Brand Review
A structured design audit that rates a finished marketing asset and turns the
ratings into a prioritized fix list before it is published.
Core insight: design review fails when "this button is the wrong colour"
(a mechanical fix) is mixed with "I am not sure about the overall aesthetic" (a
judgment call). Conflated, they produce feedback nobody can act on. This review
keeps them apart: mechanical fixes are non-negotiable, judgment calls need
discussion.
When to use
- Before publishing a social graphic, ad creative or campaign visual.
- Before launching or updating a landing page.
- Reviewing an email template design.
- Evaluating a new brand direction or asset set.
When not to use
- Copy and messaging:
seven-copy-critics for the words, campaign-brief for
the campaign behind them.
- A strategic decision about brand direction:
ceo-review.
- A product interface:
design-expert.
- A rough draft not yet ready for critique: say it is a draft and review it when
it is ready.
Where the brand comes from
Brand consistency is judged against stated brand values, never against your
sense of what looks right.
This repository's agent configuration should have been provided to you (the
## Agent skills block pointing at docs/agents/); if it is missing, suggest
the user run the setup-aclarify-skills skill. In order:
- The
Brand kit folder named in docs/agents/workspace.md, when one is
set. With the setting at none, the default, skip this step. When the
get-brand-kit skill is installed, load it (through your skill tool,
/get-brand-kit, or by reading its SKILL.md) and it reads the folder;
otherwise read the folder directly, using the file names get-brand-kit
documents. Read the parts covering colour, typography, logos, layout and
voice: color-palette.md, typography.md, logos.md (logo paths are
relative to the folder), visual-design-system.md and voice-and-tone.md;
a file the folder lacks is a dimension not yet defined. Use the values
exactly.
- Otherwise, a brand guide the user names, or one in the conventions folder
named in
docs/agents/workspace.md (default: none).
- With neither, ask the user for the brand guide or its core values:
palette, typefaces, logo rules. Ask the user, using your structured-question
tool if you have one; otherwise list the questions numbered in plain text
and stop until they reply. If they would rather you proceed, score the other
six dimensions and mark Brand Consistency not checked: no brand source.
Never audit against get-brand-kit's placeholder kit: with no folder set, it
serves stand-in values, and a score against them says nothing about the brand.
If a file in the configured folder still says Status: placeholder, tell the
user that brand consistency was checked against a placeholder, not their brand.
Step 1: identify the asset
Look at the asset itself. Open an image file directly; render a page or
template if you can, at both desktop and mobile widths. If you cannot see it,
ask the user for a screenshot rather than reviewing from a description.
Then state:
- The asset type: social graphic, landing page, email, ad.
- Its destination platform.
- Its intended audience.
- The campaign or context it belongs to.
For a set (a three-post carousel, say), score each asset on its own, then add a
note on the set's coherence.
Step 2: score seven dimensions, 0 to 10
Give each a score and a two to four sentence rationale. A 7 is not a gift and a
4 is not a punishment; the score reflects the actual quality.
| Score |
Meaning |
| 9–10 |
Exceptional: a reference example. |
| 7–8 |
Strong: minor polish needed. |
| 5–6 |
Acceptable: clear room to improve, specific fixes needed. |
| 3–4 |
Weak: issues that will hurt performance. |
| 1–2 |
Broken: fundamental problems to resolve before any use. |
- Visual hierarchy. Is the eye drawn in the right order: headline, then the
supporting point, then the call to action?
- What does the eye land on first? Should it?
- Is the headline dominant, or competing with the imagery?
- Does the layout lead to the call to action, or must the viewer hunt?
- Brand consistency. Do font, colour, imagery and voice match the brand
source above?
- Are the brand typefaces used?
- Are the colours inside the palette, or are there off-brand additions?
- Does the imagery match the established style?
- Would someone who knows the brand recognize it at once?
- Message clarity. Can the core value proposition be pulled out in under
five seconds?
- Cover the logo: can you still tell what it is for?
- What does a first-time viewer take away in five seconds?
- Is the benefit stated, or left to inference?
- Emotional resonance. Does the design evoke the feeling the context calls
for: professional, trustworthy, bold, warm?
- What does the design communicate before any text is read?
- Is that right for this audience and message?
- Does the visual tone match the copy's tone?
- Call to action. Is it specific, visible and compelling?
- Is there one? With none, the score is 1 or 2.
- Is it specific ("Book a 20-minute call", not "Learn more")?
- Can a viewer find it without searching?
- Does it match the asset's goal: awareness, conversion or nurture?
- Platform fit. Do format, ratio and length suit where it will be seen?
- Is the aspect ratio right for the placement (square or landscape feed,
9:16 story)? Check the platform's current specification when unsure.
- Is text legible at the size it renders, on mobile as well as desktop?
- For email: does it hold up in clients that block images by default?
- Would it show uncropped in a native feed?
- Distinctiveness. Does it stand out, or could any competitor's logo
replace this one?
- What single element makes it unmistakably this brand's?
