Riskified
An Israeli company that checks online orders for fraud for big retailers, and pays them back when an order it approved turns out to be fraud.
The scores
Growth3/4 ↑
Revenue grew 21.8% in Q2 2026 after 5.2% in 2025; the pickup leans on new merchants and live sports, and is under a year old.
Business quality2/4 →
Payouts held steady for three years, but private rivals and Stripe sell the same guarantee and existing merchants grow only 5%.
Financial position4/4 →
$223.6M of cash and investments, no debt, and positive free cash flow; nothing in the plan needs outside money.
Execution record3/4 ↑
2026 guidance raised twice and adjusted profit rising, but 2024's guide was cut and growth collapsed after the 2021 listing.
Per share capture3/4 →
The share count fell 10.9% in six months, but only by spending more cash than the business makes, and share pay equals free cash.
Potential2/4 →
Online shopping grows 12.2% a year, but nothing is contracted, the prize is not sized, and Q2's driver sits in a legally contested category.
Risk profile3/4 ↓
Five merchants are 32% of revenue and one 11%; event-contract exposure is contested in court and undisclosed.
Valuation2/4 ↓
22.1 times trailing free cash with nothing left after share pay: the price needs about 20% growth to continue after 2026's one-off events.
Evidence quality3/4 →
Figures are filed and fraud and e-commerce come from Nilson and Census, but the category split and win rate are management's word.
The Verdict
We rate Riskified neutral. It checks online orders for fraud and pays shops back when an approved order turns out to be fraud, a solid business that has started growing again. But the new growth comes from riskier shops that earn it less and from one busy sports summer, and the shares it pays staff eat about all the spare cash. After the shares doubled, the price is full, not cheap: it assumes that growth lasts, so a slowdown would hurt. The view hinges on whether growth near a fifth a year holds and share pay keeps falling, which the November results should start to show.
The long caseShopping keeps moving online and card fraud follows it; Riskified must keep winning shops that pay well, not only riskier ones, for that trend to reach owners.
The evidence under this - 6 claims
- The company does not file the category split; it is stated only on the call and cannot be checked against a filed table.
- Tickets were lifted by the World Cup and NBA Finals, which management itself calls a dense events calendar rather than a lasting change.
- Share pay is falling: 10.5% of revenue in Q2 2026 against 15.9% a year earlier, so the trailing figure overstates the run rate.
- Management treats share pay as non-cash; the cash cost appears only through the buybacks that offset it.
- Buybacks after 30 June would lower the share count and the market value; none is disclosed yet.
- On free cash after share pay (C10) the multiple is not meaningful, so the 22.1 flatters the price unless share pay keeps falling.
The company
When someone buys online with a stolen card, the shop usually loses the sale and the goods. Shops either run their own fraud teams, pay a scoring service that only gives advice, or hire Riskified.
Riskified looks at hundreds of details about each order and says yes or no within a fraction of a second. On the orders it approves, it takes the loss if they turn out to be fraud. Its customers are large retailers, ticket sellers and travel sites.
The evidence under this - 3 claims
How it makes money
A shop pays Riskified a set percentage of every order it approves, about 22 cents for each $100 of orders it checked in 2025. Out of that, Riskified pays back the fraud it missed.
So the business works only if it prices risk well: charge too little and payouts eat the fee; charge too much and shops leave. Big shops matter most: five of them paid 32% of the revenue.
The evidence under this - 4 claims
The CFO attributes the Q2 2026 rise to riskier new merchants priced higher and expects it to ease.
Where the money comes from, stream by stream
| Stream | Share of sales (FY2025) | How it earns it |
|---|---|---|
| Order review (ASC 606) | 59% | The fee for screening each order and approving or declining it, a percentage of approved order value. |
| Chargeback guarantee (ASC 460) | 41% | The part of the same fee booked as payment for standing behind approved orders, repaying the merchant if one proves fraudulent. |
Who runs it
Eido Gal co-founded Riskified in 2012 and still runs it and chairs the board; the finance chief has been there since 2015. Insiders hold close to half the votes through ten-vote founder shares, and Gal alone about a fifth. His yearly bonus rests on financial goals the company does not name, and his big stock award pays only if the price climbs several times over. The company has bought no other businesses and spent its spare cash buying back shares.
The evidence under this - 4 claims
- Insiders made 82 open-market sales and no purchases in the year to September 2026 (C16), so the stake is shrinking.
- Part of the group's shares are venture funds counted through their directors, such as Qumra through Erez Shachar, who disclaims beneficial ownership beyond his own interest.
