Stripe Radar Chargeback Prevention: Plans, Pricing & Rules (2026)

rétrofacturation?
Ce n'est plus votre problème.
Récupérez 4 fois plus d'rétrofacturation s et PRÉVENTION jusqu'à 90 % des messages entrants, grâce à l'IA et à un réseau mondial de 20 000 commerçants.

En bref :
- Scope: Radar helps reduce fraud disputes and early fraud warnings. It does not fix "not received", "not as described", or subscription cancellation disputes, which need operational fixes and evidence.
- Plans: Radar Lite covers card payments and card testing. Custom rules, backtesting, risk scores, and Adaptive 3D Secure start on Plus, and dynamic risk thresholds start on Pro.
- Risk settings: Maximize protection blocks payments likely to draw early fraud warnings, which Visa counts toward VAMP, so it fits merchants near a threshold.
- Rules: backtest on 6 months of payments, then roll out with traffic allocation. A rule at 0% runs in shadow mode.
- Cost and dates: On US standard pricing, Stripe lists a $15 dispute received fee and Radar pay as you go from $0.05 per screened transaction. Custom rules cover all payment methods from October 21, 2026, and new pricing for users covered by the transition notice starts January 22, 2027.
Stripe Radar chargeback prevention reduces fraud-related disputes by screening payments and letting you block, review, or authenticate risky transactions. It does not resolve delivery, product-quality, or cancellation complaints, and a low risk score does not guarantee that a payment will never be disputed. Use this guide to compare Radar plans and pricing, choose risk settings, test rules, and measure fraud prevention alongside payment acceptance.
For what Stripe's separate Chargeback Protection product covers, 3D Secure liability, and Verifi and Ethoca alerts, read Stripe Chargeback Protection and Prevention. If a dispute is already open, use the Stripe chargebacks guide. For prevention before payment acceptance, start with the Radar configuration below.
{{cta}}
What Stripe Radar Can and Cannot Prevent
Match the dispute type to the control before you spend time tuning rules. Radar acts at payment time, so it helps most where the problem is the payment itself.
| Type de litige | What Radar can do | Additional control |
|---|---|---|
| Stolen or unauthorized card | Score the payment; block, review, or request 3D Secure where supported. | Tune risk settings and rules to your traffic, and review flagged payments promptly. |
| Fraude amicale | Limited: payment screening cannot reliably establish a cardholder's intent. | Retain order, delivery, and customer history. Eligible Visa cases may use separate Compelling Evidence 3.0 deflection. See friendly fraud. |
| Product not received or not as described | Does not fix fulfillment or product-expectation problems. | Set clear shipping and return policies; retain tracking and customer communications. |
| Canceled subscription or confusing billing | Does not fix cancellation failures or an unclear statement descriptor. | Offer clear cancellation, renewal reminders, and a recognizable billing descriptor. |
| Test de cartes | Card testing defenses are included on every Radar plan. | Add checkout rate limits and bot controls to complement payment screening. |
Stripe Radar Plans: Lite vs Standard vs Plus vs Pro
Stripe sells Radar in four plans. According to the Stripe Radar documentation, businesses block 42% more fraud on average on Standard than on Lite. For chargeback prevention, these are the differences that matter.
| Capacité | Lite | Standard | Plus | Pro |
|---|---|---|---|---|
| Card fraud and card testing prevention | Oui | Oui | Oui | Oui |
| Expanded fraud prevention across supported payment methods | Non | Oui | Oui | Oui |
| Default rules and risk levels | Non | Oui | Oui | Oui |
| Risk scores (0 to 99) | Non | Non | Oui | Oui |
| Custom rules and backtesting | Non | Non | Oui | Oui |
| Manual payment review queue | Non | Non | Oui | Oui |
| Risk settings and Adaptive 3D Secure | Non | Non | Oui | Oui |
| Dynamic risk thresholds | Non | Non | Non | Oui |
If you want to write rules for your own dispute patterns, you need Plus or above, because custom rules and backtesting start there. Standard is enough if you only want Stripe's default models. Pro also adds customer abuse signals for free trial abuse, bots, and multi-account behavior, which matter mostly to subscription businesses.
