Method note
What was measured, how it was measured, and where the read stops.
A Gen Z creator panel, by design
Every creator in this panel is Gen Z by design, so every figure on this site is a Gen Z creator read.
This is creator-side data. It measures the Gen Z creators making the content, not the audience watching it. The brief asks about Gen Z Connector audiences, and audience-level demographics are not part of this dataset.
Three dig sources, and what each one covers
dig Creator Panel
- Scope
- 60 Gen Z creators, 30,079 posts, 2.85B views
- Window
- Aug 2025 to Aug 2026
- Networks
- TikTok, Instagram, YouTube
- Role
- primary source for affinity, aesthetic and driver signals
dig Category Set
- Scope
- category-scale handbag conversation, 56 tracked brands
- Window
- Oct 2025 to Aug 2026
- Networks
- six networks including paid placements
- Role
- brand share of voice at scale
dig Broad Creator Set
- Scope
- 5,556 creators, brand-attributed at source
- Window
- Jun to Dec 2025
- Networks
- TikTok, Instagram, YouTube
- Role
- independent validation of the brand gap
The dig chatbot produces qualitative analysis over the same creator content: it reads posts rather than counting tags. It is a different evidence class from the tagged measurements and is labelled separately everywhere it appears, with its own marker in the legend below.
Charts and evidence posts returned reliably; written analysis returned intermittently. Only confirmed, completed answers are carried here.
One dataset was reviewed and set aside
A fourth dataset, brand-mention monitoring across a much larger creator base, was reviewed and excluded: it carries no interest or affinity dimension and no brand attribution, so it cannot support an audience read.
How the cohort and the measures are built
Posts where the brand is named and the bag is the subject of the post, rather than a passing mention in a multi-brand list.
Affinity index = the rate of a signal inside the brand cohort divided by its rate across the full creator panel. Overlap = the share of the brand cohort carrying that signal.
An affinity of 1.00 means the brand cohort talks about a signal at exactly the panel baseline rate. Above 1.00 is over-indexing, below is under-indexing.
Every cohort was re-tested with a subject filter
Brand attribution in the creator panel is text-based, so every cohort was re-tested with a subject filter to remove passing mentions. Only signals that held under the tighter cohort definition are reported.
Affordable fashion read 1.50 before passing mentions were removed and 0.28 after, so it sits outside the reported set.
The four territory sub-scores
- Consumer relevance
- Share of the Kate Spade cohort carrying the signal, rescaled 0 to 100 across the 23 signals
- Right to play
- Kate Spade affinity index against the panel baseline, rescaled 0 to 100
- whiteSpace
- Kate Spade affinity divided by Coach affinity, rescaled 0 to 100. Higher means Kate Spade leads Coach by more on that signal
- Momentum
- Trend slope of the signal's monthly share of the Kate Spade cohort, rescaled 0 to 100
- Composite
- Unweighted mean of the four. The weighting model recomputes it against a 100 point budget.
- scaleNote
- Each sub-score is relative to the strongest signal in this set, so 100 means highest observed, not a perfect score. The raw measured value sits beside every bar.
- independenceNote
- White space replaced an earlier evidence dimension that measured the same quantity as consumer relevance. The four dimensions still intercorrelate between 0.33 and 0.88, so this ranks and compares signals rather than acting as four independent axes.
The four sub-scores are the only derived numbers in this app. Every raw value behind them is shown next to the bar it feeds.
Two themes were removed from the interest lexicon
Film & TV and Sport & fitness were removed from this lexicon set. Cross-checking against a separate corpus-wide analysis showed both were inflated by ambiguous terms: 'show' matched 'let me show you' and 'showcase', and 'run' matched 'run out'. Their corrected values appear in the cultural-passions analysis instead.
How segmentation was tested
Segmentation was attempted by clustering all 60 panel creators on their measured signal profiles, testing three through seven segments and scoring each solution for separation. The best solution placed 51 of 60 creators in a single group and isolated one creator alone, with a separation score of 0.20 against a 0.25 threshold for usable structure. The panel is too small and too homogeneous to support segments, so none are reported.
What the cultural read covers
The cultural read covers food, music, film and TV, art, books, gaming and sport, measured across 5,431 posts. Deeper cultural terrain is addressed through the audience-level pull set out in the next steps.
Scope of this read
- The creator panel is 60 creators. It is a curated panel, not a population sample.
- The panel is 35 percent US by creator and 34 percent by post volume. The brief is US only.
- This is handbag-category data. It measures how the bag conversation differentiates, not general culture.
- Brand attribution in the creator panel is text-based, so every cohort was re-tested with a subject filter to remove passing mentions. Only signals that held under the tighter cohort definition are reported.
What this read covers
- Cultural passions are read from qualitative analysis over the creator content and reported as shares of the matched set, alongside the tagged fashion and lifestyle measurements.
- Media habits cover TikTok, Instagram and YouTube, the three networks this panel publishes on.
- Community participation is read from what creators say, so it covers online expression rather than offline events.
- Sizing in this read is panel reach. US population sizing would come from a purpose-built audience pull.
- Narrative opportunities are reported as narrative ground supported by measured signals, each with a way to prove it out.
Every figure in this app is measured or analysed from the creator content described above. Nothing is illustrative.