Webull
Webull runs a trading app on which people buy and sell US shares and options without paying a commission; the trading firms that fill those orders pay Webull instead.
The scores
Growth3/4 ↑
Revenue up 46% in 2025 and 51% in Q2 2026, organically, but it narrows to trading volume and accounts grew only 8%.
Business quality2/4 →
Customers can leave at no cost, market makers set the rebate, and Robinhood has 28.4M funded customers to Webull's 5.13M.
Financial position3/4 →
$701.6M of cash against $67.6M of debt, but margin lending drained $250.9M in the first half and it sold shares this year.
Execution record3/4 ↑
Turned GAAP-profitable and grew fast since listing in 2025, but the record is short and gives no guidance to be measured against.
Per share capture2/4 →
Stock pay near 8.6% of revenue, new shares for the equity line and the Pi deal, and a founder with about 71% of the vote.
Potential2/4 →
The options market is large and growing, but nothing is committed and the upside needs trading to stay high and rebates to survive.
Risk profile2/4 →
A rebate ban is one regulatory step away, four counterparties pay half the revenue, and China links draw government attention.
Valuation3/4 ↑
At 38.7 times trailing operating income the price needs 1.6 times Q2's pace by 2029, about the plan working if the trading lasts.
Evidence quality3/4 →
Revenue, volumes, cash and the rule change are filed or the regulator's, but the market makers are unnamed and retention is the company's own measure.
The Verdict
We rate Webull neutral. It runs a commission-free trading app paid by the firms that fill its customers' orders, and nothing stops a customer leaving for a far bigger Robinhood, so the business is mixed. When the day-trading limit on small accounts was dropped in June, its trading and profit jumped. The price is fair: about right if most of that jump lasts, which is reachable, but with no cushion if the extra trading fades. The view hinges on the September quarter showing the extra trading held, because it rests on a small group of very active traders.
The long caseSmall investors trade more options, and a rule change lets small accounts day trade freely; that extra trading must last, not fade once the novelty wears off.
The evidence under this - 7 claims
- Quarterly retention was 97.3% in Q2 2026, and the average account holds over $5,500, so customers do stay (S05, S02).
- 82% of customers report prior investing experience, an active-trader base that values tools a beginner app lacks (S04).
Q2's margin came with record volumes and lower marketing; a reversion toward Q1 halves the run-rate (C05, C33).
The company
Webull is an app and desktop platform for people who pick their own investments: US shares, funds, options, futures, crypto and bets on events. It charges no commission on US shares and options, and its customers tend to be active traders who want detailed charts.
Until June 2026 a US margin account with less than $25,000 could make only three day trades in five days. FINRA, the brokers' regulator, removed that limit, and Webull built its pitch around it. Another firm, Apex Clearing, holds customers' assets and settles their trades.
The evidence under this - 4 claims
- Quarterly retention was 97.3% in Q2 2026, and the average account holds over $5,500, so customers do stay (S05, S02).
- 82% of customers report prior investing experience, an active-trader base that values tools a beginner app lacks (S04).
How it makes money
When a customer places an order, Webull sends it to a trading firm called a market maker, which fills it and pays Webull a rebate. These rebates were $113.0M of Q2 2026's $198.8M, most of it from options. Interest on customers' cash and on money lent to them added $42.8M.
That makes Webull dependent on a few payers: four firms each brought 12% to 14% of 2025 revenue.
The evidence under this - 4 claims
- Concentration is falling: three counterparties were 41%, 24% and 11% of 2023 revenue (S01).
- The 20-F does not name the four, so whether all are market makers is inferred from the risk factor (S01).
