--- title: "Configure automated risk" description: "Learn about automated risk, and how to enable it for a risk profile." url: "https://docs.adyen.com/risk-management/configure-automated-risk" source_url: "https://docs.adyen.com/risk-management/configure-automated-risk.md" canonical: "https://docs.adyen.com/risk-management/configure-automated-risk" last_modified: "2022-05-19T16:02:00+02:00" language: "en" --- # Configure automated risk Learn about automated risk, and how to enable it for a risk profile. When you use automated risk, RevenueProtect evaluates transactions based on machine learning. Our machine learning model is trained to predict the probability that a payment will result in a fraudulent chargeback. Automating part of your risk assessment can save you time managing risk. Because automated risk is optimized to evaluate payments that could result in a fraudulent chargeback, we recommend that you supplement and customize your risk profile with [manual risk rules](https://docs.adyen.com/risk-management/configure-automated-risk#automation-manual-rules) to address unique risks faced by your business. ## Requirements To be able to use automated risk: * You have to enable RevenueProtect premium in the risk settings for your company account. * Adyen needs to have processed payments for your company for some time to have sufficient data on risk refusals, fraudulent chargebacks, notifications of fraud, and false positive rates. Before you [turn on automated risk](https://docs.adyen.com/risk-management/configure-automated-risk#automation-on): * We strongly recommend that you **create a new risk profile** or **make a copy of your existing risk profile** for automated risk. This way, you can refer back to your current manual risk rule configuration settings if needed. * You can [run an experiment](https://docs.adyen.com/risk-management/configure-automated-risk#automation-experiment) to compare an automated with a manual risk profile, or to evaluate two automated risk profiles with different risk appetite settings. When you turn on automated risk, we disable ShopperDNA and velocity risk rules. If your risk profile is based on an existing manual risk profile, we also disable any referrals or consistency risk rules with a risk score below 100. You can continue to use any configured custom rules, or create new ones to [fine-tune your profile](https://docs.adyen.com/risk-management/configure-automated-risk#automation-fine-tune). ## Turn on automated risk 1. Log in to your [Customer Area](https://ca-test.adyen.com/). 2. Go to **Risk & disputes** > **Risk profiles**. 3. Select a risk profile. 4. Select the **Automated risk** tab. If you do not see the **Automated risk** tab, your company or merchant account is [not eligible](https://docs.adyen.com/risk-management/configure-automated-risk#before-you-begin-automated-risk) for automated risk. 5. Select **Enable automation**. When enabled, you can set or change your [risk appetite](https://docs.adyen.com/risk-management/configure-automated-risk#automation-appetite) for automated risk, and [fine-tune the risk profile](https://docs.adyen.com/risk-management/configure-automated-risk#automation-fine-tune) for your business using manual risk rules. ## Set your risk appetite You can influence how strict you want the decision to be that automated risk makes. You can choose the following settings: * **Dynamic**\ This is the recommended setting. All merchant accounts using the automated profile will usually start with a tolerant risk appetite. The system adjusts the risk appetite for each merchant account using this profile, depending on their fraud levels. * **Strict** (static)\ This setting allows fewer payments to block more fraud. This may lead to an increase in the number of legitimate payments being blocked. All merchant accounts using the automated profile will have the same strict setting. * **Tolerant** (static)\ This setting allows more transactions to increase your authorization rate. This may also increase fraud. All merchant accounts using the automated profile will have the same tolerant setting. ## Fine-tune your profile with manual risk rules Our machine learning model is trained to predict the probability that a payment will result in a fraudulent chargeback. This means that the model is currently not trained to address other types of risks, for example around transactions that cannot result in a chargeback such as [zero-value auth](https://docs.adyen.com/get-started-with-adyen/adyen-glossary/#zero-value-auth) requests. Similarly, automated risk does not evaluate transactions made with payment methods that cannot be disputed, or for which disputes are handled outside Adyen's dispute flow. To address the risk around these types of transactions, and any other risks specific to your business, you can configure manual risk rules for your automated risk profile. ### Enable or re-enable risk rules You can enable and configure these [risk rules](https://docs.adyen.com/risk-management/configure-manual-risk) in the **Risk rules** tab: * [Block and trust lists](https://docs.adyen.com/risk-management/configure-manual-risk/referral-rules) * [Consistency rules](https://docs.adyen.com/risk-management/configure-manual-risk/standard-risk-rules#consistency-rules) * [Custom risk rules](https://docs.adyen.com/risk-management/configure-manual-risk/configure-custom-risk-rules) If your automated risk profile is based on an existing [manual risk profile](https://docs.adyen.com/risk-management/create-and-use-risk-profiles), and contains block and trust lists or consistency risk rules with a risk score **below 100**, you can re-enable any of these rules in your Customer Area. We also recommend that you set up a number of [custom risk rules](https://docs.adyen.com/risk-management/configure-manual-risk/configure-custom-risk-rules). With custom rules, you can block transactions by specifically targeting fraudulent behavior or patterns that you know to be specific for your business, for example: * Detecting reseller fraud. * Preventing fraudulent card testing activity using zero-auth transactions. * Minimizing the amount of retries to prevent costs connected to excessive retries. ### Extra risk management features You can continue to use extra risk features such