Polaris Project / Trafficking Hotline: Intelligence Source Guide

Polaris runs the US National Human Trafficking Hotline and publishes the typologies derived from it: the most detailed open account of how trafficking is organised by industry and venue. It is a record of who called a phone line, not a prevalence estimate.

At a glance

Source	Polaris Project / Trafficking Hotline
Category	Conflict, Crime & Human Security › Human Trafficking & Child Protection
Homepage	https://polarisproject.org/
Format	`HTML`
Access	Open — no account required
Disciplines	Human Intelligence
Mission domains	Human Trafficking

US hotline statistics and typologies. — as catalogued in the platform’s own source registry.

Polaris is a US non-profit that has operated the National Human Trafficking Hotline under federal grant funding, alongside a research and data programme built on what the hotline hears. The hotline is a confidential, multilingual, round-the-clock service reachable by phone and by online channels; it takes contacts from people experiencing trafficking, from family members and community members, from service providers, and from law enforcement. Advocates on the line do crisis intervention, safety planning and referral to local services, and they connect callers to law enforcement only where the person wants that and where the situation permits. From that operational work Polaris derives an anonymised case dataset, and from the case dataset it publishes: annual and periodic statistical reports on signals and cases, breakdowns by trafficking type, venue and industry, victim demographics and recruitment method, and a body of typology research that maps how trafficking operations are structured in the United States. The best known of that research classifies trafficking into distinct business models — illicit massage businesses, agriculture, domestic work, travelling sales crews, restaurant and food service, health and beauty, construction, escort services and others — each with its own recruitment channels, control mechanisms, financial footprint and intervention points. Polaris also maintains a directory of anti-trafficking organisations worldwide and has published applied guidance for sectors, including financial institutions, that need to recognise the pattern in their own data.

The typology work is the reason this source exists in an intelligence catalogue. Almost everything else published on trafficking is either legal — treaty text, statute, tier ratings — or estimative, and neither tells you what a trafficking operation looks like from the outside. Polaris's material does. It describes, industry by industry, how people are recruited, what the control mechanism is, where the money moves, what the visible business footprint is, and which third parties — landlords, banks, staffing agencies, transport operators, licensing authorities — are structurally in a position to notice. That converts an abstract crime type into a set of observable indicators you can look for in data you already hold, which is precisely what a corporate compliance function, a financial intelligence unit, a licensing regulator or a supply-chain auditor needs and cannot get from an international report. For HUMINT work it does something else: it maps the referral landscape, so that when a case surfaces you know which channel serves the person's actual interest rather than your investigative interest. The dataset also serves as a longitudinal signal on public awareness, because the volume and composition of contacts responds visibly to media events, campaigns and viral misinformation — a property that is a nuisance for prevalence estimation and a genuinely useful measurement of information-environment effects.

Who publishes it, and why that matters

Polaris is a grant-funded non-profit, and the hotline it has operated is a federally funded programme awarded by a US government office through a competitive process. This is the single most important structural fact about the source. The operator of a federal hotline can change when a grant is recompeted, and with a change of operator come changes to intake protocols, case definitions, data retention and publication practice — any of which can break a time series without anyone announcing that it has been broken. Confirm who currently operates the National Human Trafficking Hotline and under what arrangement before treating recent figures as continuous with older ones. Beyond that, Polaris's incentives are worth understanding because they shape the data: it is a victim-services and advocacy organisation, not a research institute, and its primary obligation is to the people who call. That obligation is why the data is anonymised, why law-enforcement referral is consent-based rather than automatic, and why Polaris has publicly and repeatedly declined to let its numbers be used as prevalence estimates or as state rankings — positions that have cost it easy publicity and that are correct. Take the refusal seriously rather than working around it.