Where the money comes from, stream by stream
| Stream | Share of sales (Q2 2026) | How it earns it |
|---|---|---|
| Order-flow rebates: options | 39.9% | Market makers pay Webull per options contract its customers trade; no commission is charged. |
| Order-flow rebates: equities | 16.9% | Market makers pay a small share of each stock trade's value for routing customer orders to them. |
| Interest-related income | 21.5% | Interest on customer cash, margin loans to customers and stock lending. |
| Handling charges | 17.5% | Fees on trades outside the free core, such as futures, crypto, prediction markets and foreign markets, net of some promotions. |
| Other revenue | 4.2% | Subscriptions, market data and smaller services. |
Who runs it
Founder Anquan Wang has run Webull since it began and controls it: his special shares carry about seven in ten votes. Anthony Denier has run the US broker since 2017 and H. C. Wang the money since 2021. Insiders own about a sixth of the shares. Their pay is mostly shares that vest with time; only the founder has an award tied to the share price, and nothing rewards profit or sales.
The evidence under this - 4 claims
The chief executive's 10.9M performance units vest only at average prices of $15 to $30, so part of his pay does ride on the share price (C22, S07).
By our arithmetic he still holds about 71% of the vote after the conversion, so control does not change (S07, S01).
Why it wins, and who could beat it
Why customers come back
Nothing stops a customer leaving. Opening an account elsewhere is free, every app charges no commission, and the market makers pay all apps the same kind of rebate, so Webull cannot charge more for a trade.
Customers do stay: Webull says 97.3% of them were still there after Q2, and the average account holds over $5,500. But Robinhood sells the same app to 28.4M funded customers against Webull's 5.13M, and Interactive Brokers sells keener traders deeper tools.
The evidence under this - 5 claims
- Quarterly retention was 97.3% in Q2 2026, and the average account holds over $5,500, so customers do stay (S05, S02).
- 82% of customers report prior investing experience, an active-trader base that values tools a beginner app lacks (S04).
- Concentration is falling: three counterparties were 41%, 24% and 11% of 2023 revenue (S01).
- The 20-F does not name the four, so whether all are market makers is inferred from the risk factor (S01).
Retention is the company's own measure and its definition is not filed (S04).
The threat
Robinhood is the threat that matters. It earned $561M in Q2 2026, almost as much as Webull's revenue for the whole of 2025, and it already earns real money from event contracts and crypto, the products Webull is just starting. Outside the US, Futu's moomoo app sells the same account in Webull's Asian markets.
The evidence under this - 4 claims
Webull's prediction-market revenue was about $5M to $6M in Q2 and crypto about 1% of revenue, so it is behind on both (S05).
The rivals: HOOD, IBKR, SCHW, MS, FUTU, and 1 more that is private, in-house or listed abroad.
Who else does this, rival by rival
| Ticker | Rival | Where it competes | How far along | Priced against it |
|---|---|---|---|---|
| HOOD | RobinhoodPriced against it: Yes | US retail brokerage: a zero-commission trading app paid by market makers and interest | On sale | Yes |
| IBKR | Interactive BrokersPriced against it: Yes | US retail brokerage: a zero-commission trading app paid by market makers and interest | On sale | Yes |
| SCHW · MS | Charles Schwab and E-TradePriced against it: Yes | US retail brokerage: a zero-commission trading app paid by market makers and interest | On sale | Yes |
| Private | FidelityPriced against it: Private | US retail brokerage: a zero-commission trading app paid by market makers and interest | On sale | Private |
| FUTU | Futu's moomooPriced against it: No usable figure | International retail brokerage: the same app licensed outside the US | On sale | No usable figure |
How big it could get
How big this could get
The market is large and growing. OCC, which clears every listed US option, cleared 70.8M contracts a day on average this year to August, 22.6% more than in 2025. Webull's 213M contracts in Q2 are a small slice, though nobody publishes how big a slice, because OCC does not split ordinary investors from professionals.
The evidence under this - 4 claims
The results
The numbers
Sales grew 46% in the year to Dec 2025, faster than the 0.2% in the year before.
It keeps 4.3% of sales as profit, down from 13% in the year to Dec 2022.
Yearly sales growth has ranged from +0.2% to +46% over the last 3 years.
Free cash flow was 22.7 times net income.
It turned 98% of sales into free cash, $562M over the year.