Why it wins, and who could beat it
Why customers come back
Every shop that joins sends orders the computer models learn from, so Riskified can spot a fraudster at one retailer before he reaches the next. The sign it works: for three years it paid back a steady 40% to 42% of what it charged.
Fraud is worth guarding against. Card fraud costs about 6.43 cents on every $100 spent worldwide, and US shops bear far more than their share. But existing customers grew their spending only 5% in 2025, which is modest loyalty.
The evidence under this - 4 claims
- Retention of 105% is modest for a product customers supposedly cannot leave, and most Q2 2026 growth came from new merchants (C05, C06).
- Private rivals claim larger networks: Forter over 2 billion identities (C22).
- Gross margin fell 3.4 points in Q2 2026 as new merchants started at higher payouts (C04).
The threat
Others sell the same promise. Signifyd and Forter, both private, offer guarantees too. Stripe, which processes payments, has bundled one into its checkout since 2019 for 0.4% of each sale.
In Europe a rule requiring an extra login step moves most card fraud losses off shops altogether, which leaves less for a guarantee to cover.
The evidence under this - 4 claims
- Stripe's guarantee covers only its own checkout, so a merchant using several processors still needs an independent one.
- Riskified's growth is coming from high-risk categories and payment methods such as ACH transfers, where processor guarantees do not reach (C05).
The rivals: V, and 4 more that are private, in-house or listed abroad.
Who else does this, rival by rival
| Ticker | Rival | How far along | Priced against it |
|---|---|---|---|
| Private | Stripe Chargeback ProtectionPriced against it: Private | On sale | Private |
| Private | SignifydPriced against it: Private | On sale | Private |
| Private | ForterPriced against it: Private | On sale | Private |
| V | Visa Cybersource Decision ManagerPriced against it: One segment only | On sale | One segment only |
| Not listed | EU strong customer authentication under PSD2Priced against it: Not listed | On sale | Not listed |
How big it could get
How big this could get
Americans spent $340.2B online in Q2 2026, 12.2% more than a year earlier, according to the Census Bureau. Card fraud worldwide cost $33.41B in 2024.
Riskified keeps a fraction of a percent of the orders it checks. Nobody measures how much of online shopping is checked by outside fraud firms, so how big its slice could get cannot be counted.
The evidence under this - 3 claims
The CFO attributes the Q2 2026 rise to riskier new merchants priced higher and expects it to ease.
The results
The numbers
Sales grew 5.2% in the year to Dec 2025, about the same pace as the 10% in the year before.
Operating margin is −9.9%, up from −42% in the year to Dec 2022.
Running costs take 62¢ of each sales dollar, down from 78¢ in the year to Dec 2021.
Shareholders' money lost 8.1% a year, up from −74% in the year to Dec 2021.
It turned 9.6% of sales into free cash, $33.1M over the year.
Yearly sales growth has ranged from +5.2% to +35% over the last 5 years.
The share count rose 106% in 4.0 years, and pay in shares costs 15% 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 2021 | Dec 2022 | Dec 2023 | Dec 2024 | Dec 2025 | |
|---|---|---|---|---|---|
| Size and growth | |||||
| Sales | $229M | $261M | $298M | $328M | $345M |
| Sales growth | +35% | +14% | +14% | +10% | +5.2% |
| Profit | |||||
| Gross margin | 54% | 52% | 51% | 52% | 52% |
| Operating margin | −24% | −42% | −26% | −15% | −9.9% |
| Net margin | −78% | −40% | −20% | −11% | −8.0% |
| What growth costs, per sales dollar | |||||
| Research and development | 24% | 28% | 24% | 21% | 20% |
| Sales and marketing | 31% | 34% | 30% | 26% | 24% |
| General and admin | 23% | 32% | 23% | 20% | 17% |
| Pay in shares | 15% | 26% | 21% | 18% | 15% |
| Capital spending | 5.3% | 2.3% | 0.5% | 0.2% | 0.2% |
| Cash | |||||
| Cash from operations | −$20.3M | −$26.3M | $7.28M | $39.7M | $33.9M |
| Free cash flow | −$32.5M | −$32.4M | $5.92M | $39.1M | $33.1M |
| Balance sheet | |||||
| Cash and short-term investments | $418M | $189M | $470M | $371M | $293M |
| Returns and shares | |||||
| Shares (diluted) | 76.5M | 168M | 177M | 171M | 157M |
What growth costs
Growth has been making Riskified less unprofitable: its research, sales and office costs have shrunk while sales grew, so each new sale carries less overhead. The cost that does grow with sales is paying back merchants for fraud, and that is rising faster this year as riskier new merchants arrive.