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
`signal_count`	int	The number of contacts received — calls, texts, webforms, chats. A signal is an interaction, not a case and not a victim, and one situation can generate many signals from many people over months.	Awareness-effect analysis; never a prevalence input.
`case_count`	int	Situations of potential trafficking identified from signals, after an advocate's assessment. These are not verified, not adjudicated and not law-enforcement determinations; they are an operational classification made to route help.	Typology and venue analysis; comparison with prosecution statistics as a gap measure, never as corroboration.
`trafficking_type`	enum	Sex trafficking, labour trafficking, both, or not specified. The distribution is heavily shaped by which forms the public has been trained to recognise, and labour trafficking is systematically under-represented relative to any credible estimate of its share.	Sector targeting; identification of awareness gaps rather than of actual composition.
`venue_industry`	enum	The setting in which the trafficking is reported to occur — illicit massage business, agriculture, domestic work, travelling sales crew, restaurant and food service, hospitality, construction, health and beauty, escort services and others. This is the most operationally useful field in the dataset.	Sector risk assessment, licensing and inspection targeting, supply-chain and procurement due diligence, financial-institution typologies.
`recruitment_method`	enum	How the person was brought into the situation: false job offer, familial, intimate partner, posted advertisement, recruitment agency, smuggling debt and related categories. Recruitment is the intervention point where prevention actually works.	Prevention-campaign design, job-advertisement monitoring, recruitment-agency due diligence.
`victim_demographics`	enum	Aggregated age band, gender and citizenship or immigration status categories. Published only at a level that cannot identify anyone, and deliberately coarse for that reason.	Service-provision planning and cohort-level prevention; never individual inference.
`contact_channel`	enum	Phone, text, webform or chat. Channel mix shifts with the demographics of who is reaching out and with which channels are promoted, and a channel change alone can move apparent volume substantially.	Accessibility analysis; explaining discontinuities in the series.
`contact_role`	enum	Whether the person contacting was directly experiencing trafficking, a family or community member, a service provider, or law enforcement. Self-reports and third-party reports differ enormously in reliability and in what they contain.	Weighting and quality assessment; separating first-hand from second-hand information before any analysis.
`geography`	string	State or region associated with a case, where identifiable. Polaris has consistently warned against ranking jurisdictions on this field because it tracks population, awareness, service availability and language access rather than incidence.	Service-coverage gap analysis; never a state league table.
`referral_outcome`	enum	What the contact resulted in — referral to a service provider, safety planning, information provision, or connection to law enforcement where the person consented. The consent condition is the defining feature of the system.	Service-capacity analysis; evaluating whether a referral network exists where the need is.
`reporting_period`	string	The period a published aggregate covers. Polaris has changed publication format and periodicity more than once, and figures from different formats are not automatically comparable.	Time-series construction with explicit break markers at each format change.
`typology_class`	enum	The business-model classification from Polaris's typology research, describing how an operation is structured rather than what offence it constitutes. This is the field that translates into observable indicators.	Indicator development for compliance, licensing, financial monitoring and audit workflows.

Coverage — and what is not in it

Coverage is United States and United States territories, in the sense that the hotline serves people located there and the case data reflects situations with a US nexus — though the people involved frequently have origins, recruitment and debt relationships in other countries, and the data captures that international dimension in the recruitment and citizenship fields. Language access is broad, delivered through interpretation, and this matters more than it sounds: coverage of a population is effectively a function of whether the hotline can be used in that population's language and whether the population has been reached by outreach. Temporally the hotline runs continuously and the case data accumulates continuously, but publication is periodic and the format has changed over the years, from detailed annual statistical reports to other presentations. Industry coverage is where the source is strongest — the venue and typology breakdowns reach into sectors that no official dataset describes at this resolution, particularly domestic work, agriculture, illicit massage businesses and travelling sales operations. The weakest coverage is of populations that cannot safely make a call at all: people under close physical control, people without a phone, people whose immigration status makes any institutional contact feel unsafe, and people in situations that neither they nor those around them recognise as trafficking.

Known blind spots

Absence of evidence here is not evidence of absence. These are the conditions under which Polaris Project / Trafficking Hotline will not show you something that is nevertheless real:

