TRAFFIC Wildlife Trade: Intelligence Source Guide

TRAFFIC is the wildlife-trade monitoring NGO whose seizure records, market surveys and CITES-facing analysis form the open evidence base for illegal wildlife trade. It is where you go when you need a defensible, sourced account of what was seized, where, and in what volume.

At a glance

Source	TRAFFIC Wildlife Trade
Category	Conflict, Crime & Human Security › Environmental & Wildlife Crime
Homepage	https://www.traffic.org/
Format	`HTML`
Access	Open — no account required
Disciplines	Environmental Intelligence, Geospatial Intelligence
Mission domains	Wildlife Trafficking

Wildlife trade monitoring network. — as catalogued in the platform’s own source registry.

TRAFFIC is a non-governmental organisation that monitors trade in wild animals and plants, both the legal trade regulated under CITES and the illegal trade that runs alongside it. Its output has three distinct shapes and you should not confuse them. First, a long series of investigative and analytical reports – species-specific (ivory, rhino horn, pangolin scales, tortoises, timber, totoaba, saiga, songbirds), market-specific (a survey of physical markets in a named city, or of listings on a named e-commerce platform), and corridor-specific (a trade route between two countries). Second, the TRAFFIC Bulletin, a periodical running since the late 1970s that publishes shorter analyses, seizure round-ups and species notes. Third, an open seizure and incident database, published as the Wildlife Trade Portal, which exposes structured records drawn from TRAFFIC's internal wildlife trade information system: an incident, a date, a place, a taxon, a commodity, a quantity, and a source reference. TRAFFIC also manages the Elephant Trade Information System on behalf of the CITES Parties, which is the mandated repository for ivory seizure data reported by governments and the analytical basis for the ivory-trade assessments presented at CITES Conferences of the Parties.

The analytical job TRAFFIC does that nobody else does at the same quality is convert scattered national seizure announcements into a comparable, taxon-normalised, geolocated series. A customs press release says officers found bags of scales at an airport. TRAFFIC turns that into a record with a species-level or genus-level identification where possible, a converted weight, an inferred origin and destination, and a citation back to the original announcement. That normalisation is the whole product. Without it you cannot compare a Vietnamese seizure to a Nigerian one, you cannot build a route, and you cannot say anything about volume trend that survives contact with a critic. For ENVINT work this is the primary open collection route on wildlife crime; for GEOINT work the value is the route geometry – port pairs, transit hubs, land border crossings – that emerges once seizures are placed on a map with consistent attributes. It is also the bridge between conservation evidence and law-enforcement evidence: TRAFFIC's reports are written to be usable in CITES compliance proceedings and national prosecutions, which means they carry sourcing discipline that campaign material does not.

Who publishes it, and why that matters

TRAFFIC began in 1976 as a joint programme of WWF and IUCN and remains closely tied to both, operating today as an independent organisation headquartered in Cambridge, United Kingdom, with regional offices across Africa, Asia, Europe and the Americas. Funding comes from a mix of conservation foundations, bilateral aid agencies, the European Union, national governments and multilateral programmes, which has two consequences you should hold in mind. Coverage follows donor priorities: species and corridors that attract funding get sustained monitoring, and those that do not get episodic attention or none. And because TRAFFIC works alongside CITES management authorities and national enforcement agencies, it operates under access arrangements that constrain what it publishes – some of what it knows appears only in aggregate, or in documents circulated to Parties rather than to the public. This is an organisation with genuine longevity and a stable institutional identity, which makes it more reliable as a long-run series than most conservation data providers. It is not a neutral statistical agency and does not claim to be; it is an advocacy-adjacent research body whose advocacy is unusually well disciplined by evidence.

Provenance is the first question to ask of any dataset and the one most often skipped. Who collects it, what their incentive is, whether they publish a methodology, and whether they correct the record when they get something wrong all bear directly on how much weight a finding drawn from it can carry.

What a record actually contains

The fields you will be working with, what each one means, and whether it is something you can pivot on. Read the meanings carefully — more analysis is wrecked by misreading a field than by failing to find one, and a field that looks like an observation is often an inference.

