Research Papers, Validation, and Brand Analysis

Papers

The two anchor papers: the foundational specification and the large-scale empirical validation. Together they define what SBT claims and whether those claims hold against data.

Foundational

Empirical Validation

The research program spans 30+ papers across seven layers — from mathematical foundations and geometric proofs to applied case studies, cross-domain extensions, and meta-science methodology. The complete catalog with reading paths, audience entry points, and open problems is on the Atlas page.

Prior Art: Structural Comparison

Spectral Brand Theory builds on and differentiates from a thirty-year lineage of brand frameworks. The right comparison is not "which framework is best" — that depends on the decision being made and the ecosystem already invested in — but "which structural properties does each framework provide by design." The table below uses ten properties chosen to test what a formal coordinate system for brand perception would have to provide. The criteria are deliberately neutral with respect to popularity, training base, MBA-curriculum coverage, or consulting-ecosystem depth. They test what each framework can express, not how widely it has been adopted.

Legend: ✓ — provides by design, ∼ — partial / domain-limited, — — does not provide.

How to read this table without self-deception

The right-hand column is fully filled, and that pattern is suspicious. The reason it is filled is not that SBT is magically superior — it is that the ten properties were chosen to test the structural design of a formal coordinate system, which is precisely what SBT is trying to be. If the criteria changed to "depth of practitioner adoption," "MBA curriculum coverage," "academic citations over thirty years," or "validated empirical scale instruments," the picture would invert: Aaker, Keller, and Kapferer would be near-fully covered; Brakus and J. Aaker would have substantial coverage from their factor-analytic validation studies; SBT would have one reference dataset (the R15 study) and a stack of working papers.

The table answers a specific question — "which structural properties does each framework uniquely provide by design" — not the question "is SBT the practically better choice today." Those are different questions, and conflating them is the easiest way to oversell a young framework.

What the table actually shows

Reading row by row, almost every property exists in some partial form somewhere in the lineage. Aaker's four perspectives are an early move toward dimensional decomposition. Kapferer's six-facet prism is a richer attempt with the same intent. J. Aaker's Brand Personality Scale provides factor-analytic validation that none of the identity-side frameworks have. Brakus et al. add experiential decomposition with confirmatory factor analysis. Holt acknowledges that brands operate inside cultural moments that observers traverse in different orders. None of these properties is SBT's invention.

What is uniquely SBT is that all ten properties are unified under one coordinate system. No single predecessor covers more than three or four cells. Aaker covers identity decomposition but offers no metric, no cohort weights, no falsifiability structure. Keller covers consumer-side equity but treats the consumer as singular. Kapferer covers identity-prism facets but does not parameterize observer heterogeneity. J. Aaker validates a personality structure but cannot model the same brand having opposite personalities for different cohorts simultaneously. Brakus validates experiential dimensions but cannot model dark signals or path-dependence. Holt theorizes cultural disruption but does not provide a measurement instrument.

The honest one-sentence claim: SBT is the first attempt to provide a computable coordinate system for brand perception in which the observer's spectral profile is a first-class parameter, distances between brand positions are formally defined, and the entire apparatus is machine-readable end-to-end so AI agents can consume it directly. Everything else — the eight specific dimensions, the dark signal taxonomy, the coherence type catalog — falls out of that single architectural commitment.

Why convergence, not replacement

The relationship between SBT and the predecessors is formalization, not replacement — analogous to how behavioral economics formalized classical economics by adding the observer back as a parameterized object. Classical brand frameworks remain valid as special cases: when all observers in a market share the same spectral profile (homogeneous market with low cohort variance), SBT reduces to a single-vector representation that is empirically indistinguishable from the classical models. SBT becomes necessary when observer diversity is high — which is precisely the contemporary situation under globalization, digital fragmentation, and AI mediation. The argument is not that SBT outcompetes Aaker as a workshop vocabulary or Keller as a teaching pyramid; it is that the seams between those tools become unbearable when AI agents start consuming brand specifications directly, and at that point a unified machine-readable coordinate system stops being a luxury.

A worked formalization of Aaker's four-perspectives model into SBT's eight dimensions appears in "From Brand Identity to Spectral Identity" (2026n). A worked example of how the framework reveals what classical tools cannot is in the Dove case study (R10, 2026p), which decomposes the same brand into cohort-divergent activation profiles that any single-score personality or experience instrument would average away.

Illustrative Brand Analyses

Five brands spanning luxury, mass-market, mission-driven, technology, and hyperlocal niche. Each received a full seven-module spectral audit. These analyses are demonstrations of the analytical framework — they show what SBT produces and how it structures brand diagnosis, not empirical validation of the framework's theoretical claims. All five produced identical coherence type and grade when executed by Claude Opus 4.6 (primary) and independently replicated by Gemini 3.1 Pro, confirming results are framework-driven, not model-specific. All five brand profiles have been retroactively validated against the mathematical bounds from R1-R7: every profile passes metric axioms (R1), no metameric pairs exist (R2), positioning capacity is unconstrained (R4), and trajectory risk is low across all brands (R6), with per-dimension velocity tracking confirming no dimensions approach absorbing boundaries.

Each analysis produces two output layers: an L1 spectral profile (the full multi-dimensional tensor — eight dimension scores per cohort, weights, tolerances, cloud formation modes) and L2 rendered summaries (coherence type label, grade, narrative). The grade is a projection of disruption resilience, not a quality score. Two brands with the same grade can have structurally different spectral profiles — the same way different light spectra can produce the same perceived colour (spectral metamerism).

Cross-Model Replication

The analytical pipeline was executed independently by Claude Opus 4.6 (primary) and Gemini 3.1 Pro (replication) on all five brands. Results:

Model-sensitive variations exist in cohort granularity (Claude identifies 5-6 cohorts, Gemini synthesizes 3) and D/A ratio estimates (within 10-15 percentage points). These variations do not affect the structural diagnosis. The framework acts as a structured lens that constrains analysis toward the same architectural findings regardless of which model holds the lens.

Explore Further

Browse the full 30+ paper catalog on the Research Atlas. Learn the framework behind these analyses at the theory page. Run the same pipeline on your own brand with the audit toolkit. Read the full article series for deep dives into each brand and concept. All analytical terms are defined in the glossary.