Sensitivity scoring for S3 buckets - Amazon Macie

Sensitivity scoring for S3 buckets

If automated sensitive data discovery is enabled, Amazon Macie automatically calculates and assigns a sensitivity score to each Amazon Simple Storage Service (Amazon S3) general purpose bucket that it monitors and analyzes for an account or organization. A sensitivity score is a quantitative representation of the amount of sensitive data that an S3 bucket might contain. Based on that score, Macie also assigns a sensitivity label to each bucket. A sensitivity label is a qualitative representation of a bucket's sensitivity score. These values can serve as reference points for determining where sensitive data might reside in your Amazon S3 data estate, and identifying and monitoring potential security risks for that data.

By default, an S3 bucket's sensitivity score and label reflect the results of automated sensitive data discovery activities that Macie has performed thus far for the bucket. They don't reflect the results of sensitive data discovery jobs that you've created and run. In addition, neither the score nor the label implies or otherwise indicates the criticality or importance that a bucket or a bucket's objects might have for your organization. However, you can override a bucket's calculated score by manually assigning the maximum score (100) to the bucket, which also assigns the Sensitive label to the bucket.

Sensitivity scoring dimensions and ranges

If calculated by Amazon Macie, an S3 bucket's sensitivity score is a quantitative measure of the intersection of two primary dimensions:

  • The amount of sensitive data that Macie has found in the bucket. This derives primarily from the nature and number of sensitive data types that Macie has found in the bucket and the number of occurrences of each type.

  • The amount of data that Macie has analyzed in the bucket. This derives primarily from the number of unique objects that Macie has analyzed in the bucket relative to the total number of unique objects in the bucket.

An S3 bucket's sensitivity score also determines which sensitivity label Macie assigns to the bucket. The sensitivity label is a qualitative representation of the score—for example, Sensitive or Not sensitive. On the Amazon Macie console, a bucket's sensitivity score also determines which color Macie uses to represent the bucket in data visualizations, as shown in the following image.

A color spectrum that shows hues for sensitivity scores. Red hues for 51-100. Blue hues for 1-49. Gray for -1.

Sensitivity scores range from -1 through 100, as described in the following table. To assess inputs to an S3 bucket's score, you can refer to sensitive data discovery statistics and other details that Macie provides about the bucket.

Sensitivity score Sensitivity label Additional information
-1 Classification error

Macie hasn't successfully analyzed any of the bucket's objects yet due to object-level classification errors—issues with object-level permissions settings, object content, or quotas.

When Macie tried to analyze one or more objects in the bucket, errors occurred. For example, an object is a malformed file, or an object is encrypted with a key that Macie can't access or isn't allowed to use. Coverage data for the bucket can help you investigate and remediate the errors. For more information, see Assessing automated sensitive data discovery coverage.

Macie will continue to try to analyze objects in the bucket. If Macie analyzes an object successfully, Macie will update the bucket's sensitivity score and label to reflect the results of the analysis.

1-49 Not sensitive

In this range, a higher score, such as 49, indicates that Macie has analyzed relatively few objects in the bucket. A lower score, such as 1, indicates that Macie has analyzed many objects in the bucket (relative to the total number of objects in the bucket) and detected relatively few types and occurrences of sensitive data in those objects.

A score of 1 can also indicate that the bucket doesn't store any objects or all the objects in the bucket contain zero (0) bytes of data. Object statistics in the bucket's details can help you determine if this is the case. For more information, see Reviewing S3 bucket details.

50 Not yet analyzed

Macie hasn't tried to analyze or analyzed any of the bucket's objects yet.

Macie automatically assigns this score when automated discovery is initially enabled or a bucket is added to the bucket inventory for an account. In an organization, a bucket can also have this score if automated discovery has never been enabled for the account that owns the bucket.