- Swap in a competitor's logo: does it still feel like this brand?
- Is there a choice here a competitor would not make? Should there be?
Step 3: separate mechanical fixes from judgment calls
Mechanical fixes are objectively wrong and must be fixed before publishing.
No discussion needed. For example:
- A colour outside the brand palette (
#FF0000 where the palette says
#CC2200).
- An off-brand typeface on the headline.
- Text illegible at mobile size, or failing a contrast check.
- The wrong aspect ratio for the platform.
- A missing call to action, or one hidden below the fold.
Judgment calls are choices that may be right depending on the strategy.
Discuss before changing. For example:
- The layout feels dense rather than airy: is that intended?
- The image choice is bold: does that match this campaign's tone?
- The headline is unconventional: a calculated risk, or an oversight?
List each category separately, with specific items.
Step 4: scorecard and fix list
## Brand review scorecard
| Dimension | Score | Notes |
|---|---|---|
| Visual hierarchy | X/10 | [one line] |
| Brand consistency | X/10 | [one line, or "not checked: no brand source"] |
| Message clarity | X/10 | [one line] |
| Emotional resonance | X/10 | [one line] |
| Call to action | X/10 | [one line] |
| Platform fit | X/10 | [one line] |
| Distinctiveness | X/10 | [one line] |
| **Overall** | X/10 | |
## Mechanical fixes (resolve before publishing)
1. [Fix]
## Design judgment calls (discuss before changing)
1. [Observation and question]
## Priority fix list (the three highest-impact changes)
1. [Fix, with expected impact]
2. [Fix, with expected impact]
3. [Fix, with expected impact]
The overall score is the mean of the dimensions scored; an unchecked dimension
is left out of it, and the scorecard says so.
- 7 or above: publishable once the mechanical fixes are resolved.
- 5 to 6: needs at least the three priority fixes before publishing.
- Below 5: back to production for significant rework.
1---2name: brand-review3description: >-4 Design audit of finished marketing creative against the brand: landing pages, social5 graphics, email templates, ad creatives. Scores each asset 0 to 10 on seven dimensions,6 separates mechanical fixes from design judgment calls, and returns a scorecard plus a7 prioritized fix list. Use before publishing any visual asset, or on is this on-brand,8 audit this creative, or review this design before it goes out. Do NOT use for copy9 quality (use `seven-copy-critics`), for the campaign's strategy (use `campaign-brief`),10 to look up brand values while making an asset (use `get-brand-kit`), or to design or11 fix a product interface (use `design-expert`).12license: MIT13argument-hint: '[the asset, or where it lives]'14metadata:15 source: https://skillmd.com/skills/spencermarx/brand-review16---1718# Brand Review1920A structured design audit that rates a finished marketing asset and turns the21ratings into a prioritized fix list before it is published.2223**Core insight:** design review fails when "this button is the wrong colour"24(a mechanical fix) is mixed with "I am not sure about the overall aesthetic" (a25judgment call). Conflated, they produce feedback nobody can act on. This review26keeps them apart: mechanical fixes are non-negotiable, judgment calls need27discussion.2829## When to use3031- Before publishing a social graphic, ad creative or campaign visual.32- Before launching or updating a landing page.33- Reviewing an email template design.34- Evaluating a new brand direction or asset set.3536## When not to use3738- Copy and messaging: `seven-copy-critics` for the words, `campaign-brief` for39 the campaign behind them.40- A strategic decision about brand direction: `ceo-review`.41- A product interface: `design-expert`.42- A rough draft not yet ready for critique: say it is a draft and review it when43 it is ready.4445## Where the brand comes from4647Brand consistency is judged against stated brand values, never against your48sense of what looks right.4950This repository's agent configuration should have been provided to you (the51`## Agent skills` block pointing at `docs/agents/`); if it is missing, suggest52the user run the `setup-aclarify-skills` skill. In order:53541. **The `Brand kit` folder named in `docs/agents/workspace.md`, when one is55 set.** With the setting at `none`, the default, skip this step. When the56 `get-brand-kit` skill is installed, load it (through your skill tool,57 `/get-brand-kit`, or by reading its SKILL.md) and it reads the folder;58 otherwise read the folder directly, using the file names `get-brand-kit`59 documents. Read the parts covering colour, typography, logos, layout and60 voice: `color-palette.md`, `typography.md`, `logos.md` (logo paths are61 relative to the folder), `visual-design-system.md` and `voice-and-tone.md`;62 a file the folder lacks is a dimension not yet defined. Use the values63 exactly.642. **Otherwise, a brand guide the user names**, or one in the conventions folder65 named in `docs/agents/workspace.md` (default: none).663. **With neither,** ask the user for the brand guide or its core values:67 palette, typefaces, logo rules. Ask the user, using your structured-question68 tool if you have one; otherwise list the questions numbered in plain text69 and stop until they reply. If they would rather you proceed, score the other70 six dimensions and mark Brand Consistency **not checked: no brand source**.7172Never audit against `get-brand-kit`'s placeholder kit: with no folder set, it73serves