# Creativity ranking methodology

# Show the work.Explain the score.

Method 1.1.0 · provisional editorial review

## What is being ranked?

Six scored brands from a purposive seven-brand sample of the 148-brand directory. This is a provisional ranking of sampled creative work, not a census of DTC brands or a current-year advertising award.

## How the sample is chosen

Three specific examples per brand, across at least two formats and including a product or experience. Historical and current work may both be included; source publication dates and retrieval dates are separate. Brands were selected for varied creative mechanisms and accessible first-party evidence, not randomly. English-language and source-availability bias remain.

## Minimum evidence

Three linked examples across at least two formats, including a product or customer experience.

Every scored dimension has a rationale tied to example IDs. Unknown dimensions exclude a brand from the overall ranking.

Two explicit comparison lenses; product-level or cross-category comparisons are labeled and do not imply a full peer campaign audit.

Founding channel recorded separately from creativity. The digital-launch view includes only brands with a linked launch source; ownership and current direct-sales share may remain unknown.

## The five dimensions

## Score anchors

The reviewed evidence shows a conventional execution with no distinctive choice under this dimension; cite the comparison.

One limited distinctive choice, or participation restricted to following a supplied routine.

A coherent idea is supported, but uses an established mechanism or has limited breadth.

A specific and developed approach is supported across the relevant examples, with an explicit limitation.

An unusually strong, integrated example in this sample; rationale must identify why it exceeds 3. This is an editorial judgment, not proof of market-wide uniqueness.

## The calculation

Round(sum(weight × dimension_score / 4)) to the nearest whole point, once at the end. All five dimensions must be present. Scores are editorial 0–4 anchors, not measured performance.

Example: Liquid Death’s anchors are 2, 4, 4, 4, 3. The weighted sum is 12.5 + 20 + 25 + 15 + 11.25 = 83.75, displayed as 84/100. No dimension is filled in from brand popularity.

Equal rounded totals share the same competition rank (1, 2, 2, 4); alphabetical order breaks display ties only. Filtered rankings are recalculated within the selected cohort.

A one-point change in a dimension changes the unrounded total by 3.75–6.25 points. Small score gaps should be read as broadly comparable work, not statistically meaningful differences.

## Who made the judgment?

Single AI-assisted editorial pass, explicitly provisional. No second reviewer or independent consensus is claimed. Re-review is due 2027-01-13. A second human review, consistent campaign window and fuller category-peer samples are required before calling this a definitive DTC-native league table.

Reviewer: Codex — AI-assisted editorial review; not independently second-reviewed. Sources checked 2026-10-13. Source publication dates can be unknown; a check date never substitutes for a launch date.

## What does not earn points?

Revenue, Ad spend, Follower count, Unverified AI usage, Platform rank, Visual polish alone. Product efficacy, environmental benefits and commercial results are not validated by a creativity score.

Version 1.0 proposed a future two-reviewer, DTC-origin-qualified ranking but assigned no scores. Version 1.1 adds a visibly provisional pilot with a broader research cohort; it does not claim the stricter v1 review gate has been satisfied.

Method as JSON · Back to the ranking →