A $7.8 Billion Market in Used Jet Parts Runs on Photocopied Tags and Phone Calls
A used jet part costs 40 to 70 percent less than a new one, but the market still runs on photocopied FAA tags, emailed photos, and phone calls, and forged tags keep showing up.
Picture a regional airline with a 737 sitting on the ramp at 6 a.m., grounded because a single hydraulic pump failed overnight. Every hour that jet sits still, the airline burns money it will never get back: a cancelled flight, rebooked passengers, a crew timing out, a slot lost at a congested hub. A new pump from the manufacturer costs full list price and arrives in weeks or months. A used serviceable one, pulled from a retired 737 and recertified, costs half as much. It sits on a shelf in Florida right now. Getting it from that shelf to the airplane, with paperwork a quality inspector can check, is the whole business. It is also the hard part.
Published estimates put the global used serviceable material trade, known as USM, at $5.22 billion to $8.03 billion in 2025, heading for $8.7 billion to $15.5 billion by the early 2030s. This article models on the $7.8 billion figure, near the top of that range. At the low end, every figure derived from that $7.8 billion TAM falls by about a third.
- A used jet part costs 40 to 70 percent less than a new one, but lead times for certain new components reached 200 days and the trade still runs on photocopied tags and phone calls.
- Forged tags are a live problem: one UK broker sold more than 60,000 parts with forged certificates, and its director was sentenced to four years and eight months in February 2026.
- The US market is about $2.32 billion, served by roughly 4,000 small repair firms that buy on trust because no marketplace open to small shops checks documents or holds money.
- The business: document checks, escrow, and public prices, so a small shop buying from a stranger gets checks, escrow, and public prices it does not get today.
The Problem
Global spending on aircraft maintenance, repair, and overhaul hit $136 billion in 2025, according to the 2026 assessment from the Aeronautical Repair Station Association and Oliver Wyman. That assessment expects spending to reach about $193 billion by 2036. Oliver Wyman counts a little over 29,000 commercial aircraft in service at the start of 2025, growing to 38,300 by 2035, with a record order backlog above 17,000 aircraft. At $136 billion spread over 29,000 aircraft, the industry spends about $4.7 million per airplane per year keeping the fleet flying.
Three forces are squeezing that fleet at once: airplanes are getting older, replacements are arriving late, and the engines on the newest jets keep breaking. IATA puts the average age of the global commercial fleet at 15.1 years at the end of 2025, a record high, 1.5 years, or 11 percent, above the long-run average of 13.6. Cirium expects average retirement ages around 24 years for narrowbodies during the current delivery crunch, against a historical average near 22. Pratt and Whitney's geared turbofan problems grounded 648 aircraft in March 2025, about 28 percent of the roughly 2,300 jets flying with that engine family.
When new parts cannot arrive fast enough, operators turn to used ones. Lead times for certain new components extended to 200 days or more in 2025 and 2026, about six and a half months, according to Werner Aero chief executive Tony Kondo. Original equipment manufacturers raise list prices 5 to 6 percent a year on average, which compounds to 28 to 34 percent over five years, and USM prices keep climbing across a broad range of products because demand keeps outrunning supply.
Scarcity has an ugly side effect: it makes fraud pay. A used part may be installed only if the installer can establish eligibility and airworthiness for that specific aircraft under Part 43, commonly evidenced by an FAA Form 8130-3 airworthiness approval tag or an EASA Form 1. That tag alone does not authorize installation. The form itself states that its existence does not automatically constitute authority to install, and the installer retains responsibility for the airworthiness determination, airworthiness directive compliance, and life limits. A failed rotable can do worse than ground an airplane; it can bring one down. That is why the installer, not any marketplace, owns the airworthiness decision, and why the paperwork matters as much as the part.
Paper tags are easy to fake. Industry advisers report that counterfeiters replicate 8130-3 tags and EASA Form 1 certificates using authentic serial numbers taken from scrapped or cannibalized aircraft, altering PDF metadata to match legitimate production runs. A forged tag looks exactly like a real one to a receiving inspector reading it by eye.
Regulators have been chasing this for thirty years. The FAA's Suspected Unapproved Parts program dates to 1995, and its current guidance is Advisory Circular 21-29D. By 2004 the program had run more than 2,500 investigations. No current cumulative total is published, so that figure is twenty-two years old and should be read as a floor, not a count of today's caseload.