The evidence under this - 3 claims
- Gross margin moved the other way, from 53.7% in 2021 to 51.7% in 2025 and 45.6% in Q2 2026 (C04), because the cost that grows with sales is chargeback payouts, and management guides 2026 gross profit growth below revenue growth (C39).
- Part of the fall in cost per sales dollar is headcount cuts (C33) rather than growth; a business that cut 3.3% of staff in a 5.2% growth year has not shown what the costs do when it hires again.
Management calls the lower margin a mix effect that new merchants grow out of, so gross profit could catch up with revenue in 2027.
The last report (12 Aug)
Quarter ended June 2026, filed figures.
Q2 2026 looked like a turn. Revenue rose 21.8%, the fastest in over four years, and management raised its forecast for the year a second time.
Under the surface, profit per sale fell. Gross margin, what it keeps after paying out fraud, dropped to 45.6% from 49.0%, because new shops start out riskier. Much of the growth came from betting and prediction-market sites and from ticket sales around the World Cup. It still lost $9.1M under standard accounting.
What it did to the thesisIt confirmed growth has returned, which supports the market's view. But the profit kept on each sale fell, and much of the lift came from betting sites and one summer's sporting events. The view stays Neutral until growth holds without those boosts.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $0.02 against $0.01 expected.
The evidence under this - 4 claims
- The company does not file the category split; it is stated only on the call and cannot be checked against a filed table.
- Tickets were lifted by the World Cup and NBA Finals, which management itself calls a dense events calendar rather than a lasting change.
Management calls the lower margin a mix effect that new merchants grow out of, so gross profit could catch up with revenue in 2027.
What management did with the money
Insiders in 2026Q2: 6 purchases and 48 sales, from their Form 4 filings.
The price, and where we disagree
In short. Full. Fair if the business keeps growing near 20% a year and pays staff less in shares; too much if the jump in Q2 2026 came mostly from one summer's events.
The shares doubled in six months, and most of that rise was buyers paying more for each dollar of sales rather than sales rising. The last leg, 24.6% in three days in late September, came with no news from the company.
The day after that run ended, an appeals court ruled that states may regulate sports prediction contracts, the kind of business behind Riskified's fastest growth. The next results, due 11 November, should show whether the growth lasts.
The evidence under this - 4 claims
The company guided a growth rate rather than a dollar range, so the bar is an approximation.
How we read the multiple
For each $1 a buyer pays, the business produced about 4 cents of spare cash over the last year, the cash left after running and investing in it. It also holds $223.6M of cash and no debt.
The catch is pay in shares. Staff received $45.8M of shares over the same year, about all the spare cash, so the company spends that cash buying shares back. The price works only if growth near 20% continues and share pay keeps falling.
Against its own history. It looks cheap on cash only because it used to burn cash. On sales and on what it owns, it costs more than its own normal.
The evidence under this - 4 claims
- Buybacks after 30 June would lower the share count and the market value; none is disclosed yet.
- On free cash after share pay (C10) the multiple is not meaningful, so the 22.1 flatters the price unless share pay keeps falling.
- Share pay is falling: 10.5% of revenue in Q2 2026 against 15.9% a year earlier, so the trailing figure overstates the run rate.
- Management treats share pay as non-cash; the cash cost appears only through the buybacks that offset it.
- The August raise rests on a quarter lifted by live-sports volume, which management says will not recur at the same level.
- In 2024 management lowered its revenue guidance in August, from $323M-$335M to $320M-$325M.
Priced against its rivals
None of the listed rivals above sells the same thing to the same buyers as a large part of what it does. A price comparison needs at least two, so the price is judged against its own history instead.
Also checked and set aside: Equifax (a substitute, not the same product).
Where we disagree with the market
Riskified's growth has turned back up for good, and the doubled share price reflects a business back on track.
The growth is real but leans on new, less profitable shops and one-off events, while share pay still takes all the cash.
The market is right that growth came back. Where we part company is on its quality: existing shops grew only 5%, the new ones earn Riskified less, and live sports gave the quarter a lift that will fade.
We would be wrong if Q3 2026 shows growth near 27% with the margin recovering, which would mean the new shops are seasoning faster than we expect.
What analysts say, and the evidence
The evidence under this - 6 claims
- The company does not file the category split; it is stated only on the call and cannot be checked against a filed table.
- Tickets were lifted by the World Cup and NBA Finals, which management itself calls a dense events calendar rather than a lasting change.
- Share pay is falling: 10.5% of revenue in Q2 2026 against 15.9% a year earlier, so the trailing figure overstates the run rate.
- Management treats share pay as non-cash; the cash cost appears only through the buybacks that offset it.
