ThredUp teardown: 79% margin, still GAAP-unprofitable

Most DTC teardowns start with a story. This one starts with a 10-K, because that is the only version of ThredUp's year you can actually verify. ThredUp Inc. (Nasdaq: TDUP, the publicly traded online resale marketplace) filed its fiscal 2025 annual report on March 2, 2026, and it tells an operator a precise story: the resale model works on the gross-margin line, it finally works on the cash line, and it still does not work on the GAAP bottom line. If you run a brand in the $10M to $150M range, the value here is not the gossip about a public company. It is the playbook hiding in the line items, and the trap hiding in the non-GAAP table.

The headline: a resale model that finally generates cash

ThredUp printed the cleanest year in its public history. Revenue from continuing operations was $310.8M, up 19.5% year-over-year from $260.0M, the fastest growth since the 2021 IPO year and a clean break from two flat-to-down years (revenue was $258.5M in 2023 and $260.0M in 2024). Adjusted EBITDA from continuing operations turned solidly positive at $13.5M, a 4.4% margin, up 55.8% from $8.7M the year before.

The picture that matters is the two lines moving in opposite directions. Revenue re-accelerated while the GAAP operating loss compressed to $21.7M, the smallest in six years, down from a $40.6M operating loss in 2024 and an $89.5M trough in 2022. Losses shrank nearly 50% while revenue grew 20%. That is a fixed cost base spreading over a bigger top line, and it is the single thing that turns a story stock into a real business.

When we read a turnaround like this, the first instinct is to be skeptical of the inflection, not impressed by it. The pattern we see again and again with operators at this stage is that the cash line turns positive in one quarter and the founder declares victory, then a working-capital swing or a capex cycle pulls it back negative two quarters later. ThredUp's Q1 2026 operating cash flow was a positive $4.75M, roughly in line with the positive $5.74M in Q1 2025. That is an inflection worth respecting, but the company itself still guided FY2025 to negative full-year free cash flow after capex and timing. Inflection, not finish line.

Read the gross margin before you get excited

A 79.4% gross margin looks like software. It is not. ThredUp runs a consignment marketplace, which means it recognizes revenue net of what it pays the sellers who supply the clothing. The cost of revenue line, $64.1M in FY2025, only covers outbound shipping, fulfillment labor, and packaging. The real cost of goods, the payout to consignors, sits above the revenue line and never shows up in that 79%.

This is the most common place a financially literate operator gets misled by a resale comp. The take-rate margin and a retailer's product margin are different animals. A typical apparel brand buying inventory at roughly a quarter of retail and selling it might run a 60% to 65% gross margin and carry all the inventory risk. ThredUp's 79% looks better on the page precisely because the supply cost has been pushed off the income statement entirely and onto the seller. The number is real, but it is answering a different question than the one most people think they are asking.

Line item	FY2025 ($000)	FY2024 ($000)	% of FY2025 revenue
Revenue	310,813	260,031	100.0%
Cost of revenue	64,060	52,906	20.6%
Gross profit	246,753	207,125	79.4%
Operations, product, technology	152,859	142,210	49.2%
Marketing	58,982	48,639	19.0%
Sales, general & administrative	56,658	56,895	18.2%
Operating loss	-21,746	-40,619	-7.0%
Loss from continuing operations	-20,214	-39,999	-6.5%
Adjusted EBITDA (continuing ops)	13,524	8,679	4.4%

The cleanest line in this table is marketing. It held at 19.0% of revenue in FY2025 versus 18.7% in 2024. Spend rose 21.3% in dollars, almost exactly tracking the 19.5% revenue growth. ThredUp is not buying its growth at a worsening cost of acquisition. Revenue per marketing dollar held flat, which is the strongest single signal in the filing that the growth is closer to organic than bought.

Where the adjusted-EBITDA story breaks down

Here is the part of the filing you have to read slowly. The gap between a $20.2M GAAP loss from continuing operations and a positive $13.5M adjusted EBITDA is about $33.7M of add-backs. The largest single piece is $19.0M of stock-based compensation. Depreciation and amortization adds $12.9M, interest $1.9M, an impairment $1.1M, partially offset by a $1.5M gain on an equity investment.

Stock comp is non-cash, which is exactly why the adjusted-EBITDA convention adds it back. But it is not free. It is a real, dilutive cost paid by existing shareholders, and at ThredUp the dilution is visible: shares outstanding went from about 116M at the end of 2024 to roughly 127M a year later. When you hand out $19M of equity a year and then exclude it from your headline profit number, you are reporting a profit that your own shareholders are paying for in ownership.

When we talk to founders running a brand at this stage, the figure that keeps surprising them is how much of their own "adjusted profit" is just stock comp added back. We have sat with operators whose adjusted EBITDA looked healthy until we stripped out a seven-figure equity line, at which point the cash reality looked very different. The rule we use is blunt: if you would not be comfortable paying that stock comp in cash, do not count the add-back as profit. Apply that test to ThredUp and the positive adjusted EBITDA gets a lot more modest, which is precisely why the cash-flow line, not the non-GAAP line, is the one to track.