Stripe Radar Pricing and the 2026 to 2027 Changes
Stripe's pricing page lists Radar from $10 per month, or pay as you go from $0.05 per screened transaction. The Radar pricing page lists Standard from $10, Plus from $14, and Pro from $20 per month for businesses. Radar Lite is included at no additional charge for businesses on Stripe's standard payments pricing. These are advertised starting prices for businesses on the US pricing pages, checked October 11, 2026. Monthly subscriptions and pay-as-you-go are alternative billing options; usage, country, contracts, and platform pricing can change your cost. Confirm the plan and effective fees in your Dashboard.
Compare that with the dispute side. On US standard pricing, Stripe lists $15 per dispute received and a separate $15 fee when you respond manually; the countered fee is returned if you win. At $0.05 per screened transaction, one dispute fee equals the screening cost of 300 payments. This is a fee-only illustration, not a savings forecast: screening also covers legitimate payments and does not prevent every dispute. Lost revenue, goods, and operating costs are additional.
For users who were on Radar or Radar for Fraud Teams with standard pricing before July 29, 2026, Stripe has announced the following changes in its support pages on Radar pricing (January 2027) and Radar for LPMs and Billing (October 2026):
- July 29, 2026: users on Radar or Radar for Fraud Teams with standard pricing get a free trial of Radar Standard, and Radar for Fraud Teams users move to Radar Plus.
- October 21, 2026: Radar screens all Stripe supported payment methods, including wallets, bank debits, and buy now pay later, and custom rules apply to all of them by default. Use the
:payment_method_type:attribute to limit a rule to specific methods. Stripe says you can ask Support to opt out of screening the newly covered methods before this date. - January 22, 2027: the new pricing takes effect and per-transaction charges begin as trials end. Fee reports rename Radar usage charges to "Radar Screened Transaction Fee".
- Subscriptions: Radar pricing applies to every transaction in a subscription, not only the first. You can limit screening to a set number of billing cycles in the Dashboard. Radar always screens the first payment.
If your current rules were written for card payments, review them before October 21, 2026, because they will start evaluating other payment methods.
How to Configure Stripe Radar Risk Settings
On Plus and above, a risk setting sets the blocking thresholds for several risk controls at once. Stripe shows the estimated effect, based on your last four months of payment data, before you apply a change. Details are in the Radar risk settings documentation.
| Risk setting | Priority | Merchant use case |
|---|---|---|
| Maximize protection | Block payments likely to generate early fraud warnings. | Consider during a fraud spike or when approaching a relevant monitoring threshold; review projected acceptance impact. |
| Balance risk and revenue | Balance fraud protection with payment acceptance. | A starting point for stable merchants; compare the estimate with your own fraud and conversion results. |
| Maximize revenue | Prioritize acceptance while blocking high-risk payments. | Consider where fraud remains low and additional checkout friction would materially affect sales. |
The risk controls behind those settings each target a different outcome:
- Fraudulent card payments: blocks payments Radar judges likely to be fraud at payment time, even if they would not result in a dispute. Included from Standard.
- Fraudulent dispute: blocks payments likely to produce fraudulent disputes. You can set a custom threshold and backtest it, which puts the account in Manual mode.
- Early fraud warning: blocks payments likely to draw an early fraud warning. It is available on the Balanced and Maximize protection settings, and Stripe recommends it if you are in a card brand monitoring program that counts early fraud warnings toward VAMP.
- Adaptive 3D Secure: uses machine learning to authenticate medium-risk payments. When 3D Secure authenticates a payment, liability for fraud disputes typically shifts to the issuer.