The dataset only contains situations someone chose to report to a hotline. Populations under the tightest control are least able to call, which means the most severe cases are the least represented — the exact inverse of what an analyst instinctively assumes.
Labour trafficking is systematically under-represented relative to sex trafficking, because public awareness campaigns, media coverage and training have concentrated on the latter, and because labour trafficking victims often do not identify what is happening to them as a crime.
Volume responds strongly to media events. A high-profile case, a documentary, a campaign or a viral social-media conspiracy can produce a surge of contacts that has nothing to do with any change in trafficking, and at least one such surge has measurably degraded the service's ability to help real callers.
Geographic distribution tracks population, awareness spending, service availability and language access. A state with more cases very often has better outreach and more service providers, not more trafficking, and Polaris says so explicitly.
Cases are not verified. They are operational assessments made quickly, by an advocate, on partial information, for the purpose of routing help — a completely different standard from an investigative or evidential determination.
Consent governs law-enforcement referral, so cases where the person did not want police involvement are in the data but absent from every criminal-justice dataset. Comparing hotline cases with prosecutions measures that gap, not accuracy.
Anything below the aggregate is unavailable and will stay unavailable. There is no route to case-level data, and any analysis requiring it is not going to happen through this source.
The series is fragile at its administrative joints: grant recompetition, operator change, intake protocol revision or a change in publication format can break comparability without an obvious marker in the data.
Situations occurring entirely outside US jurisdiction are out of scope, even where a US-linked supply chain, recruiter or purchaser is involved, so the international dimension is visible only through the fragment that reaches a US contact.

Write the blind spot into the product. A statement that something “was not observed in Polaris Project / Trafficking Hotline” 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

Everything usable is published on the web as reports and statistical summaries: Polaris's own site carries the research and data publications, and the hotline site carries service information and referral routes. There is no API, no bulk dataset and no case-level access, and requests for the last of these are declined for reasons that are about the safety of the people in the data rather than about institutional caution. Researchers seeking anything beyond the published layer should approach Polaris directly with a specific question and an ethics framework, and should expect the answer to involve Polaris producing an aggregate rather than releasing records. The one access route that is open to everyone and that you should have ready is the hotline itself — for referral, not for enquiry. Never call an emergency service to ask a research question; it occupies a line that someone else needs.

Licence

Polaris's reports and typology publications are copyrighted material published for public benefit and are freely readable and quotable with attribution. Reproducing whole documents, or reusing the underlying figures as if they were an open dataset you assembled, is not appropriate. The stronger constraint is not legal but conditional: Polaris publishes with an explicit statement of what its numbers may and may not be used to claim, and reusing them in a way it has publicly disclaimed — as prevalence estimates or as jurisdictional rankings — is a misuse regardless of copyright status. If you are building a commercial product on this material, ask, and expect a conversation about characterisation rather than about fees. Confirm current terms and the current publication set directly, since both the format and the operator arrangements have changed over time.

Rate limits and fair use

There is no programmatic interface, so the etiquette is about restraint rather than throughput. Fetch published reports once and cache them; do not crawl the hotline's service pages, which exist to help people in crisis and should not be competing with a scraper for server capacity. If you monitor for new publications, weekly is ample. The one hard rule specific to this source: the hotline number is not a data endpoint. Do not call it to test a workflow, to verify a case, to ask about statistics, or to demonstrate a capability. Every occupied line is a line a person in danger cannot reach.

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 Polaris Project / Trafficking Hotline 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
Statistical and data reports	HTML	periodic, with format changes over time	The core aggregate layer: signals, cases, types, venues, demographics. Extract into your own schema and record which publication format each figure came from.
Typology research	HTML	irregular	The business-model classification work. This is documentary rather than numeric, and it is the part with the longest useful life and the greatest operational value.
Sector guidance publications	HTML	irregular	Applied indicator material aimed at specific industries, including financial services. Read these as ready-made detection logic rather than as background.
Organisation directory	HTML	continuous	A directory of anti-trafficking organisations, useful for referral routing and for mapping which geographies have any service capacity at all.
Hotline referral route	HTML	n/a	An outbound action. Record the number and the online channels in your contacts, and never treat them as a collection interface.