Field	Type	What it means	Pivot value
`Incident date`	timestamp	The date of the seizure or the date of the reported event. In a meaningful minority of records this is the date the incident was reported rather than the date it occurred, because national announcements often lag by weeks. Records derived from court reporting can lag by years.	Align against shipping schedules, seasonal harvest windows, CITES quota years and other seizures in the same corridor within a window.
`Seizure country`	string	Where the commodity was found. This is the single most reliable field and the one most often over-interpreted: seizure country reflects enforcement capability, not trafficking volume, so a country with strong customs scanning appears worse than a neighbour with none.	National CITES authority, customs administration, port and airport infrastructure, other incidents in the same jurisdiction.
`Location`	string	Sub-national place: airport, seaport, land border post, market, road checkpoint or address. Granularity varies enormously – some records name a terminal, others name only a province.	Geocode to a transit node and build a route graph; cross-reference against port call and flight data.
`Taxon`	string	The species or higher taxonomic rank identified. Identification quality is the biggest hidden variable in the whole dataset – a customs officer's naked-eye call, a specialist's morphological identification and a laboratory DNA result all land in the same field.	IUCN Red List assessment, CITES appendix listing, national protected-species schedules, range-state geography.
`Commodity or product type`	enum	What form the material was in: whole live animals, raw tusks, worked ivory, scales, skins, bones, horn, meat, timber logs or sawn wood, derivatives in traditional medicine preparations. Product form is a strong signal of position in the supply chain.	Processing infrastructure, likely end-market, and the point in the chain where the interdiction occurred.
`Quantity and unit`	string	Number of specimens, weight, or volume, with the unit as reported. Units are heterogeneous by design because the original announcements are heterogeneous, and any conversion to a common measure is an estimate with an error bar.	Convert to estimated individual animals using published conversion factors, but always report the conversion assumption alongside the figure.
`Origin, transit and destination`	array	The countries a shipment is believed to have passed through. Some of this is documentary – the manifest, the container's routing – and some of it is inference from the enforcement agency. The record rarely tells you which.	Route reconstruction, carrier and freight forwarder identification, corridor-level trend analysis.
`Transport mode`	enum	Air passenger baggage, air cargo, sea container, postal and courier, road, or concealed within a legitimate consignment. Mode is the field that most usefully predicts what enforcement intervention would work.	Carrier, route, freight documentation, and the specific enforcement agency with jurisdiction.
`Arrests and prosecution status`	string	Whether anyone was detained, charged or convicted. Overwhelmingly incomplete and asymmetrically so: arrests are announced, acquittals and discontinuances are not, so the field decays towards a false impression of enforcement success.	Court records, national judicial databases, and follow-up reporting in the TRAFFIC Bulletin.
`Source reference`	string	The citation the record rests on – a government press release, a customs bulletin, a news report, a court filing, or a TRAFFIC field observation. This is the field that determines how much weight a record can bear, and it is the first thing a serious analyst reads.	The primary document itself; go to it before you cite the derived record.
`Species protection status`	enum	The CITES appendix and, where recorded, the IUCN Red List category at the time of the incident. Both change over time, so a record's status label may reflect the listing at ingest rather than at the date of the incident.	CITES appendices and their amendment history; IUCN Red List assessment history.
`Record type`	enum	Whether the record is a seizure, a market observation, an online listing, a poaching incident or a prosecution. Mixing record types in a count is the single most common analytical error made with this data.	Filter to a single type before any trend claim; compare types only as separate series.
`Market or platform`	string	For survey-derived records, the physical market, shop or online platform where the item was observed. Present only in survey outputs, absent from seizure records, and never comparable between the two.	Platform policy and enforcement posture, vendor identity where lawfully collectable, repeat-observation analysis across survey rounds.
`Report or bulletin citation`	string	For narrative outputs, the TRAFFIC publication in which the analysis appears, with its year. Older reports remain online and are frequently cited as if current; check the publication year before quoting a figure.	The publication itself, its methodology annex, and any later edition that supersedes it.

Coverage — and what is not in it

Coverage is global in ambition and deeply uneven in practice. It is strongest where TRAFFIC has had sustained regional programmes and donor support: East and Southern Africa for ivory, rhino horn and pangolin; South East Asia and China for consumption-end markets, pangolin scales, tiger parts and reptiles; South Asia for tiger, red sanders and turtle trade; and the major transit corridors linking them. European coverage is decent for imports, eel trade and the pet and reptile market. Coverage of the Americas is present but thinner outside specific programmes such as jaguar parts, totoaba and timber. Central Asian, Middle Eastern and Pacific coverage is episodic. Temporally, the narrative record runs back to the late 1970s through the Bulletin, and the structured seizure record is deeper for the last two decades than before. Update rhythm differs by product: reports appear when a piece of work finishes, the Bulletin on a periodical schedule, and the seizure database in batches as new incidents are compiled and verified rather than in real time. Do not build any workflow that assumes same-week ingestion of a seizure. Entity types are incidents, taxa, places and commodities – not people. Individual suspects and traders are named only where they are already public through court proceedings, and often not even then.

Known blind spots

Absence of evidence here is not evidence of absence. These are the conditions under which TRAFFIC Wildlife Trade will not show you something that is nevertheless real:

Seizures are a sample of interdiction, not of trafficking. Every quantitative statement built from this data is a statement about enforcement activity multiplied by trafficking activity, and you cannot separate the two factors from within the dataset. A country that stops scanning containers appears to have solved its wildlife crime problem.
Domestic and subsistence-scale trade is almost invisible. The dataset is oriented towards cross-border commercial movement, which means bushmeat markets, local medicinal trade and small-scale harvesting – collectively enormous – are represented only where a specific survey happened to look.
Plants, timber, fish and invertebrates are systematically under-recorded relative to charismatic mammals. Timber and marine species make up a large share of the value of wildlife trade and a small share of the reporting, partly because identification is harder and partly because they attract less funding attention.
Records that a government has asked not to be published, or that arrived under an access arrangement with an enforcement agency, do not appear. Absence from the public record does not mean absence from TRAFFIC's knowledge, and this gap is largest for sensitive corridors involving state complicity.
Online trade coverage is a snapshot of whatever platforms a given survey looked at, in whatever language it searched. Closed groups, encrypted messaging, regional platforms outside the surveyor's language set, and any platform not selected for the study are simply not in the data, and their absence is invisible.
The identification chain is not recorded. A record saying pangolin scales may rest on a laboratory result or on a customs officer's assumption, and the field structure does not distinguish them, so species-level claims carry silent and highly variable uncertainty.
Prosecution outcomes are not systematically followed. The dataset captures the moment of interdiction and rarely the disposal, which means it cannot support any claim about conviction rates, sentencing or deterrence without external court research.
Corruption and state involvement are structurally hard for this source to show. Where an official facilitated a shipment, the incident is either not seized, not announced, or announced in a form that omits the facilitation, so the corridors most dependent on official complicity look quietest.
The absence of a corridor in the data frequently means nobody surveyed it. TRAFFIC's regional footprint determines what gets compiled, so a clean map of Central Asia reflects the office structure of a Cambridge-based NGO, not the behaviour of traffickers.