The share count rose 195% in 3.0 years, and pay in shares costs 7.7% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-10-04. Fiscal years end in December.
| Dec 2022 | Dec 2023 | Dec 2024 | Dec 2025 | |
|---|---|---|---|---|
| Size and growth | ||||
| Sales | $388M | $390M | $390M | $571M |
| Sales growth | - | +0.3% | +0.2% | +46% |
| Profit | ||||
| Net margin | 13% | 1.6% | −5.8% | 4.3% |
| What growth costs, per sales dollar | ||||
| Research and development | - | 8.2% | 8.4% | 6.2% |
| General and admin | 19% | 25% | 31% | 30% |
| Pay in shares | 7.7% | 7.5% | 8.4% | 7.7% |
| Capital spending | 8.1% | 1.2% | 0.6% | 0.9% |
| Cash | ||||
| Cash from operations | −$31.2M | $471M | $185M | $566M |
| Free cash flow | −$62.8M | $466M | $183M | $562M |
| Balance sheet | ||||
| Cash and short-term investments | - | $372M | $271M | $653M |
| Returns and shares | ||||
| Shares (diluted) | 134M | 138M | 139M | 397M |
What growth costs
Growth has made Webull more profitable, mostly because advertising stopped rising while sales jumped. Clearing costs rise with every trade and office and staff costs have grown faster than sales, so the gain rests on marketing staying flat.
The evidence under this - 2 claims
The filer's ResearchAndDevelopmentExpense tag, which the numbers card prints, was only $35.5M in 2025, narrower than the $79.2M technology and development line, so the two research figures use different definitions (S01).
- General and administrative rose from 24.6% to 29.5% of revenue over the same years, on stock pay, staff and office costs (S01).
- The finance chief said marketing in the rest of 2026 would match the first half, so the marketing gain depends on revenue holding (S05).
The last report (19 Aug)
Q2 2026 was the first quarter with the day-trading limit gone for part of it. Options contracts traded rose 68% from a year earlier and 34% from the first quarter, and revenue rose 51% to $198.8M. Profit was $24.4M, after a $21.7M loss in Q1.
Two things temper it. Much of the swing came from marketing, which fell to $35.0M from about $49.4M in Q1. And funded accounts grew only 8%, to 5.13M: the jump came from existing customers trading more, not many new ones arriving.
What it did to the thesisThe print showed the rule change lifting trading and profit at once, which is the bull case. It left open whether customers keep trading this much once the novelty fades, because accounts barely grew and the profit leaned on lower marketing.
The quarter in detail
Earnings per share $0.07 against $0.04 expected.
The evidence under this - 5 claims
Adjusted net income was $43.2M, so stock pay and the impairment take nearly half of the adjusted figure (S02).
Customer assets grew ten times faster than accounts, so the growth is existing customers depositing and markets rising rather than new customers (S02).
Contra revenue, promotions paid out of revenue, rose to $12.4M in Q2 from $5.1M a year earlier (S02).
What management did with the money
Management has spent shares rather than cash. It sold 11.5M new shares for $172.7M early in 2026, paid for a Thai broker, Pi Securities, mostly in shares, and pays staff about 8.6% of revenue in stock. A buyback of 1.82M shares offset little of that.
Insiders in 2026Q3: 1 purchases and 6 sales, from their Form 4 filings.
The evidence under this - 4 claims
By our arithmetic he still holds about 71% of the vote after the conversion, so control does not change (S07, S01).
The price, and where we disagree
In short. Fair. The price pays for most of the post-June trading lasting and a fatter margin, which is reachable but not protected by anything the company controls.
The shares doubled from April's low to $10.00 on 3 September, then fell to $7.34. The fall came as insiders sold without buying, the sellers of Pi Securities were cleared to sell 12.3M shares, and the chief executive converted 25M of his super-voting shares into ordinary ones.
The Q3 2026 results on 19 November are the first full quarter without the day-trading limit, so they show whether June's trading lasted.
The evidence under this - 5 claims
By our arithmetic he still holds about 71% of the vote after the conversion, so control does not change (S07, S01).