In August 2023, UK and EU regulators issued suspected-unapproved-parts notices after AOG Technics, a UK parts broker, was reported to have supplied CFM56 engine parts with falsified release certificates; the FAA followed in September 2023. Operators grounded aircraft while the extent was investigated. The sentencing remarks record that falsified parts were installed into 52 aircraft engines, including those of TAP and other client airlines. The substantive case ended in a conviction: on December 1, 2025 the company's director, Jose Alejandro Zamora Yrala, pleaded guilty to fraudulent trading, and on February 23, 2026 he was sentenced to four years and eight months in prison. A proceeds-of-crime case remains ongoing. Between January 2019 and July 2023 the company sold more than 60,000 aircraft engine parts accompanied by forged Authorised Release Certificates. This is a current event, not history: the prison sentence came down nine months ago.
On November 10, 2026, the Justice Department announced charges in the Southern District of Florida, case 26-cr-60239, against Alain R. Ruiz, alleged to have used fraudulent FAA Form 8130-3 tags and fabricated traceability records and military purchase orders in an alleged parts-and-investment scheme involving more than $1 million in alleged victim losses. An indictment is an allegation only, and Ruiz is presumed innocent until proven guilty. The $1 million figure covers the whole alleged scheme, including the investment side, not forged tags alone. Read that way, the case still shows the pattern: fake paperwork, real money, and a buyer who wired funds on the strength of a PDF.
Meanwhile the supply of used parts is itself tightening. Aircraft retirements fell to 3,400 in 2021 through 2026, down from more than 4,600 in 2015 through 2020, according to Aviation Week's 2026 fleet and MRO forecast, with 2026 a projection rather than a count. Fewer retirements mean fewer teardowns, which means fewer used parts entering the market just as demand peaks. Competition for teardown feedstock is so intense that traders are now dismantling young airplanes. Around 100 Airbus A320neo family aircraft and five Boeing 737 MAX jets have already been dismantled or committed to disassembly, and two former Spirit Airlines A320neos stripped in 2026 were three and four years old, the youngest airframes ever scheduled for teardown. Two full-life engines can now represent around 80 percent of a new aircraft's value, against roughly 20 to 30 percent two decades ago, per IATA. In some cases the engines are worth more rented separately than the whole airplane earns flying.
Demand has never been higher, supply has never been tighter, prices keep rising, and buyers and sellers still meet in subscription databases that list part numbers and phone numbers and leave trust to a fax machine. The money is in the hole between them, not in the parts themselves.
The Gap in the Market
Parts trading has incumbents. None of them verifies what it lists for the open market.
| Company | Scale | Business model | Verifies tags? | Escrow? | The gap |
|---|---|---|---|---|---|
| ILS (Inventory Locator Service) | Largest electronic aviation parts marketplace, founded 1979; owned by Hearst's CAMP Systems since 2019, bought from Boeing | Subscription | No | No | A directory with a paywall; no verification; seller checkout only, no escrow or marketplace-held payment |
| PartsBase | 90M+ parts on PBLocator | Subscription ($300 for 60 days); free buying through PartStore | No | No | Search and RFQ tool; verification and payment stay between buyer and seller |
| Locatory.com | Large European listing base, 25,000+ aviation professionals | Annual membership | No | No | A locator, not a transaction platform; no tag checks or escrow |
| Aeroxchange | Airline-to-airline trading network; co-launched the first digital 8130-3 shipment with Boeing and Southwest in October 2025 | Network fees | Yes, digital 8130-3 authentication, inside its own network only | No | Closed to large airlines; not open to small shops |
| StockMarket.aero | Smaller independent listing service | Listings | No | No | Listings only; no verification, escrow, or pricing data |
| AAR, GA Telesis, AJW Group, AerSale | Large USM traders; AAR led the global USM competitive field in 2025; GA Telesis says it is dismantling a record number of aircraft and engines | Principal trading, selling their own inventory | Their own stock only | No | They are sellers, not a neutral marketplace for the other 4,000 shops |
| Boeing Distribution Services | Global factory-backed distribution, kitting, logistics | New-parts sales | Not applicable, new parts | No | Sells new parts; does not touch the used market |
Every incumbent shares the same shape: a directory with a paywall. Buyers search a part number, see who claims to have it, and then pick up the phone. Price discovery is a negotiation between two people, one of whom may have done this a thousand times and the other never. Trust comes from the relationship, not from the platform. A first-time buyer has no way to know whether the seller's tag is real, or whether the part in the warehouse is the part in the photograph.
Small operators feel this worst. The United States has 5,006 civil MRO firms, and 80 percent of them are small or medium enterprises, about 4,000 firms. Those firms employ 449,765 people and generate $43.7 billion in MRO activity plus $24.9 billion in parts manufacturing and distribution, $68.6 billion combined. That works out to about $13.7 million per firm across all 5,006 firms and about $153,000 per employee. A five-person repair station in Tulsa does not have a procurement department. It has one buyer who also answers the phone, and that buyer spends hours chasing parts through brokers, waiting on callbacks, and reading tags by eye. Subscription databases charge for access whether or not a small shop finds anything, and the big traders would rather sell a $400,000 engine package to an airline than a $9,000 actuator to a regional operator.