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 as a periodic document source — Add Polaris in `sources.php` with a periodic cadence, so `collect.php` treats missing updates as normal rather than as a collection failure — publication here is irregular by nature.
Load aggregates with their period and format — Use `import.php` to store each published figure with its reporting period and the publication format it came from. Format changes are the main source of false trends in this series and the only defence is recording them at ingest.
Codify the typologies as indicator sets — Convert the venue and business-model descriptions into structured indicator lists in `playbooks.php` — the observable business footprint, the third parties in a position to notice, the financial pattern. This is the step that makes the source operational rather than merely informative.
Attach constraint labels to geography — Flag the state and region fields at schema level with the explicit note that they are not incidence measures, so that anything drawing a map from them is forced to render the caveat alongside it.
Wire referral, not investigation — Register the hotline and the directory in `le-contacts.php` and `agencies.php`, with a decision note distinguishing emergency situations, consent-based referral, and situations where contacting anyone could endanger the person.
Route typologies into the mission area — Surface the industry indicator sets on `human-trafficking.php` and `vulnerable-populations.php` where they can be applied to case work, rather than leaving them as reading material in a library.
Link to financial-crime workflows — Where the typology describes a money movement pattern, connect it to `financial-crime.php` so that compliance-side users encounter it in the workflow that can act on it.
Prohibit person-level records — Configure ingest so this source cannot create person, phone or location entities. Everything Polaris publishes is aggregate or documentary, and any record that looks like a person from this source is a bug.

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.

As a record of contacts, the data is careful and honestly presented. Polaris's own methodological statements are unusually direct about limitations, and it has declined public opportunities to let its figures be inflated into prevalence claims — which is a strong signal about internal data culture. The case classifications are made by trained advocates under operational conditions, which makes them good enough for their purpose and unsuitable for evidential use, and Polaris says so. The typology research is the highest-quality component: it draws on a large body of cases, it is corroborated against sector knowledge, and its descriptive claims about how operations are structured have held up well and have been adopted by financial institutions and regulators. Where quality is genuinely uncertain is comparability over time, because of format changes and the possibility of operator change under grant recompetition. Judge this source on the internal consistency of the typologies, which is high, rather than on the stability of the counts, which is not the property it was built to have.

Characteristic false positives

Reporting hotline case counts as the number of trafficking victims in a state or in the country. They are the number of situations someone reported to one phone line, and the ratio between the two is unknown and probably very large.
Ranking jurisdictions by case volume. This produces a league table of outreach spending, population and service availability, and Polaris has explicitly asked people not to do it.
Reading the sex-versus-labour trafficking split as the true composition of the problem. It is the composition of what the public has been trained to recognise and feels able to report, and it diverges sharply from every credible global estimate.
Treating a case as a verified allegation. It is an advocate's rapid operational assessment for the purpose of getting someone help, made without investigation and often on second-hand information.
Attributing a spike to a real-world increase. Spikes track media events, campaigns and viral misinformation, and one notorious online conspiracy produced a contact surge that impeded the service's actual work.
Comparing hotline cases with prosecutions and calling the difference a measure of under-enforcement. Much of the difference is people who deliberately did not want law enforcement involved, which is a designed feature of a consent-based system rather than a failure of one.
Inferring anything about an individual, a business or an address from aggregate venue categories. The published data cannot support that and the attempt endangers people.
Assuming the series is continuous across years. Publication formats have changed and the operating arrangement is grant-dependent, so a break can occur for administrative reasons with no announcement in the numbers themselves.

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

The two components age at completely different rates and should be handled separately. The statistical aggregates are period-bound facts that never change, but their interpretive value decays within a year or two because the awareness environment that generated them moves. The typology research ages far more slowly: business models in this space are durable, and a description of how recruitment and control work in domestic service or agricultural labour remains substantially accurate for years. What does shift within the typologies is the technology layer — how recruitment advertisements are placed, how payments move, which platforms are used — and that layer should be refreshed against current casework even when the structural description still holds. The referral information ages fastest of all and matters most: a service provider that has closed, a number that has changed, or an operator that has been replaced turns a referral into a dead end at the moment someone needs it. Verify contact routes on a schedule, not on demand. A stale record from this source looks like a case-count chart with no reporting period, or a referral entry nobody has dialled in two 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

Human Intelligence

Serves these mission domains

Human Trafficking

Yields these data points

How each sector uses Polaris Project / Trafficking Hotline

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

The relevant exposure is contractual and installation-level rather than operational. Base support services, construction, catering, cleaning and logistics are contracted through chains that reach into exactly the sectors Polaris's typologies describe, and recruitment-fee debt bondage among third-country nationals working on facilities is a well-documented pattern with its own compliance regime. Use the typologies as an audit lens on your own contractor base and on the labour recruitment practices beneath it. The hotline is also the referral route for anything surfacing in a US-jurisdiction facility, and it should be posted somewhere people can actually see it.