How we read the multiple
A loss-making Q1 inflates the price per dollar of last year's profit; on Q2's pace it is about 18 times.
- Valued at about $3.28B net of cash; grown 10% a year, a common rule of thumb for investors' return, is $4.37B in 2029.
- At 15 times operating profit, normal for a grown-up broker, that needs about $291M of 2029 profit.
- That is 1.6 times Q2's annualised $182M, and at a 25% margin $1.16B of revenue.
- That is revenue growth of about 19% a year from 2025, against 51% in Q2.
Against its own history. Too short to judge: Webull has traded publicly for under two years and was loss-making for part of it, so it has no normal price for its profit yet.
The evidence under this - 5 claims
Our own arithmetic gives about $3.25B of company value and 38.4 times; the pack's $3.28B is used for comparability (S06).
Q2's margin came with record volumes and lower marketing; a reversion toward Q1 halves the run-rate (C05, C33).
One or two analysts publish revenue estimates beyond 2026, so the consensus is thin (S06).
Q1 2026 carried about $49.4M of marketing, 40% of it amortising a 2025 promotion, so the trailing margin understates the run-rate (S05).
Contra revenue, promotions paid out of revenue, rose to $12.4M in Q2 from $5.1M a year earlier (S02).
Priced against its rivals
The competitors its own annual report names, each confirmed against the rival's filing. Measured on company value for each $1 of operating profit, cheapest first:
| Ticker | Company | company value for each $1 of operating profit |
|---|---|---|
| SCHW | The Charles Schwab | 11.8× |
| IBKR | Interactive Brokers Group | 13.8× |
| BULL | Webull | 38.7× |
| HOOD | Robinhood Markets | 52.3× |
| MS | Morgan Stanley not in the median | 25.4× |
| FUTU | Futu Holdings Limited held out | not priced |
The middle of the other 3 is 13.8. Morgan Stanley competes with it in one product line inside a much bigger company, so its price says little about this business and it is not counted in the middle. Webull Corporation is 2nd dearest of 4.
Rivals' figures come from a data provider, so read the spread of the group rather than any single pair.
Dearer than the established brokers on trailing profit and cheaper than Robinhood; the gap is this company's thin margin, not the group's pricing.
Private or foreign rivals that cannot be priced: Fidelity Investments.
Where we disagree with the market
The analysts are mostly positive, but the share price says the market expects Q2 to be the peak.
Some of the step will last, but it rests on a small group of traders and nothing keeps them at Webull.
The rule change is real, and Webull's management says trading will not fall back. But FINRA's own survey found only about 1.3M customers had been flagged as pattern day traders, out of about 150M, and 91% of accounts that traded never day traded at all. So the change freed a small, very active group.
We would be wrong if Q3 options contracts hold at 180M or more with operating margin at 15% or better.
What analysts say, and the evidence
The evidence under this - 6 claims
- July equity volume was softer than June on the call's own account, and customer assets fell 3% from May to June (S05, S18).
- FINRA found 91% of accounts that traded did no day trading at all, so the rule bound a small group (S10).
Customer assets grew ten times faster than accounts, so the growth is existing customers depositing and markets rising rather than new customers (S02).
What could go wrong, and how we would know
The risks
- A rebate ban. Over half of 2025 revenue came from market makers' rebates, which Congress and the SEC have questioned; a ban would cut the main income line.
- A few payers. Four firms each brought 12% to 14% of 2025 revenue, and losing one would leave a hole Webull cannot quickly fill.
- China links. Much of its engineering is in Changsha, China. A House committee called those ties a national-security risk on 7 October 2026; it ordered nothing, and Webull disputes it.
- One man's vote. The chief executive controls about 71% of the vote through shares with 20 votes each, so ordinary holders cannot change the board.
- A bigger rival. Robinhood has 28.4M funded customers and far more profit to spend on products and promotions.
- One clearer. Apex Clearing settles every US trade; a failure there would stop the business until Webull clears for itself.