stand-in values, and a score against them says nothing about the brand.7475If a file in the configured folder still says `Status: placeholder`, tell the76user that brand consistency was checked against a placeholder, not their brand.7778## Step 1: identify the asset7980Look at the asset itself. Open an image file directly; render a page or81template if you can, at both desktop and mobile widths. If you cannot see it,82ask the user for a screenshot rather than reviewing from a description.8384Then state:8586- The asset type: social graphic, landing page, email, ad.87- Its destination platform.88- Its intended audience.89- The campaign or context it belongs to.9091For a set (a three-post carousel, say), score each asset on its own, then add a92note on the set's coherence.9394## Step 2: score seven dimensions, 0 to 109596Give each a score and a two to four sentence rationale. A 7 is not a gift and a974 is not a punishment; the score reflects the actual quality.9899| Score | Meaning |100|---|---|101| 9–10 | Exceptional: a reference example. |102| 7–8 | Strong: minor polish needed. |103| 5–6 | Acceptable: clear room to improve, specific fixes needed. |104| 3–4 | Weak: issues that will hurt performance. |105| 1–2 | Broken: fundamental problems to resolve before any use. |1061071. **Visual hierarchy.** Is the eye drawn in the right order: headline, then the108 supporting point, then the call to action?109 - What does the eye land on first? Should it?110 - Is the headline dominant, or competing with the imagery?111 - Does the layout lead to the call to action, or must the viewer hunt?1122. **Brand consistency.** Do font, colour, imagery and voice match the brand113 source above?114 - Are the brand typefaces used?115 - Are the colours inside the palette, or are there off-brand additions?116 - Does the imagery match the established style?117 - Would someone who knows the brand recognize it at once?1183. **Message clarity.** Can the core value proposition be pulled out in under119 five seconds?120 - Cover the logo: can you still tell what it is for?121 - What does a first-time viewer take away in five seconds?122 - Is the benefit stated, or left to inference?1234. **Emotional resonance.** Does the design evoke the feeling the context calls124 for: professional, trustworthy, bold, warm?125 - What does the design communicate before any text is read?126 - Is that right for this audience and message?127 - Does the visual tone match the copy's tone?1285. **Call to action.** Is it specific, visible and compelling?129 - Is there one? With none, the score is 1 or 2.130 - Is it specific ("Book a 20-minute call", not "Learn more")?131 - Can a viewer find it without searching?132 - Does it match the asset's goal: awareness, conversion or nurture?1336. **Platform fit.** Do format, ratio and length suit where it will be seen?134 - Is the aspect ratio right for the placement (square or landscape feed,135 9:16 story)? Check the platform's current specification when unsure.136 - Is text legible at the size it renders, on mobile as well as desktop?137 - For email: does it hold up in clients that block images by default?138 - Would it show uncropped in a native feed?1397. **Distinctiveness.** Does it stand out, or could any competitor's logo140 replace this one?141 - What single element makes it unmistakably this brand's?142 - Swap in a competitor's logo: does it still feel like this brand?143 - Is there a choice here a competitor would not make? Should there be?144145## Step 3: separate mechanical fixes from judgment calls146147**Mechanical fixes** are objectively wrong and must be fixed before publishing.148No discussion needed. For example:149150- A colour outside the brand palette (`#FF0000` where the palette says151 `#CC2200`).152- An off-brand typeface on the headline.153- Text illegible at mobile size, or failing a contrast check.154- The wrong aspect ratio for the platform.155- A missing call to action, or one hidden below the fold.156157**Judgment calls** are choices that may be right depending on the strategy.158Discuss before changing. For example:159160- The layout feels dense rather than airy: is that intended?161- The image choice is bold: does that match this campaign's tone?162- The headline is unconventional: a calculated risk, or an oversight?163164List each category separately, with specific items.165166## Step 4: scorecard and fix list167168```markdown169## Brand review scorecard170171| Dimension | Score | Notes |172|---|---|---|173| Visual hierarchy | X/10 | [one line] |174| Brand consistency | X/10 | [one line, or "not checked: no brand source"] |175| Message clarity | X/10 | [one line] |176| Emotional resonance | X/10 | [one line] |177| Call to action | X/10 | [one line] |178| Platform fit | X/10 | [one line] |179| Distinctiveness | X/10 | [one line] |180| **Overall** | X/10 | |181182## Mechanical fixes (resolve before publishing)1831. [Fix]184185## Design judgment calls (discuss before changing)1861. [Observation and question]187188## Priority fix list (the three highest-impact changes)1891. [Fix, with expected impact]1902. [Fix, with expected impact]1913. [Fix, with expected impact]192```193194The overall score is the mean of the dimensions scored; an unchecked dimension195is left out of it, and the scorecard says so.196197- **7 or above:** publishable once the mechanical fixes are resolved.198- **5 to 6:** needs at least the three priority fixes before publishing.199- **Below 5:** back to production for significant rework.