Write the blind spot into the product. A statement that something “was not observed in TRAFFIC Wildlife Trade” is defensible; a statement that it “did not happen” is not, and the difference is what survives cross-examination.

Access, licensing and what you may do with it

Access model: Open — no account required

There is no authenticated API and no machine contract you can depend on. Access has three practical routes. The organisation's website carries the report library, and reports are published as PDFs with methodology annexes that are usually worth more than the executive summary. The Wildlife Trade Portal exposes the structured seizure and incident records through a browsable and filterable interface with export, and that is the route to take when you want rows rather than prose. The TRAFFIC Bulletin is published as a periodical archive and is the best route to short-form seizure round-ups and species notes that never became full reports. Practically, plan for a two-stage workflow: harvest structured records from the portal for counting and mapping, and read the reports for the interpretive layer that tells you why a corridor exists. Field names and export formats on the portal have changed since launch, so bind your parser to the header row rather than to column positions, and re-verify your mapping on each pull. For ETIS ivory data, the analytical products reach the public through CITES documentation rather than through TRAFFIC's own channels, and the underlying government-reported records are not public.

Licence

TRAFFIC's publications are copyright works made freely readable, and the organisation's standard practice is to permit reproduction for non-commercial research and educational purposes with attribution, with commercial reuse requiring permission. The seizure records exposed through the portal are published for research and policy use and carry their own terms of use, which have been revised since launch. Do not assume a Creative Commons grant that has not been stated on the page you are actually downloading from. Two further constraints matter more than the licence text in practice. First, many records are compiled from national government announcements that carry their own copyright and reuse conditions, so republishing a full record set may implicate rights you did not acquire from TRAFFIC. Second, data supplied to TRAFFIC by governments under confidentiality arrangements is not licensed to you at all merely because a derived figure appeared in a report. Check the terms on the current page before any redistribution or commercial product, and cite the report and its year rather than the organisation generically.

Rate limits and fair use

No published rate limits exist because there is no published API. Treat the site as a courtesy target: fetch the portal export rather than scraping the interface record by record, pull on a weekly or monthly cadence rather than continuously, identify your client honestly, and cache aggressively. Seizure records do not change on an hourly basis and nothing in this data justifies high-frequency polling. If your requirement is genuinely large-scale or ongoing, contact the organisation rather than engineering around it – research collaborations are a normal part of how TRAFFIC works and produce better data access than scraping does.

Licensing changes, and it changes without warning. A dataset that was free for research this year may not be free for commercial or evidential use next year. Confirm the current terms before you build a dependency on it, and record the terms you relied on alongside the data — the licence in force at the time of collection is part of the provenance.

Collecting it

How TRAFFIC Wildlife Trade is actually pulled, in the order you would set it up. Prefer the bulk or export interface over per-item lookups wherever one exists: it is kinder to the publisher, faster for you, and gives a reproducible snapshot rather than a series of point-in-time answers you cannot reconstruct later.

Method	Format	Cadence	Notes
Wildlife Trade Portal export	CSV	Monthly	The structured route. Filter to your taxa, region and record type, export, and store the filter definition alongside the file so the extract is reproducible.
Report library harvest	HTML	Weekly	Monitor the publications listing for new reports and briefings. Capture the PDF, the publication date and the methodology annex; the annex is what tells you whether a figure is comparable to a prior edition.
TRAFFIC Bulletin archive	HTML	Per issue	The periodical carries seizure round-ups and short species analyses that never appear elsewhere. Back issues are the deepest available narrative series on wildlife trade.
CITES documentation cross-pull	HTML	Per CITES meeting cycle	TRAFFIC's ETIS analyses and other Party-facing work surface as CITES conference and committee documents. Pull those alongside the public reports; they often contain the fuller methodological statement.
Manual analyst capture	JSON	Ad hoc	For a specific corridor investigation, hand-extract records into a case structure with the primary source document attached to each. Slower, and the only method that preserves the evidentiary chain properly.

Ingesting it into the platform

Every step below is idempotent and cursor-based: interrupt one and it resumes from where it stopped rather than duplicating rows or losing progress. Collection is recorded per source, so a feed that quietly stops publishing shows up as a stale timestamp instead of silently thinning your coverage.

