THE CULTURAL-SIGNAL INDEX

METHODOLOGY

How the score is built.

The Cultural-Signal Score is an authorship-weighted composite of four editorial dimensions — each scored 1–5 and expressed out of 100. It is a defensible, evidence-led reading. This is exactly how it works, and exactly what it is not.

THE FORMULA

Four axes, one authorship-weighted score.

Each work is read across four axes and scored 1–5 on each. The axes are normalised and combined into a weighted composite expressed out of 100. The weighting is deliberate: authorship carries more than any other dimension, because authorship is the question this publication is built to ask.

IDEA
25%
AUTHORSHIP
35%
EXECUTION
15%
CONSEQUENCE
25%

Composite = Σ ( axis level ÷ 5 × axis weight ) × 100, rounded. A brand’s Index score is the mean composite across its scored works.

THE BANDS

A rating, not a mark out of a hundred.

The composite maps to a letter band. The cut-offs are fixed and published here — most rating providers publish a score but keep the boundaries behind it private. Ours are open, so anyone can check that a number and its band agree.

AAA90–100Exceptional — authored, executed and consequential
AA80–89Strong — clear authorship, real consequence
A70–79Sound — good work, incomplete capture
BBB60–69Mixed — the idea outruns the ownership
BB50–59Weak — visibility without value capture
B40–49Poor — borrowed authorship, thin consequence
C< 40Failing — value created, value surrendered

WORKED EXAMPLES

Here is the arithmetic, on two real works.

One at the top of the scale and one near the bottom. Both are live case studies; the numbers below are read from the same data that powers the Index, so they cannot drift from what we actually published.

IDEA4 ÷ 5 × 0.2520.0
AUTHORSHIP5 ÷ 5 × 0.3535.0
EXECUTION5 ÷ 5 × 0.1515.0
CONSEQUENCE5 ÷ 5 × 0.2525.0
COMPOSITE95
IDEA3 ÷ 5 × 0.2515.0
AUTHORSHIP2 ÷ 5 × 0.3514.0
EXECUTION3 ÷ 5 × 0.159.0
CONSEQUENCE1 ÷ 5 × 0.255.0
COMPOSITE43

THE OUTLOOK

Which way the score is likely to move.

A score says what a work is worth now. An outlook says which way it is likely to move next. We publish the probability each one stands for — a watchlist without a stated likelihood and a clock is just a mood.

POSITIVE

At least a one-in-three likelihood the score rises within twelve months — typically because evidence of consequence is pending rather than absent.

NEGATIVE

At least a one-in-three likelihood the score falls within twelve months — typically because the rating rests on reported figures that may not hold, or on a position that is eroding.

DEVELOPING

A pending event could move the score either way. Used when the trigger is known and binary, and its direction is not.

STABLE

No one-in-three likelihood of movement identified. Stable is the default and the majority: it is a finding, not an absence of one.

An outlook is assigned at at least a one-in-three likelihood over the next twelve months. A work moves to Watch only at at least a one-in-two likelihood within 90 days. Every outlook carries the date it was assigned and the specific trigger that would move the score — so a later edition can mark our own calls. 11 of 71 works currently carry a non-Stable outlook, and none are on watch.

Stable is the default and the majority. That is a finding, not an absence of one: most work is not about to move, and saying so is the only way the exceptions mean anything.

The gap between them is the whole point. Both are competent, well-resourced products built for the same continent. What separates a 95 from a 43 is not craft — it is who ended up owning the thing that was built, and whether anything verifiable followed.

THE FOUR AXES

IDEA25% weight

The originality and cultural force of the central idea — how far it departs from convention and how specifically it speaks from, not merely about, the culture.

Raises it: A genuinely new form; a culturally precise insight; an idea only this author could have had.

Lowers it: A borrowed template; a trend chased late; cultural reference as decoration.

AUTHORSHIP35% weight

Who actually shaped the work — African and diaspora creative leadership versus localisation of a decision made elsewhere. The question MonoKromatik exists to ask.

Raises it: African creatives holding the pen on concept, design and direction; ownership of the IP and the upside.

Lowers it: A global team localising; African talent as the face but not the author; value captured offshore.

EXECUTION15% weight

Craft and coherence — narrative, design, product detail and the discipline of finishing. Whether the idea is realised at the level it deserves.

Raises it: Precise craft; a coherent system across touchpoints; detail that rewards attention.

Lowers it: A strong idea undercut by rough delivery; inconsistency; unfinished edges.

CONSEQUENCE25% weight

What the work actually moved — in culture and in commerce. Evidence of effect, not just ambition or reach.

Raises it: A measurable shift; a category reframed; durable commercial or cultural outcome.

Lowers it: Visibility without effect; a moment that left nothing behind; claims without evidence.

THE ROLLUP

We rate the work.

Scoring happens at the level of a single work — a specific case study, with evidence attached. That is the unit we stand behind, and it is what the Index ranks.