How they fixed it: cut Europe, drop the inventory, scale into overhead

The turnaround came from three moves an operator can copy, and none of them are exotic.

First, they killed the part of the business that was bleeding. In November 2024 ThredUp divested its European Remix business through a management buyout led by Remix's general manager, taking a $37.0M discontinued-operations loss out of go-forward results. That is why FY2025 and FY2024 are presented from continuing operations: the European drag is gone from the numbers you are diligencing. The lesson is not "exit Europe." It is that nursing a structurally unprofitable segment is a slow tax on the whole company, and cutting it cleanly is often the single highest-impact financial decision a founder makes.

Second, they got the inventory off the balance sheet. Owned inventory was $17.5M in 2022 and was reclassified as immaterial by Q1 2025 as the model went consignment-first. Asset-light by design means the working-capital risk that crushes most apparel operators simply is not there.

Third, they scaled revenue into their overhead. Sales, general, and administrative expense fell from 21.9% to 18.2% of revenue while marketing held flat. Revenue grew into a fixed cost base, which is the most boring and most reliable way a company stops losing money.

Item	FY2022 ($000)	FY2024 ($000)	FY2025 ($000)
Cash & equivalents	38,029	31,851	38,629
Marketable securities	n/d	n/d	9,500
Inventory	17,519	690	0 (reclassified)
Long-term debt	25,788	18,151	14,276
Shares outstanding (000)	101,532	116,134	127,027

The catch is the balance sheet at the bottom of that table. Roughly $48M of liquidity against a still-loss-making P&L is light. Low debt and no inventory are good things, but they also mean there is no reserve to absorb several more years of GAAP losses. The cash inflection was not optional. It was the thing that had to happen for the asset-light model to be safe rather than fragile.

ThredUp vs the resale peer set

Put ThredUp next to The RealReal and you see two roads to the same place. The RealReal is about twice the size at roughly $692.8M of FY2025 revenue, with a 74.6% gross margin. ThredUp is smaller at $310.8M but carries the higher take-rate margin at 79.4%, and it actually grew faster (19.5% versus 15.4%). The RealReal runs a shallower operating-margin loss and, on the latest figures, already generates positive free cash flow (about +$18.4M in FY2025).

Metric	ThredUp	The RealReal
Revenue ($M)	310.8	692.8
Revenue growth (%)	19.5	15.4
Gross margin (%)	79.4	74.6
Operating margin (%)	-7.0	-3.5
Cash & equivalents ($M)	38.6	172.2

For a private operator the takeaway is not which one is winning. It is that two companies with nearly identical 75-to-80% take-rate economics ended up in different places on cash because of size, capital history, and how disciplined each one was on the GAAP-to-cash gap. Scale buys you a thicker liquidity cushion. Take-rate buys you a better-looking top line. Neither, on its own, buys you GAAP profit.

What a private operator should take from this teardown

The CFO read on ThredUp is short. The resale model works on margin and now works on cash, but the GAAP loss and the $19M stock-comp add-back say the turnaround is real and unfinished at the same time.

If you are running a brand and reading this for your own business, copy four things. Kill the structurally bleeding segment rather than nursing it; the cleanest financial decision is often a subtraction. Get inventory off your balance sheet wherever the model lets you, because working-capital risk is what actually kills apparel operators. Hold marketing as a flat percentage of revenue so you can prove your growth is not just bought. And when someone hands you an "adjusted EBITDA positive" headline, including your own, discount it by the stock comp and look at the cash line instead.

The honest one-line read on ThredUp: it is cash-flow inflecting, not GAAP-profitable, and roughly $19M of its "adjusted profit" is stock comp added back. The model finally works. The income statement does not, yet. Track the cash line, discount the non-GAAP headline, and you will read your own P&L the same way.

For the forward look, ThredUp raised its FY2026 outlook after Q1 to roughly $351M to $356M of revenue at about a 6.1% adjusted-EBITDA margin. The direction is right. The job is to watch whether the cash line follows the margin line, or whether the next working-capital swing pulls the inflection back. That is the same question we put to every operator we sit with: not "are you adjusted-EBITDA positive," but "is the cash actually showing up, and what is it costing you in dilution to say you are profitable." If you want that read on your own numbers, a fractional CFO can run it in an afternoon. For the wider apparel context, our apparel financial benchmark report puts these margins next to nine public peers, and the Stitch Fix teardown covers the closest subscription-resale analog.

Sources and methodology

The primary source for this teardown is SEC EDGAR, ThredUp Inc., CIK 0001484778 (ticker TDUP, Nasdaq). Figures were pulled directly from the company's filings rather than from third-party aggregators wherever a number appears in this post.