🕵 National intelligence

This is not a collection source and should not be treated as one; its analytical contribution is structural understanding. The typologies explain how trafficking organisations finance themselves, where they interface with legitimate businesses, and which control mechanisms they depend on, which is exactly the knowledge needed to recognise the pattern in financial, corporate or travel data you hold for other reasons. The awareness-sensitivity of the contact volume is separately interesting as a measurement of how information campaigns and viral narratives propagate through a population, since the hotline records the response with a timestamp.

👮 Law enforcement

The hotline is a partner and a referral channel, not a source of leads you can query. What it provides investigators is the typology knowledge — recruitment routes, control mechanisms, financial footprints and the third parties positioned to observe — which shortens the distance between a suspicion and a viable investigative theory. Understand the consent architecture before engaging: the hotline connects people to law enforcement when they want that, and pressing for information the person did not consent to share damages the channel for everyone. The gap between case counts and prosecutions is a resourcing argument, not evidence of anything about a particular case.

🔍 Private investigation and corporate security

Corporate and supply-chain investigators get the most direct value here. The venue typologies convert into an audit programme: which labour intermediaries to examine, which payment patterns to look for, which licensing and inspection records to pull, which accommodation arrangements signal control. In domestic and family matters, the discipline is different and stricter — if an engagement surfaces indicators of trafficking, the person's safety governs, the hotline or law enforcement is the destination, and no client instruction justifies approaching or surveilling a suspected victim.

📰 Journalism and OSINT media

The typology research is the strongest journalistic asset in this catalogue for trafficking coverage, because it lets you write about a mechanism rather than about an anecdote. The numbers are where coverage goes wrong: hotline totals are not prevalence, state comparisons are not rankings, and a spike after a viral claim is a story about misinformation rather than about trafficking. Polaris will say all of this on the record if asked, which is a better use of an interview than getting a number confirmed. Coverage that inflates the figures has directly contributed to hoax-driven surges that damaged the service.

🌍 NGO, humanitarian and human rights

For service providers and advocacy organisations this is both a peer resource and a referral backbone: the directory maps who exists where, the typologies inform case identification and training, and the statistical reports support funding applications when used honestly. The most valuable single habit is to adopt Polaris's own framing discipline in your own materials — refusing to present contact data as prevalence, and refusing to rank jurisdictions — because the sector's credibility with policymakers depends on it and is repeatedly damaged by organisations that do the opposite.

🎓 University and research

Treat the published aggregates as data about a help-seeking process, and the typologies as qualitative findings with a large empirical base. Both are legitimate objects of study; neither is an incidence measure and the literature has been clear about this for years. There is no case-level access and there should not be, so research designs requiring it are non-starters. The most productive work has used the typologies as hypothesis generators for analysis of other datasets — labour inspection records, licensing data, financial reporting — where the underlying population is better defined.

Playbook: working Polaris Project / Trafficking Hotline 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 — Separate the two products before you start

The statistical aggregates and the typology research answer different questions and have different reliability. Decide which one your question needs. Questions about how trafficking is organised go to the typologies; questions about who is asking for help go to the aggregates; questions about how much trafficking exists go to neither.

Phase 2 — Establish the administrative timeline

Record when publication formats changed, when the operating arrangement changed, and when intake or classification protocols were revised. Do this before pulling a single number, because these are the joints at which the series breaks and none of them are visible in the counts themselves.

Phase 3 — Read the methodology statement and adopt its limits

Polaris publishes explicit statements about what its data can and cannot support. Treat those as binding constraints on your own output rather than as boilerplate. The organisation closest to the data has told you what the data means; overriding that requires evidence you do not have.

Phase 4 — Extract the venue typologies into indicator sets

For each industry — domestic work, agriculture, illicit massage, travelling sales, hospitality, construction — write out the observable footprint: business registration pattern, staffing intermediary, accommodation arrangement, payment flow, licensing touchpoint. This is the artefact you will actually use, and it belongs in `playbooks.php` where casework can reach it.