A brand’s position is the mean of its scored works. Today that mean is drawn from a single work for almost every brand in the Index, so treat a brand-level score as a reading of that work, not a verdict on the organisation’s record. A brand only earns a track record once it carries several scored works, and we say how many each one has beside its score.

As new case studies are decoded, positions move — and the snapshot records that movement so the ranking has a memory. Movement is only ever computed between snapshots scored under the same rubric: a re-scoring changes the ruler, not the work, and we will not publish it as though brands had moved.

THE MEASURED ANCHOR

A number you can count, beside the one we judge.

The Signal Score is editorial judgement. So every brand also carries an Evidence Strength— a second score, 0–100, that is measured, not judged: it is counted directly from the corroboration behind the work. Independently-confirmed facts weigh most, then verification status, then cited sources, then single-source reported claims — passed through a saturation curve so it rises with corroboration without running away on volume.

It answers a different question than the rating: not how good, but how well-backed. A high Signal Score with a Developing evidence base is an honest signal — a strong read that is still thinly corroborated. Tiers: Strong (75+), Moderate (45+), Developing (below 45).

This is the rating-agency pattern — a rating paired with a data-quality indicator. It gives a buyer a measured anchor for a judged number, and it gives every brand a clear way to climb: deepen the evidence.

WHAT IT IS NOT

Editorial judgement, stated openly.

The score is interpretive editorial judgement — a defensible, evidence-led reading. It is not a measured market metric, not sales data, and not a popularity count. We say so plainly because the honesty is the point: every score traces to a case study and the evidence behind it.

Axis levels are set by a named human analyst (see below), not an anonymous committee. An AI-assisted workflow — a model drafts the decode, the analyst approves it — scales the scoring without surrendering the judgement. It is never autonomous: a person signs every score.

The weighting is locked and versioned in code. If it ever changes, that change is documented — a ranking you can’t audit isn’t intelligence.

THE ANALYST

A name behind the number.

Trust in a rating is personal. Every score on the Cultural-Signal Index is set and signed off by Sibu Shangase, founder and lead analyst of MonoKromatik — who built the methodology, weights each axis against the evidence, and is accountable for the read. Where a score is wrong, the correction carries the same byline.

This is deliberate: an index is only as credible as the judgement behind it, and judgement has a name. The analyst works to the locked methodology above and the measured Evidence Strength beside each score — disclosure, not a black box.

SS

Sibu Shangase

Founder & Lead Analyst, MonoKromatik · Cultural-Signal Index

CORRECTIONS & INTEGRITY

Sourced, challengeable, correctable.

The sourcing standard.Every case study and report is web-sourced and cited. A draft that can’t meet the bar — at least two independent, real sources, every factual beat traceable, fact kept separate from interpretation — is rejected before it publishes. We do not invent quotes, statistics, dates or sources; an unverifiable claim is dropped, not padded. Verification is stated on every decode (verified · partial · interpretive) and reflected in the Evidence Strength beside the score.

Right of reply. If you are a rated brand, you can respond. Send us evidence — a correction, context we missed, or proof of an outcome — and it is reviewed against the same methodology. New evidence can move a score; that is precisely how the Index is meant to work.

How we correct. Flag an error or dispute a score at editor@monokromatik.com. Material factual errors are corrected promptly; the correction carries the same analyst byline, is dated, and — where it changes a score — is recorded in the Index’s movement history so the change is auditable. We would rather be corrected than wrong.

VERSION HISTORY

We re-grade in public.

Every snapshot of the Index is dated and stamped with two things: the rubric it was scored under, and the unit it ranks. When either changes, that is a comparability break — earlier readings are not the same kind of measurement, so we mark the break and never silently compare across it. An index that hides its own revisions is an opinion wearing a date.

Two breaks are on the record. On 20 July 2026 the whole corpus was rescored under rubric v2.0 and scores fell, particularly on consequence — the correction of an early-days generosity. On 21 July the ranking unit changed from a brand roll-up to a single scored work, because that is the unit the evidence actually attaches to.

DATEBASISENTRIES
23 Jun 2026rubric v1 · ranks brands24
29 Jun 2026rubric v1 · ranks brands25
6 Jul 2026rubric v1 · ranks brands26
7 Jul 2026rubric v1 · ranks brands42
8 Jul 2026rubric v1 · ranks brands61
13 Jul 2026rubric v1 · ranks brands64
20 Jul 2026rubric v2.0 · ranks brandsBREAK69
21 Jul 2026rubric v2.0 · ranks worksBREAK70
27 Jul 2026rubric v2.0 · ranks works70
4 Aug 2026rubric v2.0 · ranks works71

Tracked since 23 Jun 2026. Live movement between snapshots is on the movers page; the 30 September 2026 snapshot is the Edition 01 baseline.

USE THE DATA

The scores are open. Cite them.

The full ranking is available as a machine-readable feed — JSON for applications and AI engines, CSV for analysts. The licensed tiers add refresh cadence, segmentation and team citation rights.