The FY2025 numbers come from the Form 10-K filed March 2, 2026 (accession 0001484778-26-000007, report date December 31, 2025). The income-statement lines, the MD&A percentages, the non-GAAP adjusted-EBITDA reconciliation, and the discontinued-operations note were all read from the filing text, and the verbatim figures are quoted above. The quarterly and cash-flow detail comes from the Q1 2026 Form 10-Q filed May 4, 2026 (accession 0001484778-26-000016), which reported Q1 2026 revenue of $81.7M and positive operating cash flow of $4.75M.

The peer comparison uses The RealReal Inc., CIK 0001573221 (ticker REAL), FY2025 figures from its own 10-K: revenue of roughly $692.8M, a 74.6% gross margin, and a larger cash balance reflecting earlier capital raises. The two companies are the cleanest public read on resale-marketplace take-rate economics.

Three caveats matter for anyone re-checking the math. First, FY2020 through FY2023 in the revenue-versus-loss chart are as originally reported and include the European business divested in late 2024, so the pre-2024 points are not fully restated to continuing operations. Second, inventory was not literally zeroed in cash terms; remaining owned inventory was reclassified into other current assets in Q1 2025 as immaterial under the consignment-first model, so "effectively removed from the balance sheet" is the precise phrasing. Third, full-year FY2025 free cash flow was still negative after capex and working-capital timing even though quarterly operating cash flow turned positive, which is why this post calls it an inflection rather than a finished turnaround.

Channel context (store traffic, technology stack, product counts) was cross-checked against Storeleads, whose sales estimates are modeled and were deliberately not used as financial figures; the authoritative revenue number is the $310.8M from the 10-K. External confirmation of FY2025 results, the divestiture, and the raised FY2026 guidance was cross-referenced against the company's own investor-relations releases.

Frequently asked questions

is thredup actually profitable or does adjusted ebitda hide the cash burn?

It depends which line you trust. On a GAAP basis ThredUp lost $20.2M from continuing operations in FY2025, so it is not profitable. Its adjusted EBITDA was positive at $13.5M, but $19.0M of that bridge is non-cash stock-based compensation, which is a real dilutive cost even though it does not move cash this year. The honest read: cash-flow positive at the operating line, not yet GAAP-profitable.

what is thredup's gross margin and why is it so high at 79%?

FY2025 gross margin was 79.4%. It looks SaaS-like because ThredUp runs a consignment marketplace and recognizes revenue net of what it pays sellers. Its cost of revenue only covers outbound shipping, labor, and packaging. The big cost of goods, paying the people who supply the clothing, sits above the revenue line, so the 79% is a take-rate, not a retailer's product margin.

how does thredup make money, consignment or selling its own inventory?

Primarily consignment now. ThredUp transitioned to a marketplace-first model and effectively removed owned inventory from its balance sheet (it was $17.5M in 2022 and was reclassified as immaterial by Q1 2025). It earns a take-rate on items sold by consignors plus shipping and processing fees, rather than buying and reselling stock at its own risk.

why did thredup's revenue grow 20% in 2025 after being flat for two years?

Three things lined up. It divested the loss-making European Remix business in late 2024, which cleaned up continuing operations. It leaned into the consignment marketplace, which scales without inventory risk. And its Resale-as-a-Service white-label stores for brands added supply. Revenue went from $258.5M in 2023 and $260.0M in 2024 to $310.8M in 2025, up 19.5%.

how much of thredup's profit is really stock-based compensation add-backs?

Most of the gap. The bridge from a $20.2M GAAP loss to $13.5M positive adjusted EBITDA is about $33.7M of add-backs, and the single biggest piece is $19.0M of stock-based compensation. Depreciation and amortization adds another $12.9M. If you treat stock comp as a real cost, which it is for existing shareholders, the adjusted-EBITDA headline shrinks considerably.

how does thredup compare to the realreal financially?

ThredUp is smaller but higher-margin. In FY2025 ThredUp did $310.8M of revenue at a 79.4% gross margin, while The RealReal did roughly $692.8M at 74.6%. The RealReal is about twice the size and runs a shallower operating-margin loss, and on the latest figures it is already free-cash-flow positive (about +$18.4M). Two different routes to the same 75-to-80% resale take-rate economics.

how much cash does thredup have and how long is the runway?

At the end of FY2025 ThredUp held $38.6M of cash and equivalents plus $9.5M of marketable securities, roughly $48M of liquidity, against $14.3M of long-term debt. That is asset-light and carries little debt, but it is not a war chest. The cash-flow inflection had to happen, because there is no large reserve to absorb years of GAAP losses.

what should a private dtc operator copy from thredup's turnaround?

Four moves. Kill the part of the business that is structurally bleeding rather than nursing it. Get inventory off your balance sheet where the model allows. Hold marketing as a flat percentage of revenue so growth is not bought at a worsening CAC. And track the cash-flow inflection over the GAAP loss, while discounting any adjusted-EBITDA headline by the stock-comp add-back.