review-rubric.md
The eight-dimension review rubric
Read the asset against these eight dimensions in order. Structure first (does the argument hold), craft second (is it built well), polish last (is it ready to send). A failure high on the list usually explains failures lower down, so fix the top of the list first.
1. Strategic fit
- Who is this asset for: a buyer, a partner, an investor, an agency planner, an internal team, a social audience?
- What decision should it move them toward?
- Does the hierarchy lead with the right proof for that audience? A buyer wants the mechanic and the result; an investor wants the trend; a planner wants the fit. The same facts get ordered differently per audience.
If the asset does not know who it is for, nothing below matters yet. Fix this first.
2. Brand fidelity
- Correct brand, palette, logo, font, and visual world.
- No near-match colors (an almost-right hex reads as a knock-off).
- No mixed visual languages in one asset unless a combined presentation was explicitly briefed.
3. Story and hierarchy
- One clear message per slide, page, card, or graphic.
- Action titles that state the point, not generic labels ("Reach that holds through Q4", not "Results").
- A ten-second scan should reveal the argument without reading the body.
- The reader's eye should land first on the most important thing on the page.
4. Layout craft
- Elements aligned to a grid.
- Consistent margins, gutters, and repeated positions across slides.
- Roughly a third of the canvas or more left to breathe. Whitespace is a design tool, not wasted space.
- Nothing floating, crowded, or a few pixels off from its neighbours.
5. Typography craft
- A clear type scale, few weights, hierarchy from size, weight, spacing, and placement rather than from coloring everything.
- Equivalent elements use equivalent sizes and weights.
- Body copy is readable at the actual destination size (a slide seen across a room, a graphic seen on a phone).
- No overflow, cut-off text, orphaned single words, inconsistent capitalization, or punctuation drift.
6. Proof and data design
- Metrics are real, sourced, labelled, and characterized accurately.
- Metric cards show the number first, the precise label second.
- Charts have labels, units, and a source.
- No 3D, decorative, or misleading charts; no truncated axes that exaggerate.
- Number formatting is consistent across the asset.
In a client-facing asset, an unsourced or invented figure is a blocking failure, not a cosmetic one. One figure a sharp buyer can disprove discredits every honest number on the page.
7. Imagery and assets
- Screenshots are real, sharp, and placed in the correct orientation; device frames preserve aspect ratio.
- Logos have even visual weight and clear space, never stretched or recolored.
- Icons are from one consistent set, sharing one grid and stroke weight.
- Stock or generated imagery, if used, feels authentic and on-brand, not improvised filler.
8. Polish and client-readiness
- Would you send this to a top client or agency contact without flinching?
- Does it read as an intentional system, or as a filled-in template?
- Is the fix list specific enough for a designer to execute without guessing?
The severity call
Rank every issue you find:
- High - breaks the argument, the brand, or data integrity. The asset is not ready until this is fixed.
- Medium - weakens comprehension or credibility but does not break it. Fix before an important send.
- Low - polish. Fix if time allows; never let a pile of lows bury a high.