PrioritisationLivePaid plans

Predictive Risk

Predictive Risk combines CISA KEV status, severity, EPSS exploitation probability and reachability into one priority score per finding. Instead of working down a list sorted by CVSS, you start with the vulnerabilities most likely to be exploited.

Detection Coverage

Risk signals combined

EPSS Score

The Exploit Prediction Scoring System from FIRST.org: the probability that a CVE is exploited within 30 days. Refreshed daily.

CISA KEV Status

Whether the CVE is on CISA's Known Exploited Vulnerabilities catalogue. A KEV listing puts the finding at the top.

Severity

The finding's severity, from critical to informational.

Reachability

For dependency findings in repository scans, whether the vulnerable package is actually imported in your code.

Priority Buckets

Each finding is placed in Fix now, Fix soon or Watch.

Target Risk Score

A score from 0 to 100 for each target, recalculated every night from its open findings.

How it works

A score you can explain

  1. 1

    Findings arrive

    Findings from your scan engines are recorded with their severity and, where there is one, a CVE ID.

  2. 2

    Daily enrichment

    EPSS scores from FIRST.org and the CISA KEV list are refreshed every day for open findings that carry a CVE ID.

  3. 3

    Score calculation

    The priority score adds up fixed contributions from KEV status, severity, EPSS and reachability. It is a deterministic formula, not a black box.

  4. 4

    Fix these first

    The Predictive Risk dashboard lists the findings to fix first, with the EPSS distribution across your open findings.

priority-score.jsonJSON
{
  "cve": "CVE-2021-44228",
  "severity": "critical",
  "kev_listed": true,
  "epss": 0.94,
  "bucket": "fix-now",
  "breakdown": {
    "kev": 100,
    "severity": 10,
    "epss": 18.8
  }
}
FindingIllustrative score for Log4Shell: it is on the KEV list, so it lands in Fix now. The EPSS value shown is an example.

Spoorder Predictive Risk vs sorting by CVSS

CriterionSpoordersorting by CVSS
Risk modelKEV, severity, EPSS and reachabilityCVSS only
EPSS dataRefreshed daily from FIRST.orgNot used
ExplainabilityA fixed formula with named inputsAn opaque number
UpdatesRe-scored as EPSS and KEV changeFixed at scan time
What is EPSS?

The Exploit Prediction Scoring System (EPSS) is published by FIRST.org. It estimates the probability that a CVE will be exploited in the wild within the next 30 days. CVSS measures how severe a vulnerability could be; EPSS measures how likely it is to be exploited.

How often is the score updated?

EPSS and KEV data are refreshed daily for open findings with a CVE ID, and the priority score is recalculated when they change. The per-target risk score is recalculated every night.

Does Predictive Risk replace CVSS?

No. Severity is one of the inputs. Predictive Risk adds real-world context: whether the vulnerability is being exploited and how likely exploitation is.

Which findings get an EPSS score?

Findings that carry a CVE ID, which mostly come from dependency, container and CVE Monitor scans. Findings without a CVE are scored on severity.

Start scanning

Start with a free health check. CVE findings come from the engines in the paid plans, which have a 14-day trial on Builder, Pro and Scale.