The evidence under this - 6 claims
- Concentration is falling: three counterparties were 41%, 24% and 11% of 2023 revenue (S01).
- The 20-F does not name the four, so whether all are market makers is inferred from the risk factor (S01).
- The 20-F says no further requests came in more than a year, and US customer data is held in the US (S01).
- Webull's spokesperson says the report has significant inaccuracies, that it is run from St. Petersburg, Florida, and that US customer data is stored in the US (S19); the findings carry no automatic fine or ban (S20).
- 2026-10-08editedThe House Select Committee on China report of 7 October 2026 turned a question some lawmakers had asked into a published finding; read second-hand because the report itself could not be fetched.
What we still do not know
- How much of June's jump lasted into July and August. Webull published June's figures, and the call only described the months after.
- Who the four firms paying half the revenue are, and on what terms. Webull does not name them.
- What the 97.3% retention figure measures, and how many customers also keep a Robinhood account.
- How much revenue and profit comes from outside the US. Webull reports one business.
The evidence under this - 4 claims
July and August figures were not found in the sources read, so the post-June trend rests on the call's description (S05).
- Concentration is falling: three counterparties were 41%, 24% and 11% of 2023 revenue (S01).
- The 20-F does not name the four, so whether all are market makers is inferred from the risk factor (S01).
Retention is the company's own measure and its definition is not filed (S04).
Webull files one segment and does not split revenue or profit by country, so the international business cannot be valued on its own (S01).
What would change our view
- Neutral→AttractiveA price that needs only Q2's pace, with the trading holding
It needs all of:
- ○The shares fall until the company is valued at 25 times its last year's operating profit.about ≈$4.74 today, 35% below the last close
- ○Webull trades 180M or more options contracts in Q3 2026.checked at the next results, due in Nov 2026
- Neutral→UnattractiveThe trading fades, a rebate ban, or the price outruns it
It needs any of:
- ○Options contracts fall below 160M in a quarter, back to before the rule change.checked at the next results, due in Nov 2026
- ○The SEC or Congress bans or restricts the rebates market makers pay brokers.watched as it happens
- ○The shares rise until the company is valued at 55 times its last year's operating profit.about ≈$10.43 today, 42% above the last close
We would also be wrong if
- Funded accounts shrink for two quarters in a row.
- The US government restricts Webull over its links to China.
What is coming
- 30 October 2026 General meeting votes on new articles and the employee share purchase plan: Whether 5M more shares are reserved for staff, adding to the dilution the per-share score already reads; the founder's vote decides it.
Every dated event ahead
- 30 October 2026 General meeting votes on new articles and the employee share purchase plan: Whether 5M more shares are reserved for staff, adding to the dilution the per-share score already reads; the founder's vote decides it.
- 19 November 2026 Q3 2026 results: Options contracts against 180M, operating margin against 15% and marketing near the first half's pace: the first full quarter without the day-trading rule.
- April 2027 $50M of unsecured notes fall due: Whether repayment comes from cash while margin lending keeps draining operating cash flow, or needs new borrowing or shares.
Since we wrote this
- 13 OctNewsA shareholder class action now rests on the China-ties report; Webull rebuts it. View heldSecurities class action filed over China ties; Webull's 6-K rebuttal.
- 8 OctNewsHouse committee calls Webull's China ties a security riskA published finding with no order or fine; the criterion on a US restriction is closer, not triggered. View held.
- 5 OctThesis reviewThe long case now sits under the verdictAfter the five-name pilot, the owner asked for the long case on every thesis.
- 4 OctThesis writtenNeutral, medium convictionWritten six weeks after the Q2 2026 print and four months after the rule change, with the shares 27% below their September high.
- 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the 20-F and filings.
Also checked 2 times with nothing new to act on.