Register the source and its collection mode — Add TRAFFIC in sources.php with its collection route recorded as manual or scheduled export rather than live API, so that last-collected state on collect.php reflects the truth and nobody downstream assumes freshness the source does not have.
Normalise incident records — Run the export through import.php mapping incident date, seizure country, location, taxon, commodity, quantity, unit and record type into the platform's incident schema, preserving the original strings alongside the normalised values.
Geocode transit nodes — Resolve locations to ports, airports and border crossings so that incidents become route endpoints rather than country-level dots. Country-level plotting destroys the corridor signal that makes this data useful.
Attach the source document — Store the primary reference – press release, customs bulletin, court filing – against each record so that any downstream assertion can be traced to the document rather than to the aggregator.
Tag taxonomy and protection status — Use resolve-tags.php to apply species, CITES appendix and IUCN category tags, recording the listing status as at the incident date rather than as at ingest.
Correlate to geography and country risk — Push aggregates into country.php and country-risk.php so wildlife-crime exposure sits alongside other governance and crime indicators for the same jurisdiction rather than in a separate silo.
Build the corridor view — Use link-analysis.php to render origin, transit and destination as a graph, weighting edges by quantity and count separately, because the two produce different-looking networks and both are informative.
Publish to the mission surface — Expose the resulting incident and corridor set through the platform's environmental crime views and organized-crime.php, and generate case material through cases.php where an incident set has been taken up as an investigation.

Registered sources and their last-collected state are listed in sources.php, and the scheduled chain that keeps them current is in automation.php.

How it is wrong, and how to tell

Every dataset is wrong in characteristic ways. Knowing which ways is the difference between using a source and being used by one, and it is the part of source evaluation most often skipped because it is the part that takes work.

Judged as an evidence base rather than as a statistical series, this is high-quality work. Records are sourced, the sourcing is stated, methodology annexes are published, and the organisation has a long record of not overclaiming – TRAFFIC has repeatedly published findings that cut against conservation advocacy positions, including on the effects of trade bans, which is the strongest available signal that the analysis is not being reverse-engineered from a campaign message. Judged as a measure of the illegal wildlife trade, it is weak, and TRAFFIC says so itself. The fundamental problem is that the numerator is seizures and the denominator is unknown, so no rate can be computed and no trend can be cleanly interpreted. Quality also varies sharply by product line: a species-specific report with fieldwork and laboratory identification is far more reliable than a compiled seizure count assembled from news reporting in a language the compiler reads. The right posture is to trust the individual record about as far as its cited source justifies, trust the interpretive analysis in TRAFFIC's own reports substantially, and treat any aggregate count as an enforcement-activity indicator wearing a trafficking-volume costume.

Characteristic false positives

Double counting across records. A single shipment reported by a customs agency, a national newspaper and a court file can enter as multiple incidents, and multi-country seizures of one consignment can appear once per jurisdiction. Deduplicate on date, place, commodity and quantity before any count.
Species misidentification propagating as fact. A generic identification made at the point of seizure enters the record and is then cited as a species-level finding, which matters enormously when the taxon determines the legal offence and the conservation implication.
Unit and conversion error. Weights reported as gross including packaging, counts of items conflated with counts of animals, and conversions between scales and individual pangolins or between tusks and elephants each introduce multiplicative error that compounds when figures are chained.
Enforcement capacity read as trafficking intensity. The most common analytical failure with this source is inferring that a country with many seizures has a large trafficking problem, when the seizures may indicate a functioning customs service intercepting transit traffic.
Date confusion between incident, announcement and publication. Records dated by announcement create artificial clusters around press cycles and can shift an incident into the wrong quarter or year, distorting seasonality analysis.
Route inference presented as documented routing. Origin and destination attributions are frequently the enforcement agency's assessment rather than manifest evidence, and once in the record they lose the hedging that accompanied them in the original statement.
Stale legal status. CITES appendices and national protection schedules change, so a record's protection label may be wrong for the date of the incident, which can invert a legality assessment entirely.
Survey-derived observations mixed into seizure counts. Market and online observations are a different measurement with a different sampling frame; combining them with seizures produces a series whose movements reflect survey scheduling.

None of these make the source unusable. They make it a source that requires corroboration before an assertion built on it goes into a product, which is true of every source and admitted by few.

Ageing

Individual incident records do not go stale in the sense of becoming wrong – a seizure that happened, happened – but their surrounding context decays quickly in three ways. Prosecution status ages worst: a record showing arrests may correspond to a case dismissed years ago, and nothing in the dataset will tell you. Legal status ages next: species listings change at each CITES Conference of the Parties and in national law between them, so a legality assessment older than a listing cycle needs rechecking. Route relevance ages fastest of all: trafficking corridors reconfigure within months in response to enforcement, conflict, port closures and changes in air freight economics, so a corridor analysis more than two or three years old describes history rather than current geometry. A stale record looks like a confident route claim with no incidents in the last eighteen months supporting it. Reports themselves age unevenly: methodology and structural analysis remain useful for a decade, while volume figures and market prices are unreliable after two or three years.

What this source feeds

A source is only worth what it lets you conclude. These are the disciplines that collect through it, the mission domains it serves and the data points it yields — every one is a tag, so you can follow any thread from here into the rest of the library.

Collected by these intelligence disciplines

Serves these mission domains

Wildlife Trafficking

Yields these data points

How each sector uses TRAFFIC Wildlife Trade

The same dataset is worked very differently depending on who you are, what authority you hold, and what you are ultimately producing. A military analyst is supporting a commander’s decision; a journalist is meeting a publication standard; an NGO caseworker is protecting a person. The records are shared — the constraints, thresholds and outputs are not.