- Dynamic risk thresholds (Pro, cards only): temporarily blocks more elevated and high-risk payments while your account is under fraud pressure, then returns to your default.
Match the Score to the Outcome
All three scores run from 0 to 99. A higher score means a greater likelihood of that specific outcome.
Payment fraud
Predicts whether the payment itself is fraudulent, even if it never produces a dispute.
Applies to:
Card payments
:fraudulent_payment_score: Fraud dispute
Predicts whether the payment will result in a fraud-related dispute.
Applies to:
Cards, ACH & SEPA Direct Debit
:fraudulent_dispute_score: Early fraud warning
Predicts whether the issuer will report suspected fraud. This signal matters for Visa VAMP.
Applies to:
Card payments
:early_fraud_warning_score: Your custom rules remain active. Risk settings and risk controls do not override the rules you created.
The legacy default rule is retired. Selecting a risk setting disables the old high-risk block rule, which cannot be re-enabled. You can still write custom rules using the legacy risk score.
Stripe Radar Rules for Chargeback Prevention
Use the examples below as starting points, then backtest them against your own payments. A Radar rule combines an action, a condition, and the attributes that describe a payment. Stripe supports up to 200 transaction rules. The examples come from the Stripe Radar rules documentation, with the dispute-score template marked as illustrative.
Request authentication
Request 3D Secure
if :risk_level: != 'normal'
and :amount_in_usd: > 25Stripe documentation example. A 3DS request alone does not block payment methods that cannot authenticate.
Block failed authentication
Block
if :is_3d_secure:
and not :is_3d_secure_authenticated: Does not block every unauthenticated payment: cards without 3DS, exemptions, off-session payments, and wallets can pass.
Review a focused segment
Review
if :is_disposable_email:
and :card_funding: = 'prepaid'Backtest whether the matched payments justify human review.
Narrow a country rule
Block
if :card_country: != 'US'
and :risk_level: = 'elevated'Stripe example for a US-focused business. Adjust geography to your actual market.
Review predicted dispute risk
Review
if :fraudulent_dispute_score: > [threshold] Illustrative template: replace [threshold] with a numeric value chosen from your backtest.
Restrict allow rules
and :risk_level: != 'highest'Add this condition to an existing allow rule. An allow rule can override other matching rules and Stripe risk assessment. Access may require contacting Stripe.
Review rules put successful payments in a queue for your team without stopping them. Stripe's guidance is to start with a review rule when you are unsure, check for false positives, and convert it to a block rule only after you confirm the pattern. Rules apply only to future payments.
Roll Out a Rule Without Hurting Good Customers
- Backtest. The rule editor simulates the last 6 months of charges and shows how many legitimate, fraudulent, and blocked payments the rule would have affected.
- Read the mix. A good block rule stops significantly more fraudulent payments than legitimate ones. A mixed result points to a review rule instead.
- Set traffic allocation. Start with a small percentage of matching payments. Retries of the same transaction get the same outcome. At 0% the rule runs in shadow mode and reports what it would have done without affecting customers.
- Ramp up on evidence. Check the rule metrics: payments matched, estimated false positive rate and fraud prevented for block rules, and disputes from overrides for allow rules.
- Audit on a schedule. Remove rules that no longer match payments, and look for block rules whose volume rises while fraud stays flat, which suggests they now block good customers.
Consider reviewing before capture. Stripe notes that issuers must report possible fraud on captured payments even if you later refund them, but they are not required to report it for an authorization. If you reverse a suspicious authorization before capture, it is not reported. Separating authorization from capture and pairing it with review rules gives you a window to catch fraud before it can become an early fraud warning. Check authorization hold limits for your payment methods and the delay it adds to fulfillment before relying on it.
