[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-09-01。"],[],[],null,["# Evaluate your data risk management needs\n\nThis series of documents provides strategies for evaluating and mitigating data\nrisk in your organization. It also describes and compares two\nSensitive Data Protection services that help you learn more about your\ncurrent data security posture.\n\nObjectives of data risk management\n----------------------------------\n\nManaging data risk involves storing, processing, and using your data within the\nappropriate risk levels for your business. When you perform data risk\nmanagement, we recommend that you aim for the following objectives:\n\n- Your data is properly discovered and classified.\n- Risk of data exposure is properly understood.\n- Data is protected by appropriate controls or de-risked through obfuscation.\n\nAs you evaluate your data workloads you can start by asking these\nquestions:\n\n- What kind of data does this workload handle and is any of it sensitive?\n- Is this data properly exposed? For example, is access to the data restricted to the right users, in the right environment, and for an approved purpose?\n- Can the risk of this data be reduced through data minimization and obfuscation strategies?\n\nTaking a well-informed and risk-based approach can help you make the most of\nyour data without compromising the privacy of your users.\n\nExample analysis\n----------------\n\nFor this example, suppose your data team is trying to build a machine learning\nmodel based on customer feedback in product reviews.\n\n### What kind of data does this workload handle and is any of it sensitive?\n\nIn the data workload, you found that the primary key used is the customer email\naddress. Customer email addresses often contain the customers' names.\nAdditionally, the actual product reviews contain unstructured data (or *freeform\ndata*) submitted by the customer. Unstructured data can contain intermittent\ninstances of sensitive data like phone numbers and addresses.\n\n### Is this data properly exposed?\n\nYou found that the data is accessible only to the product team. However, you\nwant to share the data to your data analytics team, so that they can use it to\nbuild a machine learning model. Exposing the data to more people also means\nexposing it to more development environments where this data will be stored and\nprocessed. You determined that the exposure risk will increase.\n\n### Can the risk of this data be reduced through data minimization and obfuscation strategies?\n\nYou know that the analytics team doesn't need any of the actual sensitive\npersonally identifiable information (PII) in the dataset. However, they need\nto aggregate the data per customer. They need a way to determine which reviews\nbelong to the same customer. To address this need, you decide to tokenize all\nthe structured PII---the customer email addresses---to keep the\nreferential integrity of your data. You also decide to inspect the unstructured\ndata---the reviews---and mask any intermittent sensitive data within it.\n\nWhat's next\n-----------\n\n- [Compare Sensitive Data Protection services that help you learn about your\n data](/sensitive-data-protection/docs/learn-about-your-data) (next document in this series)"]]