🎖 Military and defence

Relevant where wildlife trafficking intersects with armed groups, border control and stabilisation tasks. The corridor data identifies smuggling routes that are usually shared with other contraband flows, which makes it a proxy for permissive terrain and for the informal logistics networks operating in a theatre. In several African contexts, ivory and bushmeat revenue has been credibly linked to non-state armed group financing, and TRAFFIC's reporting is the open-source substrate for that assessment. Use it to understand who controls movement through a border region and what the local economy of that control looks like, and do not use it as an order of battle input – it is an economic and route-geometry source, not a force disposition one.

🕵 National intelligence

This is a strategic ENVINT collection route and should be run as a standing series rather than a lookup. Its value is in structural pattern: which corridors persist, which reconfigure after enforcement pressure, which countries appear repeatedly as transit points, and which product forms indicate processing capacity in-country. Because the material is open and citable, it is also useful for building assessments that can be shared with partners who cannot receive classified reporting. Pair it with trade, shipping and financial sources for the convergence question – wildlife trafficking routes are rarely single-commodity, and the same freight forwarders, brokers and payment channels usually serve other flows.

👮 Law enforcement

The immediate operational value is corridor and modus operandi awareness: what concealment methods are current on a given route, which transport modes dominate for a given commodity, and which ports have produced repeated interdictions. TRAFFIC reports frequently describe concealment patterns at a level useful for targeting profiles. For casework, the primary source references attached to records let you go back to the original enforcement agency and open a liaison channel. Note that this source names incidents, not suspects, and it is not an evidentiary database – use it for intelligence development and targeting, and obtain evidence through mutual legal assistance and formal channels.

🔍 Private investigation and corporate security

Most private-sector relevance is due diligence rather than investigation. If you are screening a logistics operator, an exotic-pet or traditional-medicine business, a timber importer or a fishing enterprise, the seizure record and market surveys tell you whether the sector, the corridor or the specific commodity has an established illicit dimension. Client-side, the reports support supply-chain risk assessment for CITES-regulated materials in furniture, musical instruments, cosmetics and pharmaceuticals. Be careful about the boundary: attributing wildlife crime to a named company on the basis of a corridor-level pattern is defamatory in most jurisdictions and analytically unsound.

📰 Journalism and OSINT media

For investigative journalism this is the source that lets you say something quantitative without being wrong. It gives you sourced incidents you can verify independently, a corridor context to place a single seizure in, and expert reports you can cite. The discipline that separates good wildlife-crime reporting from bad is refusing to convert a seizure count into a trafficking volume claim. Use the individual records as leads to the original enforcement announcement and to court records, and use the reports for the structural story – who profits, where the demand is, why a ban did or did not work – which is almost always the better story than the tonnage.

🌍 NGO, humanitarian and human rights

For conservation and environmental-justice organisations this is the shared evidence base, and its main contribution is comparability: it lets you place your local observation in a regional context and see whether what you are seeing is unusual. It also supports advocacy that survives scrutiny, because the sourcing is transparent enough to withstand a hostile read. Community-facing organisations should be alert to the human dimension the data does not carry – poaching economies frequently involve coerced or desperately poor labour, and enforcement responses have documented human-rights costs, so the seizure record should never be the only input into a programme design.

🎓 University and research

This is the standard citation base for wildlife trade research and the reason a large empirical literature exists at all. Methodologically, it is best treated as an observational dataset with a non-random, enforcement-driven sampling mechanism, which means the interesting econometric work is about modelling the detection process rather than assuming it away. The archive depth of the Bulletin supports genuine long-run analysis, which is rare in this field. Cite the specific report and edition rather than the organisation, record your extract date and filter definition for the portal, and state the identification-quality limitation explicitly rather than in a footnote.

Playbook: working TRAFFIC Wildlife Trade end to end

A repeatable sequence from first pull to finished product. Each phase states what you are trying to establish, not merely what to click — the objective is a defensible chain of reasoning, not a completed checklist.

Phase 1 — Define the unit of analysis before you touch the data

Decide whether you are studying incidents, specimens, mass, corridors or actors, and commit to it. Each produces a different-looking picture from the same records, and most of the published confusion about wildlife trade comes from analysts silently switching between them. Write the unit down and put it in the caption of every chart you produce.

Phase 2 — Establish the enforcement baseline first

Before interpreting any seizure pattern, characterise the detection capability of each jurisdiction in scope – scanning infrastructure, customs staffing, whether a dedicated wildlife unit exists, and whether any of that changed during your window. This is the denominator problem stated concretely. A corridor that lights up in year three may have acquired an X-ray machine, not a trafficker.

Phase 3 — Scope the taxa and their legal status at the incident date

Pull the CITES appendix and national protection status for every taxon in your set, with effective dates. Legality is time-dependent and jurisdiction-dependent, and an analysis that applies today's listing to a five-year window will misclassify a meaningful share of the records. Record the listing history alongside the taxon.

Phase 4 — Harvest structured records with a reproducible filter

Export from the portal using an explicit, saved filter definition – taxa, region, record type, date range – and store the filter alongside the file with the extract date. The dataset changes as records are added and revised, so an unversioned extract cannot be reproduced and any finding built on it cannot be defended six months later.