Sources 20
Sources - 20
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-04-09Item 4 business and competition, PFOF share and risk factors, revenue and receivable concentration, single clearer, China links, control and share classesbacksC01C03C04C14C18C21C27C28C38C40C41C42C43C44C45C46
- S02primarycompany release2026-08-19revenue by line, expenses, net income, balance sheet, buyback, operating metrics and international accountsbacksC01C02C05C07C08C13C17C19C21C24C26C46
- S03primarycompany release2026-03-042025 revenue by line, operating income, net income and net depositsbacksC03
- S04primarycompany filing2026-09-11the eight-quarter operating table, first-half statements and cash flow, the rule change named as a growth factor, Apex and the clearing licence, the equity linebacksC05C08C10C19C20C23C24C28C36C46
- S05primarytranscript2026-08-19Q2 2026 earnings call transcript (FMP), held 2026-08-19data/cache/transcripts/BULL-2026Q2.jsonvolumes after the rule change, marketing plans, margin before marketing, retention, account adds, prediction markets and crypto
It will not revert, or volumes will not revert to pre-PDT levels.
backsC07C10C17C33 - S06secondaryvendor data2026-10-04Research pack (FMP market tier, FINRA short interest, daily closes, cohort multiples)content/BULL/research-pack.jsonprice, market value, company value, multiples, short interest, analyst split and estimatesbacksC32C35C47
- S07primarycompany filing2026-10-02the chief executive's conversion and performance units, insider sales and the 10b5-1 flagbacksC22C34
- S08primarycompany filing2026-09-01the Pi Securities purchase, its consideration shares, price protection and the shares registered for resalebacksC23C25
- S09primarycompany filing2026-10-01the two resolutions: new articles and the 2026 employee share purchase planbacksC24C47
- S10primaryregulator2026-01-09FINRA's count of pattern day traders among member firms' customers and the share that day trade at all
Together, these members identified approximately 1.3 million current customers that were designated as PDTs.
backsC09 - S11primaryindustry data2026-07-02OCC June 2026 monthly volume data, as published on Mondovisione The Options Clearing Corporationindustry options volume in the first month after the rule change
OCC reported June 2026 total cleared contract volume of 1,608,649,778 contracts - up 45.0 per cent from June 2025.
backsC11 - S12primaryindustry data2026-09-03OCC August 2026 monthly volume data, as published on Mondovisione The Options Clearing Corporationthe year-to-date industry rate
OCC cleared more than 1.44 billion contracts during August, while year-to-date average daily volume increased 22.6% to 70.8 million contracts.
backsC11 - S13primarycompany filing2026-07-30Robinhood Markets Form 10-Q, quarter ended June 30, 2026 Robinhood Markets, Inc.Robinhood's revenue, net income, funded customers and transaction revenue mixbacksC13C37
- S14primarycompany filing2026-02-18Robinhood Markets Form 10-K, fiscal year ended December 31, 2025 Robinhood Markets, Inc.Robinhood's description of the same product sold to the same buyerread, but no claim rests on it
- S15primarycompany filing2026-02-27Interactive Brokers Group Form 10-K, fiscal year ended December 31, 2025 Interactive Brokers Group, Inc.about 4.4M brokerage customers on one electronic platformbacksC15
- S16primarycompany filing2026-04-02Futu Holdings Form 20-F, fiscal year ended December 31, 2025 Futu Holdings Limitedmoomoo, the international app sold to US and Singapore retail investorsbacksC15
- S17primarycompany filing2026-02-24Charles Schwab Form 10-K, fiscal year ended December 31, 2025 The Charles Schwab Corporationthe Investor Services segment descriptionbacksC38
- S18primarycompany release2026-07-15June 2026 trades, net deposits and customer assets
Equities DARTs were 1.3 million (up 57% from the end of May 2026, and up 110% year-over-year).
backsC39 - S19secondarynews2026-10-07the committee's report and Webull's answer, second-hand: the report itself was not read
It is deeply disappointing that the Select Committee published a report containing significant inaccuracies and unsupported conclusions without ever seeking clarification from Webull
backsC27 - S20secondarynews2026-10-07findings only, no order; named ownership, workforce, data routing, financing and customer-cash arrangements
The committee's findings carry no automatic fine or ban
backsC27