Picture the buyer at that Tulsa station on a Tuesday morning. A hydraulic pump is needed by Friday, and if it does not arrive, a flight cancels. The database shows three sellers, and the first two do not answer the phone. The third emails a photocopy of a tag, and the buyer spends an hour comparing the serial number on the tag with the serial in the listing, calling the repair station named on it, and waiting for a callback that comes after lunch. Every step is manual, every check depends on someone picking up the phone, and a tag that looks fine on a photocopy can still be fake.
Verification is the missing layer, and the industry knows it. In February 2024, aerospace leaders formed the Aviation Supply Chain Integrity Coalition to recommend actions against unapproved parts, focused at first on propulsion parts. Its top recommendation was wider use of digital authorized release certificates. Boeing, Southwest, and Aeroxchange completed the industry's first parts shipment accompanied by a digital 8130-3 certificate in October 2025, using X.509 security and public-key encryption to authenticate the signer and protect the document. That shows the technology can work inside a closed system. It does not help a small shop, because the digital certificate only travels inside the closed systems of the companies that built it. Nobody has opened it up as a marketplace feature for the other 4,000 firms.
The Solution
A marketplace for used serviceable aircraft parts that checks documents, holds money, and publishes prices, built for the long tail first: small and mid-size repair stations, regional airlines, charter operators, and business aviation shops. Airlines and big traders can use it too, but the product is designed for the buyer who has one mechanic waiting on one part and no procurement team behind them.
One limit first: the platform checks documents, not airworthiness. A passed check means the paperwork was consistent on the day it was checked. It does not mean the part is airworthy, and it does not authorize installation. That determination stays with the installing repair station under Part 43, as the previous section explains.
Every listing carries a document check, and the badge says exactly what ran. A seller uploads the 8130-3 or EASA Form 1 with the listing. Software reads each field and cross-checks it three ways. It matches the named repair station against FAA records, flagging tags dated after a known surrender or revocation where that history is available. It checks the serial number against every other listing here, so the same serial, as entered, cannot sell twice. And it checks the part number against published FAA and EASA unapproved-parts notices. Listings that fail a check are held for human review and do not go live until a person clears them, which means the check was found erroneous, not waived. A listing cleared this way carries the failed check and the reason on its page, so the badge never hides an earlier failure. Listings that pass carry a badge reading "Document checks passed" with the checks that ran and the date they ran, never "verified part" or "verified tag."
The Tulsa buyer would see the checks that ran and pay into escrow instead of spending the morning on the phone.
What verification cannot detect. A forged tag built on an authentic serial number taken from a scrapped aircraft will pass a field-consistency check, a repair-station check, and a serial-duplicate check on this platform. A genuine tag paired with a different physical part will pass too. A serial sold off-platform cannot be caught by a platform-internal check. FAA notices cover specific part numbers and batches, not every serial in existence, and they arrive after the fact. A passed check raises the cost of crude forgery. Forgery built on authentic serials still gets through. It does not prove the part is genuine, and every listing page carries that limitation in the same block as the badge, not only in the terms of service.
Money moves through escrow held by a licensed partner. A buyer pays into the licensed partner's custodial account, not the seller and not the startup's own bank account. Holding buyer funds for a third-party seller is a regulated activity in most states, so escrow runs through licensed escrow and money-transmitter partners, licensed in each state where the sale requires it, with the required bonds, custodial accounts, and anti-money-laundering checks. Funds release only when the buyer's receiving inspector confirms that the part number and serial match the listing and the tag matches the checked copy on file. The buyer has five business days to inspect. If there is no confirmation by then, the escrow partner notifies both parties on day ten, and the sale unwinds on day twenty if the part is still unconfirmed. That confirmation is a document and physical match check, not an airworthiness finding.