A score of 50 can also indicate that the bucket's permissions settings prevent Macie from accessing the bucket or the bucket’s objects. This is typically due to a restrictive bucket policy. The bucket's details can help you determine if this is the case because Macie can provide only a subset of information about the bucket. For information about how to address this issue, see Allowing Macie to access S3 buckets and objects.

51-99 Sensitive

In this range, a higher score, such as 99, indicates that Macie has analyzed many objects in the bucket (relative to the total number of objects in the bucket) and detected many types and occurrences of sensitive data in those objects. A lower score, such as 51, indicates that Macie has analyzed a moderate number of objects in the bucket (relative to the total number of objects in the bucket) and detected at least a few types and occurrences of sensitive data in those objects.

100 Sensitive

The score was manually assigned to the bucket, overriding the calculated score. Macie doesn't assign this score to buckets.

Monitoring sensitivity scores

When automated sensitive data discovery is initially enabled for an account, Amazon Macie automatically assigns a sensitivity score of 50 to each S3 bucket that the account owns. Macie also assigns this score to a bucket when the bucket is added to the bucket inventory for an account. Based on that score, each bucket's sensitivity label is Not yet analyzed. The exception is an empty bucket, which is a bucket that doesn't store any objects or all the objects in the bucket contain zero (0) bytes of data. If this is the case for a bucket, Macie assigns a score of 1 to the bucket and the bucket's sensitivity label is Not sensitive.

As automated sensitive data discovery progresses each day, Macie updates sensitivity scores and labels for S3 buckets to reflect the results of its analysis. For example:

  • If Macie doesn't find sensitive data in an object, Macie decreases the bucket's sensitivity score and updates the bucket's sensitivity label as necessary.

  • If Macie finds sensitive data in an object, Macie increases the bucket's sensitivity score and updates the bucket's sensitivity label as necessary.

  • If Macie finds sensitive data in an object that's subsequently changed, Macie removes sensitive data detections for the object from the bucket's sensitivity score and updates the bucket's sensitivity label as necessary.

  • If Macie finds sensitive data in an object that's subsequently deleted, Macie removes sensitive data detections for the object from the bucket's sensitivity score and updates the bucket's sensitivity label as necessary.

  • If an object is added to a bucket that was previously empty and Macie finds sensitive data in the object, Macie increases the bucket's sensitivity score and updates the bucket's sensitivity label as necessary.

  • If a bucket's permissions settings prevent Macie from retrieving information about or accessing the bucket or the bucket’s objects, Macie changes the bucket's sensitivity score to 50 and changes the bucket's sensitivity label to Not yet analyzed.

Analysis results can begin to appear within 48 hours of enabling automated sensitive data discovery for an account.

If you're the Macie administrator for an organization or you have a standalone Macie account, you can adjust the sensitivity scoring settings for your organization or account:

  • To adjust the settings for subsequent analyses of all S3 buckets, change the automated sensitive data discovery settings for your account. You can start including or excluding specific managed data identifiers, custom data identifiers, or allow lists. You can also exclude specific buckets. For more information, see Configuring automated discovery.

  • To adjust the settings for individual S3 buckets, change the automated sensitive data discovery settings for each bucket. You can include or exclude specific types of sensitive data from a bucket's score. You can also specify whether to assign an automatically calculated score to a bucket. For more information, see Managing automated discovery for individual S3 buckets.

If you disable automated sensitive data discovery, the effect on existing sensitivity scores and labels varies. If you disable it for a member account in an organization, existing scores and labels persist for S3 buckets that the account owns. If you disable it for an organization overall or a standalone Macie account, existing scores and labels persist for only 30 days. After 30 days, Macie resets scores and labels for all the buckets that the organization or account owns. If a bucket stores objects, Macie changes the score to 50 and assigns the Not yet analyzed label to the bucket. If a bucket is empty, Macie changes the score to 1 and assigns the Not sensitive label to the bucket. After this reset, Macie stops updating sensitivity scores and labels for the buckets, unless you enable automated sensitive data discovery for the organization or account again.