How Radar Connects to Visa VAMP and Mastercard Monitoring
Early fraud warnings are messages that issuers on Visa, Mastercard, and JCB send when they suspect fraud. They are not disputes, but Visa counts them in VAMP, alongside disputes. According to the Stripe monitoring programs documentation, a fraudulent transaction that appears in both the TC40 fraud report and the TC15 dispute report is counted twice. That is why blocking fraud at payment time matters more than winning the dispute later.
| VAMP metric | Non-compliant level in Stripe docs | Excessive: US, Canada, EU, AP, LAC | Excessive: CEMEA |
|---|---|---|---|
| Monthly fraud and dispute count | 5 | 1,500 | 150 |
| VAMP ratio: count / captured Visa payments | 0.5% | 1.5% | 2.2% |
| Monthly fraud and dispute volume | Sans objet | Sans objet | $75,000 |
These are the regional criteria listed in Stripe's monitoring documentation as checked on October 11, 2026. Evaluate the applicable criteria together and confirm your account's status with Stripe; a percentage alone does not establish program placement.
Visa excludes disputes resolved through pre-dispute products, and fraud reports that qualified for Compelling Evidence 3.0. On the Mastercard side, ECP compares chargebacks raised in the current month with captured Mastercard payments in the preceding month. Stripe lists the Excessive Chargeback Merchant program at 100 to 299 chargebacks and a 1.5% to 2.99% rate, with fines starting at $1,000 in month two and rising in later months. Stripe also flags dispute activity above 0.75% as a general warning level. It is not a universal network threshold, and sudden increases can prompt Stripe to act earlier. For the full threshold set, see Chargeback Thresholds 2026 and our VAMP rules guide.
Refunds do not remove a dispute from these counts. Stripe says monitoring programs do not consider refunds or dispute outcomes, and that all disputes, won or lost, count toward the rate. Radar keeps fraud from settling. Disputes already in motion need the alert and resolution tools covered below.
3D Secure is not immunity either. Visa's Secure Excessive Fraud Program applies to US-based businesses with US-based custom accounts and domestic Visa 3D Secure transactions on US-issued cards. Stripe lists monthly thresholds of $75,000 in early fraud warning volume and a 0.9% fraud rate for those transactions. The consequence is losing liability shift on domestic 3D Secure transactions until you exit. Stripe allows authenticated 3D Secure payments by default, so consider rules that block authenticated payments Radar flags as high risk.
Measure Radar With the Right Ratio
Stripe reports two different numbers, explained in its measuring disputes documentation. Dispute activity counts disputes by the date they arrive. Dispute rate counts them by the date of the original charge. Say you process 1,000 payments in a week and receive 10 disputes, but only 3 come from that week's payments. Dispute activity is 1% and dispute rate is 0.3%. Card networks use the activity calculation for monitoring programs.
Use dispute rate to judge a rule change, because it shows which sales produced disputes. Allow for lag: cardholders can generally dispute a charge within 120 days of payment, and sometimes later, so recent dates keep changing. Track the review approval rate, disputes from approved reviews, estimated false positive rate, and authorization rate together. A rule that cuts disputes while authorization rate falls is trading fraud for lost sales. For the formulas, see how to calculate your chargeback ratio.
Cover the Gaps With Chargeback Alerts and Recovery
Radar covers prevention at payment time. Two other layers handle what it misses. Stripe's dispute prevention lets you set resolution rules through Radar to refund specific disputes automatically, and disputes resolved this way do not count toward your dispute rate and do not incur a dispute received fee. On US standard pricing, Stripe lists Visa resolution and Compelling Evidence 3.0 blocks at $15 each and Mastercard resolution at $29 each. For disputes that still arrive, Smart Disputes submits evidence for eligible cases and charges 30% of the disputed amount on wins, per Stripe's pricing page. See what Smart Disputes covers and misses and the Stripe chargeback protection guide for alerts and 3D Secure liability.
Pair Radar with a post-purchase workflow. Chargeflow Alerts uses Verifi and Ethoca alerts to resolve eligible cases before they become chargebacks. Chargeflow Automation collects evidence and submits responses for disputes that still arrive through the Stripe integration. Check alert coverage and assign each dispute to one submission workflow to avoid duplicate responses.