as [case management](https://docs.adyen.com/risk-management/case-management) and [Dynamic 3D Secure](https://docs.adyen.com/risk-management/dynamic-3d-secure). You configure [Dynamic 3D Secure](https://docs.adyen.com/risk-management/dynamic-3d-secure) rules separately for each merchant account, and these rules are not part of the risk profile. You can use custom rules to trigger a Dynamic 3D Secure rule. If a risk profile is based on a copy of an existing risk profile, any existing custom rules are also copied, but the link with Dynamic 3D Secure is lost. This means that for each merchant account using a new risk profile, you have to re-configure any Dynamic 3D Secure rules that should trigger on a custom rule. Where needed, you have to re-add the custom rule to the Dynamic 3D Secure rule. To check if your merchant account already uses Dynamic 3D Secure: 1. Log in to your [Customer Area](https://ca-test.adyen.com/), and switch to the merchant account that uses the risk profile. 2. Go to **Risk & disputes** > **Dynamic 3D Secure**: * If there are no existing rules, you can add new [Dynamic 3D Secure](https://docs.adyen.com/risk-management/dynamic-3d-secure#configuring-dynamic-3d-secure-rules) rules. * If there are existing rules, make sure that you check if they should trigger on a custom rule. If so, you have to re-add the applicable custom rule in the **Risk check** field, and then select **Save**. ## View automated risk results You can see the automated risk decisions on the **Risk results** page. To open the risk results page: 1. Log in to your [live Customer Area](https://ca-live.adyen.com/), and switch to a merchant account using an automated risk profile. 2. Go to **Payments** > **Payment list**. 3. Select the **Risk score** for a payment from the payments overview to open the risk results page. You can see the decision outcome at the very top of the page. The decision outcome can be **Looks fraudulent** or **Looks safe**. You can [analyze the fraud signals](https://docs.adyen.com/risk-management/configure-automated-risk#automation-signals) that contributed to the decision for that payment from the same page. To see a list of payments that were refused by automated risk: 1. Log in to your [live Customer Area](https://ca-live.adyen.com/), and switch to a merchant account using an automated risk profile. 2. Go to **Risk & disputes** > **Risk profiles**. 3. Select the **Automated risk** tab. 4. Select **View transactions list**. ## Analyze fraud signals You can see an explanation of the reasoning behind the decision outcome in the **Fraud Signal Analysis** section. These elements help explain the decision outcome: the [signal bar](https://docs.adyen.com/risk-management/configure-automated-risk#automation-signal-bar), the [signal category](https://docs.adyen.com/risk-management/configure-automated-risk#automation-signal-category), and the [signal name](https://docs.adyen.com/risk-management/configure-automated-risk#automation-signal-name). ### The signal bar The **signal bar** indicates the deductive reasoning behind the model's decision for a given payment. Any payment will have positive and negative characteristics. Negative signals move the decision towards a 'Looks fraudulent' classification, positive signals push the model's reasoning toward a 'Looks safe' classification. ![signal bar](https://docs.adyen.com/media/6985ebb2f51079862a4345632904a31867128dee346a3e5b48568ce4bb663376.png) By hovering over the signal bar, you can assess which signal category contributed the most to a certain decision. ### The signal category **Signal categories** are a common set of payment properties that are grouped together to explain why a certain decision was taken for a payment. Select a signal category to expand it into the signal names that contributed towards a fraudulent or legitimate classification. ### The signal name A signal category consists of multiple **signal names**. Signal names are the basic unit of a payment's information. All the signal names for a payment are processed by our machine learning engine to produce a 'fraud' or 'legitimate' outcome from that payment. We use the following design language to clarify a signal name's contribution towards the outcome of the signal category: ![explainability glossary legend](https://docs.adyen.com/media/fd2c893b82ed44cd47e687bf4127451bcbe24e98b3e758badac4961a896c1e2c.png) ## Optional: Run an experiment You can [set up and run an experiment](https://docs.adyen.com/risk-management/run-ab-test-experiments) to compare the performance of two risk profiles. To run an experiment, you create and configure two risk profiles. Then, you decide which profile is the **Base profile**, and which is the **Variant profile**, and set up an experiment. After some time, you can decide which profile performs better. You can run experiments with automated risk profiles. To be able to set up an experiment in which the base or the variant profile is automated, the merchant account running the experiment has to be [eligible for automated risk](https://docs.adyen.com/risk-management/configure-automated-risk#before-you-begin-automated-risk), and automated risk must be [turned on](https://docs.adyen.com/risk-management/configure-automated-risk#automation-on). For example, you can: * Compare a fully manual to an automated risk profile. * Compare two automated risk profiles with different [risk appetite](https://docs.adyen.com/risk-management/configure-automated-risk#automation-appetite) settings. ## Turn off automated risk In case you want to disable automated risk, and use a manual risk profile instead, you can turn automated risk off. 1. Log in to your [Customer Area](https://ca-test.adyen.com/). 2. Go to **Risk & disputes** > **Risk profiles**. 3. Select a risk profile. 4. Select the **Automated risk** tab. 5. Select **Disable automation**. When you disable automation, any previously configured [ShopperDNA](https://docs.adyen.com/risk-management/configure-manual-risk/standard-risk-rules#shopperdna-rules) and [velocity](https://docs.adyen.com/risk-management/configure-manual-risk/standard-risk-rules#velocity-rules) risk rule settings on this risk profile are restored. Any other manual risk rule configurations, including referrals and custom risk rules remain as is.