book illustration
academic cap illustration
microscope illustration
lab illustration

What is Riskified’s business model?

Riskified is performance-based, meaning retailers are charged only for orders approved and guaranteed by our system. This ensures Riskified’s goals are aligned with yours – safely growing online revenues. We take responsibility for our decisions by covering every approved transaction with a full chargeback guarantee. This is coupled with an approval rate guarantee, under which Riskified commits to certain rate of approved orders out of the overall volume of orders submitted for fraud review.

Which eCommerce platforms and integrations do you support?

Riskified is available to all merchants through our API Integration. We also have extensions available for the Magento and Demandware eCommerce platforms and one-click integration with Shopify.

Which industries and markets does Riskified serve?

Riskified’s solution is designed to meet the needs of global enterprise retailers. We work with the world’s leading brands and merchants in a wide variety of online verticals, including apparel, luxury goods, cosmetics, gift cards, electronics, travel, remittance, ticketing, sneakers, jewelry, and more. We have a proven track record effectively managing fraud in online orders across geographical regions, with localized data sources and experience reviewing transactions from over 186 countries.

Does your solution support expedited/same day shipping?

Riskified supports expedited and same-day business flows, as well as in-store pickup. This is enabled by our accurate, instant decisioning, which leaves merchants enough time to fulfill orders in due time.

What tools are available for tracking and reporting Riskified data?

Merchants using Riskified gain access to our comprehensive, user-friendly web app which allows merchants to review performance KPIs in real time, track order populations, approval rates, chargeback rates and more. Filters can also be used to segment data based on multiple criteria (e.g. geographies, channels, device types, operating systems, and payment methods). All data is exportable, which makes reporting simple and efficient

Can Riskified prevent ATO attacks?

Riskified leverages advanced behavioral analytics and device fingerprinting capabilities to detect and preempt Account Takeover attacks. We collect a wide range of data points anytime a user logs in, updates an account, resets a password or uses loyalty points. Any discrepancies between standard behavior, devices used or geographical indicators will trigger an alert.

How am I notified regarding Riskified’s decision?

Riskified sends decisions automatically to your notification endpoint. In the event of a “decline” decision, Riskified provides decision insights explaining what factors contributed to the decision. You can also search for a specific order in your Riskified account anytime.

How long does it take to decide an order?

Riskified reviews orders in real time and provides “approve” or “decline” decisions instantly.

How do you decide whether an order is fraudulent or not?

Riskified reviews every submitted order using machine learning models and proprietary analytical tools. Our algorithms consider data such as the customer details, proxy usage, IP and geolocation, device and browser fingerprinting, and behavioral analytics. This information is then cross-checked against millions of other orders to determine a risk profile, and decide whether to approve or decline.

How does the chargeback guarantee work? What does it cover?

We cover all chargebacks related to unauthorized card usage or stolen financial information. This includes the order amount, shipping costs, and Riskified’s approval fee. We do not cover chargebacks for reasons such as “item not received” or “item not as described,” as they are not related to unauthorized credit card usage.

How does Riskified’s pricing work?

Riskified’s pricing is performance-based, meaning merchants only pay for revenue-generating orders we approve. Merchants never pay for orders that Riskified does not approve. We charge a percentage of the value of each approved transaction, dependant on sales volume and industry-specific risk profiles. To receive a custom quote, complete the form on our Pricing page.

What sets Riskified apart from other solutions?

Riskified has unparalleled outreach across geographies, industries, and products (both digital and physical). Our automated process delivers decisions – not scores or recommendations – instantly. And we stand behind these decisions with a full chargeback guarantee.
We develop machine learning technology to enhance performance over time and relieve our merchants of the need to maintain a system, so they can focus on their core business.
Riskified is transparent, providing a complete review of performance 24/7, including decision insights, data segmentation & analysis, and full reporting capabilities. Finally, Riskified is future proof – constantly advancing to address market trends, evolving business rules, and changing regulations – a true partner for growth and expansion.

What types of transactions does Riskified review and guarantee?

Riskified can review all card-not-present orders placed using a credit card, debit card, PayPal, Apple Pay, Alipay, Amazon Pay, and more. We manage payments in all currencies, from all channels – desktop, mobile, and phone orders of both tangible and digital goods. Orders placed with payment methods that cannot incur a chargeback should not be submitted to Riskified for review (e.g. Bitcoin, Cash on Delivery).

Does Riskified reach out to my customers?

Riskified believes in a frictionless order review process to promote a smooth shopping journey, and therefore does not contact customers as part of the order review process. Our process is completely invisible to end customers and our analysis tools reach decisions without having to call, email or text customers.

How does Riskified’s machine learning differ from other machine learning solutions?

Riskfied’s machine learning models relieve merchants from having to engineer features, change thresholds, retrain models or invest in experimentation for optimization. Our models are customer-specific, and modular. Orders are assessed by the most relevant models, which makes learning more efficient – as the models only learn based on performance from similar industries, products, and geographies.