What could go wrong, and how we would know
The risks
- A few shops matter a lot. Five merchants paid 32% of revenue in 2025 and one paid 11%; big shops can move orders in-house.
- A contested category. Courts are split on whether states can stop sports prediction contracts, which drove much of Q2's growth.
- Payment firms bundle it. Stripe and card networks sell fraud screening with payments, which can make a separate guarantee unnecessary.
- Insiders only sell. Insiders sold 8.20M shares in a year and bought none, though most sales were scheduled in advance.
- Founders keep control. Shareholders from before the listing hold 84.34% of the votes through ten-vote shares.
The evidence under this - 7 claims
- Riskified does not say how much revenue comes from event contracts, and digital finance also includes money transfer and payments merchants.
- An adverse ruling may move volume between licensed operators rather than end it.
- Stripe's guarantee covers only its own checkout, so a merchant using several processors still needs an independent one.
- Riskified's growth is coming from high-risk categories and payment methods such as ACH transfers, where processor guarantees do not reach (C05).
What we still do not know
- How much revenue comes from betting and prediction-market sites; the company gives only the category's growth.
- Who Customer A is, the unnamed shop that paid 11% of 2025 revenue, and when its contract renews.
- What drove the 24.6% rise in late September; no filing or company news explains it.
The evidence under this - 3 claims
- The company does not file the category split; it is stated only on the call and cannot be checked against a filed table.
- Tickets were lifted by the World Cup and NBA Finals, which management itself calls a dense events calendar rather than a lasting change.
What would change our view
- Neutral→AttractiveThe price falls back on an unchanged business
It needs all of:
- ○The share price falls to about 15 times the cash the business produces.about $5.23 today, 32% below the last close
- ○Revenue keeps growing at least 15% a year.checked at the next results, due by 11 Nov
- Neutral→AttractiveGrowth holds and the margin and share pay mend
It needs all of:
- ○Gross margin climbs back to 49% or more in a quarter.checked at the next results, due by 11 Nov
- ○Share pay falls below 8% of quarterly revenue.checked at the next results, due by 11 Nov
- ○Revenue keeps growing at least 20% a year.checked at the next results, due by 11 Nov
- Neutral→UnattractiveThe base weakens
It needs any of:
- ○Existing shops spend less than a year earlier, taking retention below 100%.checked at the next results, due by 11 Nov
- ○The shop that pays over 10% of revenue leaves or takes its orders in-house.watched as it happens
We would also be wrong if
- Share pay stays above 10% of revenue after buybacks stop shrinking the share count.
What is coming
- 11 November 2026 Q3 2026 results: Whether revenue reaches about $104M, up 27%, with gross profit up 13% or more, and whether event-contract volume held after the Sixth Circuit ruling.
Every dated event ahead
- 11 November 2026 Q3 2026 results: Whether revenue reaches about $104M, up 27%, with gross profit up 13% or more, and whether event-contract volume held after the Sixth Circuit ruling.
- March 2027 Q4 and full-year 2026 results, with 2027 guidance: Whether growth near a fifth survives without 2026's sports calendar, and whether free cash beats share pay once the buybacks slow.
- March 2027 Form 20-F for 2026: The first disclosure of what the PSU program pays on, the year's top-merchant concentration, and how many shares insiders still hold.
Since we wrote this
- 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 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 the proxy.
- 29 SepThesis writtenNeutral, medium convictionGrowth has returned but leans on new, lower-margin merchants and one-off events, share pay equals the free cash, and the price already doubled on the recovery.
Also checked 2 times with nothing new to act on.