Phase 5 — Deduplicate ruthlessly

Match on date proximity, place, commodity and quantity to collapse the same physical shipment reported through multiple channels or in multiple jurisdictions. Do this before counting anything. Expect the duplicate rate to be material, expect it to vary by country, and expect the countries with the most press coverage to be the most affected.

Phase 6 — Separate record types into distinct series

Split seizures, market observations, online listings, poaching incidents and prosecutions into separate datasets and never combine their counts. They have different sampling frames, different collection cadences and different meanings. Any chart mixing them is measuring survey scheduling as much as criminal activity.

Phase 7 — Read the reports for the corridor, not the numbers

Go into the report library for every corridor in scope and read the narrative analysis and methodology annex. This is where you learn why a route exists – the freight economics, the diaspora trading relationships, the specific concealment methods, the regulatory gap being exploited – and none of that is in the structured records. Analysts who only take the rows produce maps with no explanation.

Phase 8 — Geocode to nodes and build the route graph

Resolve locations to specific ports, airports and crossings and build a directed graph with origin, transit and seizure nodes. Weight edges by both incident count and quantity and look at both, because a corridor with few high-mass seizures and one with many small ones represent different trafficking structures and demand different interventions.

Phase 9 — Test the corridor against independent logistics data

Take your top corridors and check them against shipping, air cargo and trade data. A claimed route with no corresponding freight relationship is probably an inference someone made, not a documented movement. A route with heavy legitimate freight and repeated seizures is a genuine concealment corridor and a targeting opportunity.

Phase 10 — Follow a sample of records back to primary sources

Pick a random sample of records – ten per cent is defensible for a corridor study – and retrieve the original press release, customs bulletin or court document. Score each on identification quality, quantity precision and route evidence. That score is your honest confidence statement for the whole set, and it is the difference between an assessment and an assertion.

Phase 11 — Assess the human and governance dimension explicitly

Wildlife trafficking sits inside a governance context: corruption at ports, coerced or impoverished labour at the harvesting end, and enforcement practices that in several documented cases have produced serious human-rights violations. Bring in governance and human-rights sources before recommending an enforcement response, and record which of your corridors depend on official complicity, because those need a different intervention entirely.

Phase 12 — Write the uncertainty into the product, not the annex

State in the body of the assessment that seizures measure interdiction, give the deduplication rate you applied, give the identification-quality score from your sample, and name the jurisdictions where you believe enforcement capability changed during the window. A wildlife-crime product that reads as more certain than the data allows will be dismantled by the first specialist who reads it, and rightly.

The platform ships this as a step-checked workflow in playbooks.php, so progress is recorded against a case rather than held in someone’s head.

What to pair it with

No single source carries a finding. These are the datasets that corroborate, extend or contradict this one — and a source that contradicts is worth more than one that agrees, because it is the only thing that will tell you when you are wrong.

Source	Relationship	What it adds
CITES Trade Database	corroborates	The legal trade record reported by Parties. Essential context: illegal trade routes frequently shadow legal ones, and mismatches between exporter-reported and importer-reported volumes are themselves an indicator.
CITES	prerequisite	The treaty framework that determines what is legal. You cannot classify an incident without the appendix listings, the reservations Parties have entered, and the national implementing legislation status.
IUCN Red List	extends	Conservation status and range data for every taxon in the seizure record, which converts an incident into a conservation-impact statement and constrains plausible origin geography.
UNODC World Wildlife Crime Report	corroborates	The UN's periodic global assessment, built partly on its own seizure database. Useful as an independent aggregation to check TRAFFIC's regional picture against.
Environmental Investigation Agency	extends	Undercover investigative reporting on trafficking networks, naming companies and individuals in a way TRAFFIC generally does not. Higher risk, higher specificity; read alongside rather than instead of.
Wildlife Justice Commission	extends	Investigation-led analysis focused on the criminal networks and their financial structures rather than on trade volumes, which fills the actor-level gap in the seizure record.
INTERPOL environmental crime programme	corroborates	Operational law-enforcement view including coordinated operations whose results provide burst-mode seizure data with a known collection cause.
Global Fishing Watch	extends	Vessel behaviour data relevant to marine wildlife trafficking, IUU fishing and the transhipment patterns that move marine products, which the terrestrial-focused seizure record covers poorly.

Legal, ethical and operational constraints

Using this source is legally uncomplicated in itself – it is published research – but three constraints bite. First, copyright and terms of use: reports and portal records are copyright works with attribution and non-commercial conditions that have been revised over time, and many underlying records carry the rights of the government agency that issued the original announcement. Second, defamation: seizure and market records name places, sectors and occasionally companies, and in most jurisdictions asserting that a named business is engaged in wildlife crime on the basis of a corridor-level pattern is actionable and analytically indefensible. Keep the distinction between an incident occurring at a facility and the facility's operator being complicit. Third, and most importantly, downstream use: wildlife enforcement is an area where intelligence products have contributed to real human-rights harms, including violent anti-poaching practices documented by human-rights investigators. If your product will inform an enforcement response, consider the human-rights due diligence that response requires, and be aware that in some jurisdictions the communities living alongside protected areas are also the communities most exposed to enforcement violence. Where personal data appears – a named individual in a prosecution record – data protection law applies in most jurisdictions on the ordinary terms, and the fact that the information is public does not remove the obligation.