Phase 5 — Map the third parties who are positioned to see

For each typology, list the institutions that necessarily interact with the operation: banks, landlords, licensing boards, transport providers, staffing agencies, health services. These are your realistic detection points and your realistic intervention partners, and identifying them is more useful than any amount of victim-side analysis.

Phase 6 — Test the indicators against data you already hold

Apply the financial and corporate patterns to your own records — transaction monitoring, supplier registers, contractor rosters — rather than trying to find new data. The typologies were built to be recognisable in ordinary business records, which is what makes them valuable to a compliance or audit function.

Phase 7 — Separate first-hand from third-hand signals

When using the aggregates, always split by who was contacting. Self-reports carry different information, different reliability and different urgency from community reports and provider reports, and a combined figure conceals all of it.

Phase 8 — Overlay the awareness timeline on any volume series

Plot known campaigns, major media events and viral misinformation episodes against contact volume. Most of the variance will be explained. What remains after that overlay is the only part of the series worth interpreting, and it is usually much smaller than the raw movement suggests.

Phase 9 — Build the referral decision tree before you need it

Write down what your organisation does when an indicator appears: when it is an emergency for police, when it is a consent-based hotline referral, when contacting anyone would endanger the person, and who in your organisation makes that call. Put it in `playbook-library.php` and rehearse it. Improvised responses in this domain get people hurt.

Phase 10 — Verify the referral routes on a schedule

Dial and check the contacts you have registered, at least annually. Providers close, numbers change and operating arrangements shift. A referral list that has not been verified is worse than no list, because it produces confident wrong action at the worst moment.

Phase 11 — Cross-read against supply-chain and enforcement sources

Set the typologies alongside forced-labour goods listings, import-enforcement actions and international tier assessments in `correlate.php`. The typologies tell you the mechanism, the trade sources tell you the commodity and the country, and only together do they support a targeted due-diligence programme.

Phase 12 — Write the constraint into every output

Any product citing this source states, in the body, that hotline data measures contacts rather than incidence and that geographic distribution reflects awareness and service availability. This is not a disclaimer; it is the finding. Products that omit it have contributed to real operational harm at the hotline.

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
National Human Trafficking Hotline	prerequisite	The service itself and its referral routes. Belongs in your contacts before any analysis begins, and should be verified periodically.
US State Department TIP Report	extends	Country-level policy assessment and tier ratings, which supply the international and governmental frame that hotline data cannot.
US DOL List of Goods	extends	Country and commodity pairs for goods produced with child or forced labour, which connects the typologies to specific supply chains.
UNODC Global Report on Trafficking in Persons	corroborates	International detection and prosecution statistics compiled from national reporting, useful for testing whether the US composition looks unusual.
Global Modern Slavery Directory	extends	A directory of anti-trafficking organisations worldwide, which is how a referral becomes possible outside the jurisdictions you know.
ILO forced labour work	corroborates	International labour standards and estimates that provide the independent basis for judging how badly labour trafficking is under-represented in hotline data.
Walk Free Global Slavery Index	contradicts	An estimative prevalence product with a very different methodology and its own contested assumptions. Reading it alongside hotline data clarifies what each can and cannot claim.
US Office on Trafficking in Persons	prerequisite	The federal office administering anti-trafficking services and the funding arrangements under which the hotline operates.

Legal, ethical and operational constraints

The published material carries no unusual legal restriction and can be read, cited and analysed anywhere. The constraints that matter here are about people rather than about data. Trafficking cases involve individuals whose safety depends on information not moving, and in most jurisdictions there are victim-confidentiality protections, immigration-relief processes that can be jeopardised by disclosure, and privilege attaching to victim-advocate communications. Do not attempt to identify, locate, contact or surveil a suspected victim; do not disclose that someone may be a trafficking victim to an employer, a landlord, a family member or a foreign consulate; and do not assume that involving law enforcement is automatically in the person's interest, because for people with irregular immigration status it frequently is not. Mandatory reporting duties vary and often turn on whether a minor is involved — know the rule in your jurisdiction before you are in the situation. If you are a corporate or supply-chain investigator, your findings will engage disclosure and remediation obligations under an expanding set of due-diligence regimes, and those are questions for counsel. The hotline's consent-based design is a legal and ethical architecture, not an obstacle to work around.