If the part does not match, the buyer returns it at the seller's expense and the money goes back in full: return shipping comes out of the seller's settlement, never out of the buyer's refund. If the seller disputes the mismatch, the arbiter decides on photos and the tag copy before the return ships, with the losing party bearing the arbiter fee. If the buyer never inspects, funds stay held until the day-twenty unwind; money never moves on the checked copy alone, and the listing says so in plain language. If the buyer does not respond at all, the escrow partner notifies both parties after ten business days, and after ten more the sale unwinds: the part goes back at the seller's expense and the buyer is refunded. The seller absorbs return freight on a no-fault unwind; that is the disclosed cost of the marketplace's buyer protection. If a silent buyer will not return the part, the escrow partner arranges collection, and the cost comes out of the seller's next settlement, never out of the buyer's refund. That is an unwind, not a release: the seller is never paid for an unconfirmed part. Disputes go to an independent arbiter named in the terms, with arbiter fees capped at 2 percent of order value and borne by the losing party, and loss or damage in transit is covered by shipping insurance, on both legs, charged through to the buyer; it does not cover a forged or mismatched part.
Prices are public. The marketplace publishes a price index by part number, built from completed transactions: median realized price, number of transactions, days on market, and the discount against the current OEM list price. Buyers stop guessing whether $18,000 is fair for a starter-generator with 4,000 hours left. Sellers stop pricing from memory. That history is also a data product: lessors, insurers, and teardown investors will pay for realized prices no incumbent publishes. The index publishes medians and counts only, never a seller's individual transaction prices.
The marketplace starts narrow. Narrowbody rotables first: A320 family and 737NG components, the highest-volume, most liquid used parts in the world. Narrowbody jets hold 53.7 percent of the USM market in 2025, about $4.2 billion at the $7.8 billion figure, and the A320 and 737 families dominate retirements and teardowns. Two airframe families, a few thousand part numbers, and a document-check pipeline tuned until it works. Widebodies, engines, and business aviation come later, once the narrowbody loop is tight.
An AOG lane for grounded aircraft, with identical checks. When an airplane is grounded waiting for a part, speed beats price. Sellers within a day's shipping distance get a dedicated lane with contracted response times and courier booking built in. AOG changes shipping speed and response times only. Verification steps, hold times, and rejection criteria are identical, and no AOG listing may bypass a check. AOG never shortens the five-business-day inspection window. The higher AOG fee is a conflict worth naming. The platform earns more when AOG moves faster. So AOG verification is audited on a separate track by the same independent firm and reported quarterly, and the 8 percent fee is shown on the listing itself, not buried in the terms.
Governance has to manage a built-in conflict: the platform profits from approvals. A marketplace earning 6 to 8 percent of gross merchandise volume earns more by approving more listings, which pulls directly against strict gatekeeping. Three separations manage that conflict. Verification decisions are audited by an independent firm appointed by the board for a fixed term, with rotation. It is paid from a fixed budget with a charter-set minimum the verification team cannot influence, and the board cannot lower that minimum without auditor consent. The auditor publishes its findings directly, without board approval. False-pass and false-fail rates, sample sizes, and raw counts are published quarterly. Rates are measured on sampled re-checks and downstream reports; a forgery never detected is never counted as a false pass. The verification team is paid on salary, not on volume, with no listing quota. And when a human reviewer suspects a sold part is fraudulent, the platform runs a defined recall procedure. The seller is suspended, and other listings from the same seller and batch are frozen. Notification goes to the buyer and any known downstream holder within 24 hours of suspicion. A voluntary suspected-unapproved-parts report (Form 8120-11) is filed with the FAA Hotline. A public notice names the part number and serial, and names the seller once suspension is in effect. Filing is voluntary, and a failed software check alone is not a finding.
A 2% Share Is a Real Company
Published estimates disagree by a wide margin, so here they are side by side instead of blended into one confident number.
| Source | 2025 market | Forecast | Implied growth |
|---|---|---|---|
| Market.us | $7.7B | $12.3B by 2035 | 4.8% a year |
| Dataintelo | $7.8B | $14.6B by 2034 | 7.2% a year |
| SNS Insider | $7.80B | $12.11B by 2035 | 4.5% a year |
| Verified Market Reports | $5.22B | $8.67B by 2034 | 5.8% a year |
| Reanin | $8.03B | $12B by 2032 | 5.9% a year |
| Growth Market Reports | $7.81B | $15.48B by 2034 | 7.9% a year |
Growth figures are recomputed from each source's endpoints over 2025 to the forecast year: Verified Market Reports states 6.01 percent and Growth Market Reports states 7.8 percent, so the table shows 5.8 percent and 7.9 percent instead.
This article models on the $7.8 billion figure, near the top of the published range. North America holds about 38 percent of that market (Market.us; other houses put it at 35 to 37 percent), roughly $3.0 billion in 2025. The United States alone was about $2.32 billion in 2025, heading for $3.57 billion by 2035, according to SNS Insider, about 4.4 percent growth a year. Use the US figure as the serviceable addressable market.
