Scoring Guide

How audit scores are calculated across different content types, including weights, penalties, strictness rules, caps, and verdict thresholds.

How The Final Score Is Built
The system starts with quality scores, subtracts risk, and then applies guardrails for serious issues.
1

Every audit starts with five dimension scores.

2

The platform combines those scores using content-type-specific weights.

3

It subtracts issue penalties from risky or off-brand findings.

4

It applies stricter or more forgiving rules depending on content type.

5

Serious issues can cap the maximum score before the final verdict is assigned.

Verdict Scale
On Brand85-100

On Brand: Strong fit with brand guidance and low risk.

Needs Review65-84

Needs Review: Mostly usable, but a person should review the flagged areas.

Off Brand0-64

Off Brand: High risk or poor brand fit. Rewrite before using.

Content Type Policy
Each content type uses the same scoring ingredients, but changes how much each part matters.

What This Content Type Cares About

Apply balanced brand standards across tone, approved messaging, restricted language, audience fit, and clarity. Treat the content as general-purpose copy unless the submitted text clearly implies a narrower format.

Dimension Weights

Higher percentages have more influence on the weighted dimension score.

Tone alignmentHighest weight

How closely the content matches the brand's voice and tone.

25%
Messaging alignmentHighest weight

How well the content supports the brand's positioning and approved messages.

25%
Banned phrase safetyHighest weight

Whether the content avoids restricted language, risky claims, and off-policy phrasing.

25%
Audience fit

How well the content fits the intended reader's context, expectations, and level of detail.

15%
Clarity and trust

How clearly and credibly the content communicates without overstatement.

10%

Issue Strictness

Multipliers adjust the base penalty for this content type.

Mild style
Standard1x
Hype phrase
Standard1x
Banned phrase
Standard1x
Absolute claim
Standard1x
Direct contradiction
Standard1x

Score Caps

Caps prevent risky content from receiving a high score even when other parts look good.

One hype issueMax 84
One banned phraseMax 84
One severe issueMax 64
Multiple severe issuesMax 44
Issue Type Reference
These are the finding types that can subtract points from the weighted dimension score.

Mild style

What it means
The wording is close, but the voice or style does not fully match the brand.
Why it matters
Small style mismatches can make otherwise accurate content feel inconsistent.
Usually triggered by
Overly formal language, casual phrasing, awkward rhythm, or wording that does not fit the brand voice.

Hype phrase

What it means
The content uses promotional or exaggerated language that may feel overstated.
Why it matters
Hype can reduce trust, especially in formats where readers expect clear proof.
Usually triggered by
Phrases like breakthrough, revolutionary, best-in-class, game-changing, or similar unsupported claims.

Banned phrase

What it means
The content uses language the brand has restricted or marked as risky.
Why it matters
Restricted wording can create compliance, positioning, or trust problems.
Usually triggered by
Known banned phrases, risky claims, off-policy terminology, or language the brand explicitly avoids.

Absolute claim

What it means
The content makes a claim that sounds certain, universal, or guaranteed.
Why it matters
Absolute claims are high risk because they often need strong proof and may overpromise.
Usually triggered by
Words like always, never, guaranteed, eliminates, proven for everyone, or claims with no room for context.

Direct contradiction

What it means
The content conflicts with approved brand messaging or stated brand rules.
Why it matters
Contradictions can mislead readers and weaken the brand's public position.
Usually triggered by
Statements that oppose approved positioning, claim unsupported capabilities, or say something the brand guidance rejects.
Base Penalty Table
Actual penalty = base penalty x content type multiplier.
Issue typeLowMediumHigh
Mild style357
Hype phrase81215
Banned phrase152025
Absolute claim253240
Direct contradiction253545
Worked Example
A simple example showing how one serious finding can change the final score.

Say the content is Ad copy. Its dimension scores combine to 82, but the audit finds a high absolute claim. A high absolute claim has a base penalty of 40. For ad copy, that issue is much stricter, so the penalty is multiplied by 1.45x.

The actual penalty is 40 x 1.45 = 58. The score after penalty is 82 - 58 = 24.

Final score

24/100
VerdictOff Brand