Operational security

Reading Polaris's publications reveals nothing. The exposure in this domain is not about your own network footprint but about the people you might expose. Searching for a business, an address or a person in connection with suspected trafficking generates records at the search provider, at the platforms you touch and at any subscription service you use, and in a small community that pattern is discoverable by exactly the people who should not discover it. Traffickers monitor. If your enquiry becomes visible, the immediate consequence is usually that a person is moved, isolated further or harmed. Conduct sensitive enquiries through channels that do not touch the subject's environment, do not contact the business, do not visit, and route anything actionable to the hotline or to law enforcement rather than pursuing it yourself. Internally, records that a case involved suspected trafficking are among the most sensitive data your organisation will hold and should be compartmented accordingly.

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 Polaris Project / Trafficking Hotline is contributing anything, and they are worth baselining now so the answer is available later.

Whether every published output citing this source states, in the body text, that the data measures contacts rather than incidence.
Whether your referral contact list has been dialled and verified within the last twelve months, with the date recorded.
The number of typology indicator sets that have actually been applied to internal data, as opposed to filed as reading material.
Whether applying those indicators has produced any case that a conventional screen missed, which is the only real test of whether the typologies are earning their place.
How many administrative break points — format changes, operator changes, protocol revisions — are recorded against your time series.
Whether an analyst has ever called the hotline for a non-referral purpose, which should be zero and should be treated as a serious process failure if it is not.
Whether your organisation has a written, rehearsed decision rule for what happens when a trafficking indicator surfaces, and whether it was followed the last time one did.

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:

The most severe cases are the least likely to appear. Control that prevents a person from calling is control that keeps them out of the dataset, so the data is biased against exactly the situations you most want to see.
The typologies are the product; the counts are context. Analysts who invert this end up arguing about numbers that were never designed to bear the weight, and miss the material that would actually change their casework.
Every jurisdictional comparison in this dataset is a comparison of outreach budgets. If you find yourself drawing a choropleth, stop and ask what you would expect the map to look like if trafficking were uniformly distributed — it would look like this one.
Consent is not a formality. A referral made without the person's agreement can trigger immigration consequences, retaliation or flight, and the hotline's design reflects hard experience about which interventions actually help.
A surge in contacts after a viral claim is a measurable harm, not a measurement of trafficking. The service has limited capacity, and hoax-driven volume displaces real callers, which is a reason for analysts and journalists to be careful about amplification.
Labour trafficking hides inside legitimate business structures with real payroll, real invoices and real licences. The typologies are useful precisely because they describe what that looks like in ordinary records, which is where you will find it if you find it at all.
Recruitment is the intervention point. Control mechanisms are hard to disrupt from outside; recruitment channels — job advertisements, agencies, community networks — are visible, documented and regulable, and that is where the typologies point.
Never treat a hotline case as corroboration of anything. It is an operational assessment made in minutes to route help, and using it as evidence misrepresents both the advocate and the person who called.
Check who currently operates the hotline before citing a multi-year trend. A grant recompetition can change definitions, intake and publication in a way that no footnote in the data will tell you about.

Questions analysts actually ask

Can I use hotline case counts to estimate how many trafficking victims there are?

No, and Polaris says so directly. The counts measure situations reported to one phone line by people able and willing to call. The relationship between that and prevalence is unknown, unstable and almost certainly very large. If you need an estimate, use an estimative source with a stated methodology and cite its uncertainty range rather than borrowing a hard-looking number from an operational dataset.

Which US states have the worst trafficking problem according to this data?

The data cannot answer that. Case volume tracks population, outreach spending, service availability and language access, so states with more cases usually have better anti-trafficking infrastructure. Polaris has publicly asked people to stop constructing these rankings, and every one that circulates is a rediscovery of the same artefact.

Why does the data show more sex trafficking than labour trafficking when global estimates show the reverse?

Because it is a help-seeking dataset. Public awareness campaigns, media coverage and training in the US have concentrated overwhelmingly on sex trafficking, so people recognise and report it; labour trafficking victims often do not identify their situation as a crime, and third parties rarely recognise it either. The split measures recognition, not composition.

Can I get case-level data for research?