The savings case, in one calculation. Start with the 2026 MRO figures from ARSA and Oliver Wyman: engines $70.1 billion, airframe $27.8 billion, components $24.0 billion, line maintenance $17.7 billion. That totals $139.6 billion (70.1 + 27.8 + 24.0 + 17.7), up from $136 billion in 2025. Engines and components, $94.1 billion combined, are the parts-addressable base. Suppose 10 percent of that base shifts to used parts over the next decade, at a 55 percent discount to new, the true midpoint of the 40 to 70 percent savings range the market reports. The arithmetic is 0.10 times $94.1 billion times 0.55, about $5.2 billion a year in avoided purchase cost. That is a gross number, before verification, failures, groundings caused by bad parts, and liability, not net social benefit. It is also an upper bound in one direction: the $94.1 billion base still includes overhaul labor, which no used part can replace. A 10 percent shift is $9.4 billion of parts at new prices; at a 55 percent discount that is about $4.2 billion of used-parts spend, against the $7.8 billion 2025 global market.
| Share shift | 40% discount | 55% discount | 70% discount |
|---|---|---|---|
| 5% | $1.9B | $2.6B | $3.3B |
| 10% | $3.8B | $5.2B | $6.6B |
| 15% | $5.6B | $7.8B | $9.9B |
All figures in that table are gross avoided purchase cost on the $94.1 billion engines-plus-components base, which still includes overhaul labor, before verification, failures, groundings caused by bad parts, and liability.
Its own numbers are smaller, deliberately so. Grow the US market three years at 4.4 percent and it reaches about $2.64 billion. Two percent of that is about $53 million in annual gross merchandise volume, or roughly 1,100 to 10,600 orders a year at $5,000 to $50,000 per rotable. A 6 percent take rate on that volume yields about $3.2 million in transaction revenue. Add 300 professional subscribers at $249 a month, about $0.9 million a year, and price-index licensing to a dozen lessors and insurers at an average of $25,000 a year, another $0.3 million. Year-three revenue lands near $4.4 million on $53 million of GMV. That is a small share of a large market, which is the point: the plan does not need to beat ILS or outbid GA Telesis for teardown feedstock. It needs to become the checked checkout for the thousands of small transactions the incumbents never see.
| Line | Value |
|---|---|
| Global USM market (TAM) | $7.8B (2025) |
| US serviceable market (SAM) | $2.64B (year three; $2.32B in 2025, grown at 4.4% a year for three years) |
| Platform share, 2% (SOM) | $53M GMV (year three) |
| Year-three economics | |
| Transaction revenue, 6% take rate | $3.2M |
| Subscriptions, 300 at $249 a month | $0.9M |
| Price-index licensing, 12 at $25,000 | $0.3M |
| Total revenue | $4.4M |
| Operating cost, year three | ~$2.4M ($200,000 a month) |
| Average monthly GMV, year three | ~$4.4M |
| Break-even GMV, contribution basis | ~$2.8M a month |
Unit economics per order work because the ticket sizes are large. A typical narrowbody rotable sells for $5,000 to $50,000. At a 6 percent take rate, the marketplace earns $300 to $3,000 per completed order. Document checks are mostly software: reading fields, registry checks, and serial matching run for cents per listing once built, but human review of held listings is the real cost and is not yet priced in this plan. Escrow and payment processing run about 1.5 to 2 percent of order value, and shipping insurance is charged through to the buyer. Gross margin on transaction revenue is modeled at 60 percent at run rate. That is a target, not a measured figure: 6 percent take, less about 2.4 points of processing and support. It is before the cost of human review of held listings, which is not yet priced in this plan. Break-even on a contribution basis sits near $2.8 million in monthly GMV, before the unpriced human-review cost: $100,000 a month from subscriptions and licensing, plus 60 percent of the 6 percent take, or 3.6 percent of GMV, on transaction fees, against a $200,000 monthly burn. Year-three average GMV of about $4.4 million a month clears it, so the business turns contribution-positive during year three rather than after it.
On supply: AerFin says a single A320neo teardown yields between 1,200 and 1,400 reusable components. The roughly 100 A320neos dismantled or committed to disassembly imply 120,000 to 140,000 components, using that range, and that counts aircraft committed to disassembly but not yet dismantled. Most of those components are low-value items, not $5,000 rotables. The marketplace's addressable slice is the subset with a part number, a tag, and a buyer.
Free to List, Paid When It Sells
Transaction fees carry the business. Buyers pay nothing to search or to place an order. Sellers list for free. The marketplace takes 6 percent of completed order value from the seller, 8 percent on AOG rush orders. Free listing matters: the long tail of sellers holds one or two of a part, not a warehouse of them, and a listing fee would keep their inventory off the platform. A take rate on completed sales aligns the marketplace with sellers instead of taxing them for showing up.