Sources 22
Sources - 22
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 filing2026-03-06Item 4 business and pricing model, merchant concentration (top three, top five, Customer A), revenue by region and by ASC 606 / ASC 460, CTB ratio, net dollar retention, GMV share of eMarketer's global figure, competition risk factors (classes only), in-house risk, insurance-classification risk, dual-class voting, buyback authorisation, share pay and free cash flow reconciliationsbacksC01C02C03C06C08C09C10C14C17C23C27C31C33C34C37C38C42C43C45C46
- S02primarycompany filing2026-08-12statements, GMV, gross margin, non-GAAP reconciliations, free cash flow, buyback, share counts, cash and investments, raised 2026 guidance, win ratebacksC04C07C10C11C12C13C14C25C32C38C40C46
- S03primarytranscript2026-08-12Q2 2026 earnings call transcript (FMP)data/cache/transcripts/RSKD-2026Q2.jsoncategory and regional growth, digital finance driven by event contracts and gaming, gross margin explanation, net dollar retention expectation, Q3 growth guide, gross profit guide, opex guide, buyback price; 7 analyst turns from 4 analysts, none on concentration, share pay or the event-contract category's legal status
Our digital finance category grew approximately 180% year-over-year, driven primarily by the ramp of multiple new merchants onboarded in the quarter to the event contracts and gaming of vertical
backsC05C06C07C11C30C39C40 - S04primarycompany filing2026-05-13Q1 revenue and GMV growth, first raise of 2026 revenue guidance to $376M-$384M, Q1 buybackbacksC07
- S05primarycompany filing2026-03-04initial 2026 guidance of $372M-$384M revenue and $26M-$34M adjusted EBITDA, 2025 buyback totalbacksC07
- S06secondaryvendor data2026-09-29RSKD research pack (FMP market tier, FINRA short interest, headcount), frozen 2026-09-29content/RSKD/research-pack.jsonclose of $7.70 on 2026-09-28, 126- and 252-session price decomposition, daily bars, analyst split and target drift, short interest, headcount by 20-F, earnings datesbacksC13C14C15C16C30C33C34C35C36C41
- S07primarycompany filing2026-09-29every filing since the Q2 2026 release: Forms 4 and 144 only, no 6-K between 2026-08-12 and 2026-09-29backsC15
- S08primarycompany filing2026-09-1782 open-market sales of 8.20M shares, zero purchases; the Rule 10b5-1 flag on eachbacksC16
- S09primarygovernment data2026-08-19US retail e-commerce sales, growth and share of total retail
the estimate of U.S. retail e-commerce sales for the second quarter of 2026, adjusted for seasonal variation, but not for price changes, was $340.2 billion
backsC18 - S10primaryindustry data2026-01-07worldwide card fraud losses for 2024, losses per $100 of volume, the US share of volume and of losses
Payment card fraud losses worldwide dipped 1.2% to $33.41 billion in 2024
backsC19 - S11primaryregulator2025-12-15ECB press release: joint EBA-ECB report on payment fraud European Central Bankeffect of strong customer authentication on card fraud in the European Economic Area; 2024 card fraud losses
Notably, card payment fraud was 17 times higher when the payment recipient was outside the EEA, where SCA is not legally required and often not used.
backsC20 - S12primarycompany release2019-06-03a payment processor's own bundled fraud guarantee and its price
Chargeback Protection costs just 0.4% per transaction.
backsC21 - S13primarycompany release2026-09-29a private rival's guarantee on approved orders
Approved orders are backed by a 100% financial guarantee, with any fraudulent chargebacks resulting from a guaranteed order reimbursed within 48 hours.
backsC22 - S14primarycompany release2026-09-29Forter home page Fortera private rival's identity network size
With a network of 2B+ identities, Forter helps you recognize trusted customers instantly and stop bad actors before they strike.
backsC22 - S15primarycompany filing2025-11-06Visa sells its own merchant fraud tool, Cybersource Decision Managerread, but no claim rests on it
- S16primaryregulator2026-09-25US Court of Appeals for the Sixth Circuit, KalshiEX LLC v. Schuler, Nos. 26-3196/5235 US Court of Appeals for the Sixth Circuitthe court's holding that states may apply gambling law to sports event contracts
We hold that Kalshi has not shown that its sports-event contracts satisfy the statutory
backsC28 - S17primarycompany filing2024-08-142024 revenue guidance lowered to $320M-$325MbacksC34
- S18primarycompany filing2024-05-152024 revenue guidance of $323M-$335M, the range the August cut came frombacksC34
- S19primarycompany filing2026-10-04Riskified XBRL company facts, 2020 to 2025, frozen as content/RSKD/financials.jsoncontent/RSKD/financials.jsonrevenue, gross profit, research and development (ResearchAndDevelopmentExpense), sales and marketing, general and administrative, share pay, capital spending, operating income, cash and diluted share counts per yearbacksC42C46
- S20primarycompany filing2026-06-30beneficial ownership and voting power at the 17 June 2026 record date, the compensation methodology (share of CEO pay that is variable, the PSU program, the FW Cook peer group), board composition and committeesbacksC43C44C45
- S21primarycompany filing2026-04-03the private repurchase of 1,000,000 Class A shares from Pitango funds at $4.26, about $4.3M, approved by the audit committee and boardbacksC46
- S22secondaryvendor data2026-10-04Earnings calendar saved by the daily routine, content/calendar/2026-10-04.json (FMP earnings, per symbol)content/calendar/2026-10-04.jsonthe next results date, 11 November 2026
"RSKD": {"archived": false, "next": {"date": "2026-11-11"
read, but no claim rests on it