No. Case records concern identifiable people in ongoing danger and are not released, and a research design requiring them will not proceed through this source. Approach Polaris with a specific question and expect them to consider producing a bespoke aggregate. That constraint is a feature of a victim-centred system, not an institutional obstacle.

Should I call the hotline to check whether a situation I found is trafficking?

Call it if a person may need help and you are prepared to be routed to services or law enforcement as appropriate. Do not call it to validate an analytical hypothesis, to test an integration, or to ask about statistics. It is a crisis line with finite capacity, and an occupied line has a direct human cost.

How do I use the typologies if I work in compliance rather than investigation?

Read them as detection logic for records you already have. Each business model implies a payment pattern, a corporate structure, a labour intermediary and a licensing touchpoint. Turn those into screening rules against your supplier register, transaction monitoring or contractor roster. This is the use case the sector guidance was written for and it is where the source pays off fastest.

A contact spike appeared in the data after a viral online claim. How should I treat it?

As an information-environment event, and as an operational harm. It reflects people reporting things they saw described online, and the volume displaces real callers on a capacity-limited service. It belongs in an analysis of misinformation propagation, not in a trafficking trend line, and it is a reason for caution about amplifying unverified claims.

How does this compare with the State Department's TIP Report?

They are complementary and not comparable. The TIP Report assesses what governments do, at country level, on an annual policy cycle. Polaris describes how operations are structured inside the US, from a service perspective. Use TIP for the diplomatic and legal frame and Polaris for mechanism, and do not expect the two to validate each other.

Is the hotline data continuous across years?

Not reliably. Publication formats have changed, and the hotline is a federally funded programme subject to grant competition, so intake protocols, case definitions and publication practice can change with an administrative decision. Confirm the current operating arrangement before presenting a multi-year trend as a single series.

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:

The Palermo Protocol definition of trafficking in persons, which supplies the act-means-purpose structure underlying every classification in this space.
US Trafficking Victims Protection Act framing and its prosecution, protection and prevention paradigm, which shapes how US cases are categorised.
Polaris's own typology of trafficking business models, which functions as a de facto sectoral taxonomy adopted well beyond the organisation itself.
Financial-sector suspicious activity reporting typologies for human trafficking, which translate the business models into transaction-monitoring rules.
ILO forced labour indicators, which provide the labour-side vocabulary that hotline categories map onto imperfectly.
Victim-centred and trauma-informed practice standards, which govern the consent architecture and constrain what may be done with any information.
Anonymisation and small-cell suppression conventions, which are why the published demographic breakdowns are as coarse as they are.

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.

Polaris — Polaris. The organisation, its programmes and the current state of its research and data publications.
Data and research — Polaris. The published statistical and analytical layer, together with the methodological statements that constrain how it may be used.
The Typology of Modern Slavery — Polaris. The business-model classification that turns trafficking from an abstract offence into a set of observable indicators.
National Human Trafficking Hotline — National Human Trafficking Hotline. The service itself: referral routes, channels and the information a person in danger needs. Belongs in your contacts list.
Office on Trafficking in Persons — US Department of Health and Human Services. The federal office administering anti-trafficking services and the funding under which the hotline has operated.
Global Modern Slavery Directory — Polaris. Directory of anti-trafficking organisations worldwide, and the practical basis for referral outside your own jurisdiction.
Global Report on Trafficking in Persons — UNODC. International detection and prosecution data compiled from national reporting; the comparative frame for judging US composition.
Forced labour, modern slavery and human trafficking — International Labour Organization. The labour-standards vocabulary and the international estimates against which hotline composition should be read.
Trafficking in Persons Report — US Department of State. The governmental and diplomatic layer that hotline data cannot provide.
Global Slavery Index — Walk Free. An estimative prevalence product with contested assumptions; useful mainly for understanding what a prevalence claim requires and why hotline data is not one.

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 turns Polaris's business-model typologies into applied indicator sets in `playbooks.php` and `financial-crime.php`, stores the contact aggregates as period-bound statistics that cannot be rendered as a map without their constraint label, and keeps verified referral routes in `le-contacts.php` — creating no person or phone records from this source at all.. Browse the full source catalogue, or follow any tag above into the rest of the library.