Qualitative analysis of artist interviews

Researchers and UX teams often sit on rich artist interviews that are impractical to analyze at scale. This Lenscratch conversation (Emma Ressel ↔ Tanya Marcuse, published Aug 19, 2025) is a 6, 000+ word qualitative goldmine covering practice, methods, materials, and place-based observation. In this post you’ll learn a reproducible workflow for qualitative analysis of artist interviews using AI: how to ingest transcripts, auto-code emergent themes, compare segments (e.g., practice vs. place), and produce stakeholder-ready visuals. We show exactly how to map excerpts like “Big Night” amphibian migration, diorama research, and Portent staging into thematic reports, and how Evidano accelerates each step while preserving privacy. See the original interview on www.lenscratch.com and try the process on www.evidano.com.

Fast take + source

What happened: Emma Ressel interviewed Tanya Marcuse in a long-form artist-to-artist conversation that Lenscratch published on Aug 19, 2025. The dialog traces Marcuse’s multi-part Book of Miracles (Parts I–IV, ongoing through 2025), methods (studio tent, outdoor tableaux), and recurring themes (decay, dioramas, vultures, amphibian migration).

Original source: www.lenscratch.com/2025/08/photographers-on-photographers-emma-ressel-in-conversation-with-tanya-marcuse/
Why it matters: interviews like this are rich for thematic, frequency, and cross-segment analysis when processed with AI-enabled qualitative tools
Try it on Evidano: ingest the interview, run thematic clustering, and generate visuals at www.evidano.com

Findings snapshot

Date	Interviewers	Key topics	Representative themes	Notes
Aug 19, 2025	Emma Ressel & Tanya Marcuse	Book of Miracles; Portent; Woven; dioramas; vultures; amphibian migration	Nature-artifice tension; cycles of decay/growth; field-based observation; staged tableaux	Long-form text (≈6k+ words); rich quotes for coding
2005–2025	Marcuse body of work	Fruitless (2005), Woven, Fallen, Book of Miracles (Parts I–IV through 2025)	Studio construction vs. in-situ staging; material specificity to Hudson Valley	Use dates and project names to create segment filters
Examples	,	Big Night amphibian migration; Valley of the Vultures; diorama research	Practice-level motifs and logistics detail (lighting, freezing, stitching photography)	Good test cases for method / logistics coding

What happened & why it’s useful for qualitative research

Nuts-and-bolts: the Lenscratch piece is a verbatim artist conversation that mixes studio description, fieldwork anecdotes, technical workflow notes (e.g., stitching 30–40 frames), and cultural reflection (museum dioramas, conservation). It contains precise dates (e.g., Marcuse began Fruitless in 2005; Book of Miracles unfolds 2021–2025) and project-level terms that make it ideal for automated thematic extraction.

Data type: long-form interview / published transcript (~6, 000+ words) with embedded project names, dates, and place-specific terms (Hudson Valley, Bard College)
Why AI helps: recurring motifs (decay, diorama, staging) are latent across paragraphs; AI-enabled coding reduces manual tagging and surfaces co-occurrence patterns
Research outputs: thematic clusters, code frequencies, quote banks, co-occurrence networks, and segment comparisons (e.g., studio vs. field quotes)

Implications for researchers, curators, and UX teams

For qualitative researchers

This interview exemplifies multi-level coding needs: descriptive (dates, techniques), interpretive (themes like hubris and preservation), and procedural (materials, failure modes). Automated thematic analysis saves hours of hand-coding and enables iterative codebook refinement.

Use case: extract all mentions of techniques (e.g., 'stitching 30 or 40 frames') to estimate resource intensity across artist practices.

For curators & catalogers

Turn quotes and project metadata into exhibition labels and provenance notes quickly. For example, pull canonical quotes about 'Valley of the Vultures' and pair them with image metadata for captions and conservation rationales.

For UX / product teams studying creator behavior

Map pain points (equipment constraints, failure tolerance) and rituals (martial arts discipline, field trips) to feature opportunities, e.g., tools that support long-term observational workflows or field-to-studio handoffs.

Do more, faster with Evidano (mapped to this use case)

Ingest & clean

Problem: long, mixed-format interviews with inline project names and dates.

Evidano: import transcript or URL, auto-detect speaker turns, apply custom dictionary (e.g., 'Portent', 'Woven', 'Big Night') and redact PII if needed.

Auto-code thematic clusters

Problem: manual thematic coding is slow and inconsistent.

Evidano: AI-assisted code suggestions plus codebook import; run thematic and frequency analyses to surface dominant motifs like 'diorama', 'decay', 'staging'.

Cross-segment comparisons & quote banks

Problem: comparing studio vs. field statements across long interviews is tedious.

Evidano: filter by segment (e.g., references to 'studio tent' vs 'vernal pool'), export top exemplar quotes and frequency tables for reports.

Visualizations for stakeholders

Problem: stakeholders want digestible evidence, not raw text.

Evidano: generate word clouds, co-occurrence networks, hierarchical code→subcode trees, and downloadable CSVs for slide decks.

Secure, research-first platform

Problem: sensitive interviews and unpublished notes require strict controls.

Evidano: end-to-end encryption, private LLMs tuned for qualitative research; customer data is never used to train third-party models.

Checklist: 7-step workflow to reproduce this analysis

Step-by-step: a compact runbook to transform the Lenscratch interview into stakeholder-ready insights.

1) Import: paste the transcript or upload the article PDF to Evidano; set a custom dictionary for project names and place terms.
2) Auto-segment: detect speaker turns and tag metadata (date published Aug 19, 2025; source Lenscratch).
3) Seed codebook: add codes for 'diorama', 'vultures', 'Big Night', 'stitching', 'obsession', 'failure tolerance'.
4) Auto-code: run AI-assisted coding, then review and merge suggested codes.
5) Analyze: run thematic frequency, co-occurrence network, and segment cross-tabs (studio vs. field).
6) Extract quotes: compile a quote bank of top exemplar quotes per theme for labels and captions.
7) Export & share: generate visuals and a one-page executive brief for curators or stakeholders.

FAQ: qualitative analysis of artist interviews

How do I compare segments (studio vs. field)?

Tag passages during import or run a keyword-based segmenter in Evidano, then produce cross-segment frequency and representative quotes.

Can I preserve nuance when using AI coding?

Yes. Use AI suggestions as a first pass and lock or adjust codes during human review; Evidano preserves source offsets so quotes map back to original text.

Is this secure for unpublished interviews?

Evidano uses end-to-end encryption and private models; data is not used to train third-party models, suitable for sensitive research.

Wrapping up & next steps

If your team analyzes interviews, artist conversations like the Lenscratch piece are low-hanging fruit for thematic discovery and stakeholder storytelling. Use the 7-step checklist above to produce reproducible insights and visuals in days, not weeks.

Start: open the Lenscratch source at www.lenscratch.com/2025/08/photographers-on-photographers-emma-ressel-in-conversation-with-tanya-marcuse/ and export the text.
Try it on Evidano: ingest the interview, run thematic and cross-segment analysis, and generate visual reports at www.evidano.com, book a demo to map this workflow to your corpus.

Keep reading

Browse all articles