Operational security

Reading published reports leaks little. Systematic querying of the seizure portal leaks more than analysts expect: query patterns, filter combinations and export requests reveal which taxa, corridors and countries you are working on, and that is visible to the site operator and to anyone in the network path. If your investigation concerns a specific corridor where TRAFFIC has an operational relationship with a national enforcement agency, be aware that your interest in that corridor is being expressed to an organisation that talks to that agency. Use ordinary infrastructure separation for sensitive casework rather than querying from an attributable corporate range. Do not contact TRAFFIC staff with case-specific questions unless you have decided that the organisation is a partner in the case, because it is a networked NGO with a professional obligation to work with the authorities in the countries concerned, which is normally an asset and occasionally a problem. Finally, be careful with derived products: a corridor map published with enough specificity is also a targeting product for the people running the corridor, and there is a documented history of traffickers reading conservation research.

Two rules that hold regardless of jurisdiction. Collection that is lawful is not automatically proportionate, and a dataset assembled for one purpose does not carry consent for another. Where the records concern identifiable people, the question is not only whether you may hold the data but whether holding it serves the purpose you are accountable for.

Is it earning its place?

Sources accumulate. Feeds get added during an incident and are never reviewed again, and a decade later the pipeline is carrying dead weight that nobody dares remove. These are the measures that show whether TRAFFIC Wildlife Trade is contributing anything, and they are worth baselining now so the answer is available later.

Share of your incident records that you have traced to a primary source document, tracked over time. If this is below half, your corridor conclusions are resting on an aggregator you have not audited.
Duplicate rate detected during deduplication, broken down by jurisdiction. A rising rate means your matching is improving; large cross-country variation tells you where press-driven double counting is distorting the picture.
Proportion of records with species-level as opposed to genus- or family-level identification, per corridor. This is your ceiling on any conservation-impact or legality claim.
Median lag between incident date and record availability, measured per country. This tells you what freshness claims you can honestly make and where a real-time monitoring requirement cannot be met by this source at all.
Number of corridors in your model with independent logistics corroboration versus those resting on seizure records alone. The second group is your fragile inventory.
Count of assessments you have had to revise after a CITES listing change or a report supersession, which measures whether your legal-status refresh cycle is fast enough.
Analyst time spent on normalisation versus analysis. If normalisation dominates, the fix is a better ingest mapping, not more analyst hours.

Beware of volume. Indicator counts rise easily and say almost nothing. Unique contribution — findings this source produced that no other source in your stack would have — is the measure that matters, and it is usually far lower than anyone expects.

Tradecraft notes

The distinctions that separate a competent analyst from a fast one:

Read the methodology annex before the executive summary. The annex tells you the sampling frame, the identification method and the conversion factors, and it is where you discover that two reports you were about to compare are not comparable.
Never convert seizures into trade volume without stating the interception-rate assumption you used, and never source that assumption to a figure whose own provenance you have not checked. A great deal of published wildlife-trade tonnage traces back to a single unsupported estimate repeated for two decades.
Treat the seizure country and the trafficking country as different concepts and keep them in separate fields. Transit states dominate seizure counts precisely because they are functioning transit hubs with functioning customs, and reporting them as source countries is both wrong and diplomatically damaging.
Watch for enforcement operation artefacts. A coordinated multi-country operation produces a burst of seizures with a common cause; if you do not tag those records, the burst enters your trend line as a change in trafficking rather than a change in looking.
Product form tells you position in the chain. Raw tusks near a range state, worked pieces near a consumption market and semi-processed material in between – a corridor where the product form changes mid-route implies in-country processing capacity, which is a targetable fixed asset.
Distinguish observation from inference in every route claim. Manifest evidence, container routing and an officer's assessment all end up in the same origin field, and only the first two survive an evidentiary challenge.
Follow the legal trade to understand the illegal one. Mismatches between exporter-declared and importer-declared volumes in the CITES trade record, permit anomalies and quota over-use are frequently better leads than seizures, because they surface laundering rather than smuggling.
Be sceptical of market price figures. Prices quoted in wildlife-trade reporting are often single-observation, end-market and headline-oriented, and chains of citation have inflated some of them substantially over time. Use price direction, not price level.
Keep the human-rights lens attached. Poaching labour at the source end is frequently coerced or economically desperate, and enforcement at the source end has a documented record of abuse; an analysis that treats the harvesting end purely as criminality will produce recommendations that cause harm and eventually discredit the work.

Questions analysts actually ask

Can I use TRAFFIC seizure data to estimate how much illegal wildlife trade there is?

Not directly, and you should resist pressure to. Seizures give you the intercepted fraction, and the interception rate is unknown and varies by country, commodity and year. If you must produce an estimate, do it explicitly as seizures divided by a stated interception assumption, present a range rather than a point, and cite the origin of the assumption. Most published tonnage figures fail this test.

Is there an API?

No published, supported API. The realistic collection routes are a filtered export from the Wildlife Trade Portal, harvesting of the report library, and the Bulletin archive. Build your pipeline around scheduled exports with an explicit filter definition, and do not architect anything that assumes a stable machine contract.

