Researchers analyzing online wildlife trade face two problems: scale (thousands of views and posts) and nuance (species, local terms, and buyer intent). A recent study of TikTok in Togo (80 videos posted between November 2022 and April 2024) flagged >3, 500 animals from at least 27 species and nearly 1.8 million total views (source: www.theconversation.com/tiktoks-online-wild-meat-sellers-study-finds-endangered-species-on-offer-in-west-africa-262807). In this post you’ll get a repeatable workflow for a qualitative analysis of wildlife trade on social media: how to ingest, code, surface demand signals, and produce secure, stakeholder-ready outputs using AI-enabled tools like www.evidano.com.
Findings snapshot (from the study)
| Metric | Value | Source / Note |
|---|---|---|
| Videos analysed | 80 | TikTok posts, Nov 2022–Apr 2024 |
| Individual animals advertised | >3, 500 | Authors’ content analysis (Nature Conservation repository) |
| Identifiable species | ≥27 | Mammals 78% • Birds 15% • Reptiles 7% |
| Total reach | ≈1.8 million views | Some videos up to 216, 000 views |
| Notable endangered finds | White-bellied pangolin (IUCN endangered; CITES App I) | Illegal trade risk highlighted |
What the authors did and why it matters
The research team sampled 80 open TikTok videos from two creators in Lomé and coded visible carcasses, species identity, and viewer comments to estimate demand and reach. The method reveals how social platforms magnify local markets into broader demand signals that can accelerate species decline and public-health risks.
- Scope: Nov 2022–Apr 2024; visible posts only (open accounts).
- Key outputs: species lists, counts, viewer engagement, and comment-driven demand for additional species (e.g., lions, leopards).
- Implication: social media acts as a storefront that can shift urban consumption patterns and scale illegal trade.
So what for researchers, policy teams and UX/ops analysts
For conservation researchers
Use social-media corpora to triangulate market scope and capture species-level evidence at scale.
Track comment-level demand signals to identify emerging species targets before supply records appear in markets.
For policy & enforcement
Quantify reach and probable distribution channels (courier vs. in-person pickup) to prioritise surveillance zones and legal responses.
Pair digital evidence with on-the-ground patrols and public-awareness campaigns.
For UX/product & platform teams
Improve automated detection by training classifiers on local imagery and language variants; policy rules must reflect regional species and slang.
Design moderation flows that surface suspected illegal wildlife listings for expert review rather than relying solely on automated takedowns.
How Evidano maps to this use case
Ingest & clean the corpus
Evidano scrapes social posts and ingests videos, captions and comments into a searchable corpus. For multilingual or local terms, custom dictionaries improve species matching and reduce false negatives.
Rapid thematic & frequency coding
Run automated thematic extraction to surface species mentions, selling language, and demand phrases; frequency tables show which species and terms recur across time and creators.
Cross-segment and co-occurrence analysis
Compare urban vs. rural commenters, time windows, or hashtags; generate co-occurrence networks (species ↔ buyer intent ↔ price mentions) to spot high-risk clusters for enforcement.
Document-level review and evidence packs
Export clickable quotes, timestamps and annotated screenshots for legal teams or policymakers. Evidano’s secure reports bundle provenance (post URL, date, view count) for chain-of-evidence needs.
Ethics & data security
Data is encrypted and not used to train third-party models; all analyses are research-focused. When exploring public health signals, treat findings as non-diagnostic and for policy/research only.
Two-week practical workflow (reproducible)
This lightweight workflow turns raw TikTok/Instagram content into actionable insight in 10–14 days.
- Day 1–2: Define scope and build a keyword/dictionary list (local species names, slang, trade terms).
- Day 3–5: Scrape posts, captions and comments; ingest media into Evidano with metadata (date, URL, view count).
- Day 6–8: Run automated coding (themes, species, pricing, buyer intent) and validate with a 5–10% manual sample.
- Day 9–11: Cross-segment analysis (commenter geography, time trends) and co-occurrence mapping to find demand clusters.
- Day 12–14: Produce an evidence pack (clickable quotes, images, timelines) and a short decision memo for enforcement or platform moderation.
FAQ: common questions when doing qualitative analysis of wildlife trade
Can short-form video be reliably coded for species?
Yes, with a combination of automated object recognition tuned to local fauna and human validation. Use confidence thresholds and spot checks to manage false positives.
How do you compare segments (e.g., urban vs. rural viewers)?
Extract commenter metadata where available, cluster by language/hashtags/timezone, and run cross-tab frequency and sentiment comparisons in Evidano.
How do you handle sensitive or potentially illegal content?
Flag suspected illegal listings for expert review, redact PII, and maintain secure storage. Evidano supports PII redaction and encrypted data handling.
Conclusion: what to do next
The TikTok study (Nov 2022–Apr 2024) shows how social platforms can amplify wildlife markets and surface endangered species in plain sight. For researchers and policy teams, AI-enabled qualitative workflows let you move from scattered posts to quantified demand signals and evidence-ready reports.
- Start by building a local dictionary and a two-week pilot corpus (50–200 posts).
- Then run automated coding, validate results, and produce a stakeholder-ready evidence pack.
- Want to pilot this workflow? Explore a demo or start a secure trial at www.evidano.com and turn social-media noise into policy-ready insight.
Keep reading
- Commentary on NewsTwo Definitions: Climate Change Acceptance for UndergradsHow a PLoS One Delphi study (Aug 25, 2026) defined climate change acceptance for undergraduate science students, and how AI-enabled qualitative analysis applies it.
- Commentary on NewsResearcher-in-the-loop: AI-enabled UX researchHow the researcher-in-the-loop model governs AI-enabled UX research. Learn practical governance, stats from the August 2026 piece, and how Evidano supports this workflow.
- Commentary on NewsResearcher-in-the-Loop: Governance for AI UX ResearchGovern AI in qualitative UX research with the researcher-in-the-loop model from Jennifer L. Bowie (Aug 25, 2026): practical rules, risks, and tool mappings.