflates data advantage with transaction liquidity. Knowing which suppliers are compliant ≠ being the marketplace where they trade. Supplyframe took 18 years to reach $70M revenue. The marketplace opportunity is real but requires a fundamentally different GTM motion than compliance SaaS.
What primary research would sharpen each number?
| Opportunity | Research Needed |
|---|---|
| Compliance | 10–15 compliance officer interviews (validate ACV, buying process, switching costs) |
| Insurance | 3+ specialty insurer/reinsurer conversations (Munich Re, Zurich, Descartes Underwriting). Key question: is idiosyncratic semiconductor risk insurable via parametric structures, and at what premium? |
| Trading/Benchmarks | Interview 5+ auto-OEM procurement officers on index demand. Research all prior failed semiconductor index attempts in detail. Talk to BMI’s Simon Moores about minimum conditions for PRA viability |
| Government | Talk to CHIPS Act program office and 1–2 defense primes. Assess FedRAMP timeline and costs. Explore prime contractor partnership model |
| Marketplace | 5+ broker/distributor interviews. Understand surplus inventory market mechanics and take-rate economics |
| Prediction markets | 1–2 conversations with Kalshi/Polymarket BD on data partnership models |
The Non-Fungibility Problem
This is the single biggest structural risk to the grand slam vision and deserves direct confrontation. Oil’s entire financial ecosystem — futures ($200B+/day notional on CME alone), options, swaps, price benchmarks, parametric insurance — rests on the fact that a barrel of WTI crude is interchangeable with another barrel of WTI crude. Semiconductors are extraordinarily heterogeneous: 5nm vs. 28nm, logic vs. memory vs. analog vs. power, TSMC vs. Samsung vs. Intel, custom vs. commodity. No universal “chip grade” exists.
The honest framing by opportunity:
- Compliance wedge and Government: Do not depend on fungibility. These work with heterogeneous, relationship-based data. No structural barrier.
- Insurance: Partially depends on it. Parametric triggers can be event-based (earthquake near TSMC, entity added to BIS list) rather than price-based. Viable as long as product design avoids requiring a chip price index.
- Trading/Benchmarks: Heavily dependent on some degree of standardization. The narrowest viable approach: lead-time indices and broker-premium benchmarks for commodity-like segments only (commodity DRAM, trailing-edge MCUs). This is a real but much smaller opportunity than “Platts for chips.”
- Prediction markets: Depends on whether supply chain events can be structured into tradeable contracts — a question about event standardization, not product standardization. Potentially viable but untested.
Questions for Bliss & Dustin
-
Sequencing. The compliance wedge is high-confidence and the natural Day-1 play. Which second opportunity do you lean toward? Insurance (needs reinsurer conversations) vs. government (needs clearance infrastructure) vs. benchmarks (needs 10+ year patience)?
-
The stuck-as-SaaS question. The most likely failure mode is getting stuck as a compliance tool and never transitioning to financial products. What’s the concrete trigger for building the next layer — revenue threshold, data volume, specific partnership?
-
Non-fungibility: accept or solve? Should financial products be designed to work around non-fungibility (event-based triggers, narrow commodity-chip indices) — or do you believe a broader standardization bridge can be built? This shapes the entire product roadmap.
-
Government: now or later? Government demand is real and funded but requires fundamentally different infrastructure (clearances, DC presence, prime partnerships). Is the CHIPS Act compliance track (selling to commercial recipients, not agencies) a viable near-term entry that builds toward the broader government play?
-
Kill criteria. If the next 5 insurer conversations all say “systemic semi risk is uninsurable,” does insurance narrow to event-based parametric only, or drop entirely? If 10 procurement officers say “we would never pay for a chip price index,” does benchmarking drop? What evidence would remove an opportunity from the roadmap?
All market size estimates represent ranges across multiple analyst firms — treat as order-of-magnitude, not precise. SOM estimates are projections based on comparable company trajectories and addressable customer counts, not validated market research. Per RDI methodology: this document prepares inputs. The judgment about which opportunities to pursue belongs to Bliss and Dustin.