What is a Business Impact (BI) Score?

PentaTrail Team···8 min read
Contents

The same CVSS 9.0 vulnerability lands very differently on a payment platform than it does on a test box waiting to be decommissioned.

The Business Impact (BI) Score puts a number on that difference from the asset side, so it can be multiplied against technical severity.

01 / The BI Score is the sum of three axes, an integer from 5 to 13

Purpose, data classification and availability each carry a value, and their sum is the BI Score. Because the axes span different widths, the total bottoms out at 5 and tops out at 13.

Diagram of the three BI axes and their ranges. Purpose runs 1 to 5 and asks whether the host is reachable from outside, data classification is either 3 or 5 and asks whether sensitive data is handled, availability runs 1 to 3 and asks whether the host can be taken down. The total spans 5 to 13, and an untagged host defaults to 9

Figure 1: The width of each axis, and the range the total falls into

Deep dive: where the ceiling and the floor come from

The maximum of 13 needs all three at once: publicly exposed (purpose 5), sensitive data (classification 5) and round-the-clock operation (availability 3). The minimum of 5 comes from decommissioned (purpose 1), public-only data (classification 3) and safe to stop (availability 1). With no tags at all, each axis falls back to 3 and the host starts at 9.

02 / Data classification is the one axis that cannot say "no business impact"

Purpose and availability both go down to 1, while data classification stops at 3. Even a host that only serves public content is never treated as weightless.

Diagram comparing the floor of each axis. Purpose runs continuously from 1 to 5 and availability from 1 to 3, while data classification takes only the two points 3 and 5, never 1, 2 or 4

Figure 2: Only data classification has its floor held at 3

Deep dive: data classification has two steps, not five

Personal data, payment data or confidential scores 5; public-only or unclassified scores 3; nothing sits between them. That is a coarser step than purpose (five levels) or availability (three), but drawing a hard line on whether sensitive data is handled keeps the judgment from drifting.

03 / Three tags are the whole interface, and the score recomputes itself

Tag a host and the score is redrawn on the spot, all the way through to the TER band. What the team maintains is tags; nobody types in a number.

Diagram of how tagging drives recomputation. An untagged host starts at the default of 9; applying the purpose, data classification and availability tags redraws the score immediately, and it flows into the TER band decision

Figure 3: Tagging redraws the score, which feeds the band decision directly

Deep dive: which tags actually trigger a recompute

Only purpose, data_classification and availability trigger it; adding or removing any other tag leaves the score untouched. Removing one of the three resets just that axis to its default of 3 and recomputes.

04 / The score collapses into three tiers and becomes the business axis of TER

The 5-to-13 integer is not used directly; it folds into high, medium and low. The cuts sit at 10 and 5: 11 and above is high, 6 through 10 is medium, and only 5 is low.

Diagram of the BI score folding into three tiers. 11 to 13 is high BI covering public storefronts, payment platforms and customer databases; 6 to 10 is medium BI covering internal business systems and limited-access apps; 5 is low BI covering systems awaiting decommission and internal tools that can be stopped

Figure 4: The 5-to-13 integer folded into three tiers

Deep dive: the default of 9 lands in the middle tier

An untagged host starts at 9, so doing nothing leaves it treated as medium BI. Lifting a heavy asset to high, or dropping a decommission candidate to low, both take a deliberate tagging step.

05 / The same vulnerability changes rank depending on what it sits on

Identical technical rank, different BI, different band. One CVSS 9.0 vulnerability with a high exploitation probability comes out as S on a high-BI host and B on a low-BI one.

Diagram of one vulnerability splitting into different bands by host. A CVSS 9.0 vulnerability with a high exploitation probability reaches the top technical rank of TDL5 on both, but lands in band S on a host with BI score 13 and band B on a host with BI score 5

Figure 5: Same technical rank, different destination by asset weight

Scoring your own assets and re-ordering the queue is free to try for 14 days.

Appendix: values and tags for the three axes

Table 1: purpose — 1 to 5

Value Tag Examples
5 public / remote_access Public web, API, VPN
4 restricted / mail_infra Limited-access systems, mail
3 development / unknown Dev environment, unclassified assets
2 internal_only Internal-only systems
1 archived Decommissioned or pre-restart systems

Table 2: data_classification — 3 or 5

Value Tag Examples
5 personal_data / payment_data / confidential Customer data, payments, restricted data
3 public_only / unknown Public-only content, unclassified

Table 3: availability — 1 to 3

Value Tag Examples
3 mission_critical / unknown Round-the-clock systems
2 business_hours Operating during business hours only
1 non_critical Back office, temporary systems

Table 4: the boundary totals

Total Breakdown Meaning
13 5 + 5 + 3 Public system × sensitive data × round-the-clock
9 3 + 3 + 3 Default for an untagged host
5 1 + 3 + 1 Decommissioned × public-only × safe to stop

The values above were checked against PentaTrail's own scoring routine on 2026-08-17.

Visualize your attack surface with PentaTrail CTEM/ASM

From discovery to vulnerability validation and remediation — all powered by the CTEM framework.

Get Started

See pricing/Compare and choose