Professional subscriptions add a second stream. For $249 a month, a shop gets unlimited RFQs, price-index history by part number, watch alerts when a part it needs gets listed, and priority placement in ordinary search, marked as paid placement in search results and never affecting a listing's verification status. Paid placement never applies in the AOG lane, where ranking is strictly by fit and distance. Comparable products cost more and do less: PartsBase charges $300 for 60 days of locator access. A $249 tier undercuts that while adding document checks neither offers. The 300-subscriber year-three target and the dozen data customers are planning targets, not forecasts, and the 6 percent take rate is modeled on comparable marketplaces, not on interviews with buyers. Repair stations operate on thin margins, and a fee that looks small against a $30,000 order looks large against a $900 consumable. Pricing will need testing with real sellers.
Data licensing is the third stream. Realized transaction prices by part number are valuable to aircraft lessors pricing lease returns, insurers setting hull values, and teardown investors deciding which airframes to buy. Twelve customers at $25,000 a year is a conservative year-three target, and the data gets better as volume grows, which is the rare revenue line that compounds instead of flattening.
Financing is deliberately out of scope for year one. Advancing payment to sellers on checked orders earns a spread later, but lending against aircraft parts is a regulated activity that needs a partner bank, not a feature to improvise.
Startup Costs
| Line item | Year one | Notes |
|---|---|---|
| Engineering (4 people) | $720,000 | $180,000 per person all-in; marketplace, document-check pipeline, escrow integration; senior-heavy, remote-first |
| FAA counsel and compliance | $150,000 | Budget allowance, not a quote; outside aviation counsel, terms of service, liability structure, SUP reporting workflow |
| Verification data and registry access | $80,000 | FAA repair station records, UPN feeds, OEM list-price data licenses |
| Insurance (E&O, product liability) | $120,000 | Budget allowance, not a quote; does not cap exposure and does not transfer airworthiness responsibility |
| Escrow via licensed partner | $160,000 | Float, surety bonds, custodial compliance, money-transmitter partner fees |
| Go-to-market (2 sales, trade shows) | $180,000 | Two reps at $90,000 all-in, base plus commission, which is lean and only works with commission on closed sellers; MRO Americas and regional trade shows |
| Cloud, tooling, office | $60,000 | Hosting, document processing, monitoring, admin |
| Total year one | $1,470,000 | About $122,500 a month; pre-seed plus a bridge; ships in 9 months |
Year one burns $1.47 million to reach a working marketplace with checked listings for A320 and 737NG rotables, escrow live, and the first fifty paying sellers. Year two adds business aviation part numbers and the AOG lane, at roughly $1.8 million. Year three runs about $2.4 million as headcount grows to ten.
The Digital Tag Finally Exists
The pilot is real. Boeing, Southwest, and Aeroxchange shipped the first digital 8130-3 in October 2025. Until then, arguing for machine-checked tags meant arguing for a standard nobody used. That pilot shows a digital certificate can be issued on an FAA-authorized electronic system. It is a pilot, not an open standard: each issuer still needs FAA authorization for its electronic system, and a marketplace that is not itself a Part 145 repair station cannot issue a maintenance-release 8130-3 at all. It can only receive, store, and check them. The remaining work is authorization, interoperability, and adoption, not inventing the format.
Fraud, scarcity, and enforcement are converging. The AOG Technics prison sentence and the Miramar charges move buyers toward document checks they can see instead of a broker they have known for years. Retirements are down, lead times for certain new components reached 200 days, OEM prices rise 5 to 6 percent a year, and airlines are parting out three-year-old airplanes for their engines, so buyers search wider and buy from sellers they have never used. A marketplace for strangers is most valuable exactly when the relationship network cannot supply the part. The Aviation Supply Chain Integrity Coalition, formed in February 2024, gave the industry a shared vocabulary for the problem and a recommendation list that starts with digital certificates, and the FAA's SUP program keeps publishing Unapproved Parts Notifications. A platform that keeps an audit trail of every check and files voluntary reports where a human reviewer suspects an unapproved part creates a record regulators can request.
The Platform Profits From Approving Listings
Liability could kill the company outright. If a checked part fails and an aircraft is damaged, the marketplace is likely to be sued, whatever its terms of service say. Insurance is budgeted as an allowance in the plan for that reason, and document checks must be presented honestly: the platform checks documents, not airworthiness. The issuer that signed the tag owns the return-to-service or production approval it certified. The operator, the repair station, and the installer remain responsible at installation. Confusing those things in marketing copy would be both dishonest and legally dangerous. Every badge needs plain language about what was checked and what was not. Cover is unquoted and may be unavailable or cost multiples of the allowance; until cover is bound, order value is capped at $25,000 per order and a litigation reserve of $100,000 is held separate from operating burn.