{{cta}}
Stripe Radar Chargeback Prevention FAQs
Does Stripe Radar prevent chargebacks?
Stripe Radar helps reduce fraud-related chargebacks by blocking, reviewing, or authenticating risky payments. It does not guarantee that an accepted payment will never be disputed, and it does not fix delivery, product-quality, or subscription-cancellation problems.
How much does Stripe Radar cost?
Stripe lists Radar from $10 per month, or pay as you go from $0.05 per screened transaction. Its Radar pricing page lists Standard from $10, Plus from $14, and Pro from $20 per month for businesses. Radar Lite is included at no extra charge for businesses on standard payments pricing. These are US advertised starting prices checked October 11, 2026. For users covered by Stripe's transition notice, new pricing starts January 22, 2027; confirm your effective rate in the Dashboard.
What is the difference between Radar Lite, Standard, Plus, and Pro?
Lite covers card payments and card testing. Standard adds all payment methods, default rules, and risk levels. Plus adds risk scores, custom rules with backtesting, manual reviews, risk settings, and Adaptive 3D Secure. Pro adds dynamic risk thresholds and customer abuse signals.
Which Radar risk setting should I choose?
Choose Maximize protection if you are near a monitoring program threshold or just had a fraud spike. Balance risk and revenue suits most stable merchants. Maximize revenue fits low fraud history and conversion-sensitive checkouts. Stripe shows an estimate based on your last four months of payments before you apply a change.
How do early fraud warnings affect my account?
Visa counts early fraud warnings in VAMP alongside disputes, and a transaction that appears in both the fraud report and the dispute report is counted twice. Turn on the early fraud warning risk control if you are in a monitoring program.
Can I test a Radar rule before it goes live?
Yes, on Plus and above. The editor backtests against the last 6 months of charges. Traffic allocation lets you apply a rule to a small share of matching payments, or run it at 0% in shadow mode.
Does Radar stop friendly fraud?
Radar cannot reliably identify a legitimate customer's intent to dispute a purchase. Screening repeat abuse patterns may help, but eligible Visa Compelling Evidence 3.0 deflection is a separate evidence-based control. Retain order, delivery, and customer history for the disputes that still arrive.
Do Radar rules apply to payment methods other than cards?
From October 21, 2026, Stripe says custom Radar rules apply to all supported payment methods by default. Use the :payment_method_type: attribute to limit a rule to specific methods.
Does 3D Secure prevent every Stripe chargeback?
No. Successful 3D Secure authentication typically shifts liability for fraud-related disputes to the issuer, subject to network rules and exceptions. It does not stop disputes about delivery, product quality, or canceled subscriptions.
Is Stripe Radar the same as Chargeback Protection?
No. Radar screens payments to reduce fraud risk. Stripe Chargeback Protection is a separate product with account-specific availability and terms for qualifying protected charges. Chargeback alerts and evidence submission address different stages of the dispute lifecycle.
Tune Radar First, Then Cover the Gaps
Start with the risk setting that matches your monitoring position, add rules one at a time with backtesting and traffic allocation, and track dispute rate rather than raw dispute counts. Then add alerts and an evidence workflow for the disputes Radar cannot stop. For the wider prevention playbook, see our ecommerce fraud prevention guide and the 30/60/90 plan to reduce ecommerce chargeback rates.
Win Back Chargebacks on Stripe, Automatically
Connect your Stripe account to Chargeflow's automated dispute workflow and recover chargebacks without manual evidence work.

rétrofacturation?
Ce n'est plus votre problème.
Récupérez 4 fois plus d'rétrofacturation s et PRÉVENTION jusqu'à 90 % des messages entrants, grâce à l'IA et à un réseau mondial de 20 000 commerçants.













.png)