How does TRAFFIC relate to CITES?

They are separate. CITES is an intergovernmental treaty with a Secretariat and Parties; TRAFFIC is an NGO that provides analysis into that process, including managing the Elephant Trade Information System on behalf of the Parties. TRAFFIC's ETIS analyses reach the public as CITES meeting documents, and the underlying government-submitted seizure records are not public.

Why does one country appear in so many records?

Almost always because it is a major transit hub with a customs service that actually inspects, sometimes because a specific TRAFFIC programme has been compiling that country intensively, and only sometimes because trafficking there is unusually heavy. Check the enforcement baseline and the programme footprint before you draw the obvious conclusion.

Does this source name traffickers?

Rarely. It records incidents, places, taxa and quantities. Individuals appear only where already public through prosecution, and even then inconsistently. For network and actor-level work you need investigative organisations that do that deliberately, court records, and corporate registry work – the seizure record will give you the geography, not the people.

How current is the data?

Not real time. Records are compiled and verified in batches, and the lag between an incident and its appearance varies from weeks to much longer depending on the country and the source type. If your requirement is same-week awareness of seizures, this source cannot meet it; monitor national customs announcements directly and use TRAFFIC for the normalised historical series.

Can I combine market survey data with seizure data in one trend line?

No. They are different measurements with different sampling frames. Market surveys happen when someone funded a survey, so the series moves with fieldwork scheduling. Keep them as separate series and compare their shapes rather than adding their counts.

Is the species identification reliable?

It varies from laboratory-grade to a customs officer's best guess, and the record does not tell you which. Where the species determines the legal offence or your conservation conclusion, go to the primary source and look for evidence of how the identification was made. Where it does not, aggregate at genus or family level and say that you did.

How should I handle records where enforcement itself is the concern?

Some corridors persist because officials are paid to let them, and some enforcement responses have produced serious abuses against local communities. Both are analytically relevant and neither is in the fields. Bring in governance, corruption and human-rights sources explicitly, and flag corridors where your assessment is that state complicity is a structural feature, because those require a fundamentally different response than more enforcement.

Standards, formats and interoperability

What this source speaks natively, and what it has to be translated into before a partner can consume it. Work that arrives in a recognised format is easier to defend, easier to hand over and easier to automate against:

CITES appendices I, II and III, and the permit and certificate system that determines the legality of any given movement.
IUCN Red List categories and criteria, which supply the conservation-status dimension for any taxon in the record.
Scientific binomial nomenclature, with the caveat that the record frequently holds common names and higher-rank identifications rather than binomials.
Harmonized System commodity codes, which are how wildlife products appear in customs and trade data and therefore how you join this source to trade statistics.
ISO 3166 country codes for seizure, origin, transit and destination normalisation.
UN/LOCODE for ports, airports and border crossings, which is the right key for corridor and route analysis.
STIX 2.1 and MISP as export carriers within the platform when wildlife-crime incidents need to move alongside other criminal-network reporting.

References

Primary documentation and authoritative references for this source. Publishers revise and retire material, so treat the retrieval date as part of the citation and re-check before relying on any of it in a formal product.

TRAFFIC — TRAFFIC International. The organisation's own site: report library, regional programmes and the current statement of what it does. Start here and note the publication date on anything you take.
TRAFFIC publications — TRAFFIC International. The report archive. The methodology annexes are the most valuable part and the least read; go to them before quoting any figure.
Wildlife Trade Portal — TRAFFIC International. The structured seizure and incident records with filtering and export. This is the route to rows rather than prose, and the place to read the current terms of use.
CITES — CITES Secretariat. The treaty, the appendices, the Conference of the Parties documentation and the compliance record. Prerequisite reading for classifying any incident as legal or illegal.
CITES Trade Database — CITES Secretariat and UNEP-WCMC. Party-reported legal trade records. The comparison set for illegal trade analysis and the place where laundering through permits becomes visible.
IUCN Red List of Threatened Species — IUCN. Conservation status and range for the taxa in the seizure record, which converts an incident count into an impact statement.
UNODC — United Nations Office on Drugs and Crime. The World Wildlife Crime Report series and the UN's own seizure aggregation, useful as an independent check on regional patterns.
Environmental Investigation Agency — EIA International. Undercover investigations into trafficking networks and named actors, covering the actor-level ground that TRAFFIC's incident record does not.
Wildlife Justice Commission — Wildlife Justice Commission. Network- and finance-focused investigations, valuable for understanding the criminal organisation behind a corridor rather than the corridor itself.
INTERPOL environmental crime — INTERPOL. The law-enforcement coordination view, including operations that produce seizure bursts you need to tag as collection artefacts.

Link integrity: every reference above was verified with a live request when this page was generated. Where a publisher had moved or withdrawn a document, the link was repointed at a preserved copy in the Internet Archive and marked as archived. Anything with no reachable copy anywhere had its link removed rather than left to rot — the source is still credited, it simply cannot be linked.

Put it into practice

The Quantus Intel threat intelligence platform operationalises this source: it normalises seizure and market records into geolocated incidents, builds the origin-transit-destination graph, keeps the primary source document attached to every claim, and carries the interdiction caveat through to the finished assessment.. Browse the full source catalogue, or follow any tag above into the rest of the library.