Cold start is the operating problem. A marketplace with no listings attracts no buyers, and no buyers attract no sellers. Starting narrow helps: a few thousand A320 and 737NG part numbers, a few dozen teardown sellers, and a sales team that onboards them one by one. Seeding the marketplace with consigned inventory from two or three teardown shops, in exchange for a reduced take rate in year one, is the practical way through the cold start. The reduced rate buys no verification or ranking preference. Buying inventory outright is the wrong way: it turns a marketplace into a parts dealer with a warehouse problem.
Disintermediation leaks the fee. Once a buyer and seller have transacted once, nothing stops them from dealing directly the next time and skipping the fee. Escrow and document checks fight this, because the second transaction needs the same protections as the first, but some leakage is inevitable. The defense is to keep adding value that a direct deal cannot match: the price index, the AOG lane, and a transaction history that follows the seller's reputation. For 6 percent, a seller has to get something a handshake cannot buy: escrow, a price history, and a reputation that follows the seller.
Incumbents can copy the feature set. ILS could add document checks to its listings, and PartsBase could add escrow. Both have the distribution this startup wants. The startup's window is the time it takes a subscription business to ship a feature that changes its revenue model and takes on verification liability: probably two to three years.
Export controls are the risk most likely to be underestimated. Aircraft parts cross borders, and screening buyers against denied-party lists is not enough. The work starts with classification: ITAR USML Category VIII for defense articles, or EAR ECCNs 9A610 and 9A619 for military items, 9A991, or EAR99 for most civil parts. Then destination, end use, and end user, including military end-user rules; OFAC SDN screening plus the 50 percent rule; Russia and Belarus aviation controls; EEI and AES filing; and recordkeeping. The marketplace may be the exporter or facilitator even if the seller ships. Military components, like the F-5 parts in the Miramar case, may be ITAR or 600-series and are out of scope for a civil narrowbody launch. Export counsel comes before the first cross-border listing, not after the first international order.
Limitations
This analysis has real limits, and they should be visible. Market-size estimates for USM disagree by a wide margin. Published reports put the 2025 figures anywhere from $5.22 billion (Verified Market Reports) to $8.03 billion (Reanin), and the forecasts run from $8.67 billion by 2034 (Verified Market Reports) to $15.48 billion by 2034 (Growth Market Reports). Growth rates in the same reports range from 4.5 to 7.9 percent a year, across mixed horizons of 2032, 2034, and 2035. Nothing in this article resolves that spread; it only reports it. Modeling on the $7.8 billion figure is a choice, and at the low end every figure derived from that $7.8 billion TAM falls by about a third. Year-three figures derived from the US market scale with the US market, not mechanically with the global TAM.
Several inputs rest on single sources. That 200-day lead time comes from one chief executive's public statement. Nor is the 40 to 70 percent discount range broken down by part category. Market.us, citing AviTrader, puts savings at 60 to 80 percent; this article models on Dataintelo's 40 to 70 percent range. No incumbent shared transaction data, so realized prices and seller concentration are estimated from public material rather than measured.
Treat the savings calculation as a scenario, not a forecast. It assumes 10 percent of the engines-plus-components base shifts to used parts and that the average discount is 55 percent. If the shift is slower, or if OEMs cut list prices to defend share, or if regulators tighten USM rules after the next fraud case, the savings shrink. What the calculation does prove is narrow: at modest assumptions, the prize is measured in billions, which is enough to justify building the marketplace. It does not prove any particular startup will capture a share of it.
One more gap: this article did not prove that small repair stations will pay a 6 percent take rate. The 6 percent figure is modeled on comparable marketplaces, not on interviews with buyers. The same thin-margin problem applies to the fee itself, as noted in the revenue section: a 6 percent take looks trivial on a large rotable and punitive on a small consumable, and no buyer interview in this article tests where that line falls. Pricing will need to be tested with real sellers, and the take rate may need to vary by order size.
Strongest Counterargument
CAMP Systems already owns the largest marketplace, and it just watched the verification problem get solved. ILS lists the parts, CAMP has the seller relationships and the distribution, and it has no new-parts business to protect, so nothing stops it from adding document checks and escrow and owning this market itself. ILS has already added SalesEdge Commerce, an e-commerce layer that lets sellers on Corridor or Quantum ERP run branded storefronts with custom pricing and online purchasing built in, which shows the direction of travel. Add the digital 8130-3 that Boeing shipped with Southwest and Aeroxchange in October 2025, and the startup looks like a company charging for something the incumbent will soon bundle for free. Why would a small shop pay a 6 percent take rate to a startup when the directory it already pays for adds the same checks?
That case is strong, and it deserves a straight answer. ILS has not become a transaction platform already because its revenue is subscriptions: sellers pay to be found. A document-checked marketplace earns transaction revenue from sales that complete, and takes on verification liability, dispute handling, and escrow compliance. Those are different businesses with different staffing, insurance, and incentives, and a company rarely volunteers to swap a working subscription business for a harder one. ILS adding an e-commerce layer for sellers is not the same as showing its work on every document check: a checkout button is not escrow, and a listing is not a document check.
A badge does not turn a directory into a marketplace.
Escrow, dispute resolution, price transparency, and reputation are a different product from listing inventory. A directory wants as many listings as possible, while a checked marketplace wants listings whose documents it has checked. CAMP would have to choose, and the choice means taking responsibility for documents its sellers provide today with no liability at all.
Neutrality is the separate advantage. The big USM traders are sellers, not marketplaces. AAR, GA Telesis, and AJW Group compete with each other for teardown feedstock and for airline contracts. None of them will list inventory on a platform owned by a rival, and none of them will open their pricing to the small shops they currently serve through brokers. A neutral marketplace has one advantage no incumbent can copy: it does not compete with its own sellers. That is the reason the long tail would list here instead of on a trader's website.
None of that guarantees the startup wins. CAMP could decide the used market is worth owning and buy the startup instead of building the feature, which is a fine outcome for a founder and a real possibility. What the counterargument does not do is show the gap closing on its own. Somebody still has to build document checks, escrow, and pricing for the roughly 4,000 small and mid-size firms, and the companies with the distribution have reasons not to.
What You Can Do
If you run a repair station: Check tags the hard way today, before any marketplace exists to do it for you. For every used part you receive, check the repair station certificate named on the tag against the FAA's records, and against any surrender or revocation history you can obtain for the tag date. Photograph the serial number on the part and compare it with the serial on the tag, character by character. Search the FAA's Unapproved Parts Notifications for the part number before you install it. Keep a log of every broker you have bought from and every tag you have rejected, and share the rejections with other shops in your region. One rejected tag, filed with the FAA as a suspected-unapproved-parts report (Form 8120-11), protects the next shop the broker calls.
If you operate a small fleet: Ask your maintenance provider what share of your rotable spend goes to used parts and what checks it performs. If the answer is a shrug, you are paying new-part prices for used-part risk, or used-part prices with no checks at all. A standing requirement that every used part arrive with a tag checked against FAA records takes real labor, and a good provider will charge for it. That cost is small next to a grounded airplane. Put it in the contract at renewal.
If you are building this: Start with one teardown seller and one airframe family. Get fifty checked A320 rotable listings live, with tags machine-read and escrow working, before you write a line of pricing-index code. Sell the document checks to three repair stations by hand, watch where they hesitate, and fix that before you add a second airframe. The AOG lane comes after the first hundred completed orders, not before. And hire the FAA counsel before the first listing goes live, not after the first dispute.
The Bottom Line
Airplanes are getting older, lead times for certain new components reached 200 days, and the used parts that fill the gap trade on photocopied tags. Forged certificates put more than 60,000 suspect parts into the supply chain, some of them installed on airline engines, and earned one man a prison sentence this year. That is not a market waiting for a better database; it is a market waiting for someone to publish what the document check found and hold the money with a licensed partner until the part is confirmed to match.
Companies with the distribution sell new parts or subscriptions and have little incentive to build it. Traders with the inventory compete with their own sellers. That leaves the field open for a neutral marketplace built for the roughly 4,000 small and mid-size repair firms that keep old airplanes flying, and the math says a 2 percent share of the US used-parts trade supports a real company.
Remember the 737 on the ramp at 6 a.m. Its pump is on a shelf in Florida. What separates a cancelled flight from a departure is three things: a tag whose checks a buyer can see, money held by the licensed partner until the buyer confirms the part matches, and a price the buyer can check. If the part is not confirmed, the sale unwinds. Document checks come before escrow, escrow before the price index, and none of them is a promise that a part is airworthy. That determination belongs to the installer who decides to fit it, based on the tag, eligibility, airworthiness directive compliance, and life limits.
Check the documents, hold the money, publish the price.
Related Startup Ideas
For adjacent angles on the same theme, see general aviation hangar rate intelligence, FBO fuel price and fee transparency, and used EV battery certification intelligence, which tackles the same trust problem, checked used components, in a different vehicle fleet.