U.S. flag

An official website of the United States government, Department of Justice.

Unconventional Wisdom: Research Shakes Up Assumptions About Sex Trafficking Clues in Online Escort Ads

Scientific study finds that many widely recognized indicators of trafficking in online escort ads are likely not reliable. Research also produces new practice guide.
Date Published
May 8, 2024

Many investigators fall back on conventional wisdom when searching for indicators of sex trafficking in online escort ads. But recent research supported by the National Institute of Justice (NIJ) reveals that many assumptions are likely unwarranted.

A study by the Justice Information Resource Network [1] (JIRN) found many long-held beliefs regarding sex trafficking indicators were not valid. See Table 1 for four key examples.

Table 1 : Key examples of long-held beliefs regarding sex trafficking indicators that the study found not to be valid.
Inaccurate BeliefFinding

Ads for minor victims of sex trafficking usually list their ages between 18 and 23.


A stated age of under 23 was not a significant predictor of trafficking.


Language in ads suggesting the physical movement of victims/service providers is consistent with frequent movement to isolate and disorient victims. Examples include "limited time," "new in town," and "leaving soon.”


Movement language in an ad was 70% less likely to be associated with sex trafficking than conventional sex work.
Client screening language in an escort ad is associated with trafficking because traffickers carry out screening on behalf of victims.Client screening language was not usually a predictor of trafficking.

Photos of suspiciously young-looking individuals in escort ads indicate sex trafficking activity.



Photos of young-looking individuals were more often not a predictor of trafficking. (See sidebar “Little Trafficking Behind Ads With Young-Looking Subjects,” below.)


The researchers identified seven additional types of language often believed to be likely indicators of sex trafficking that did not correlate with trafficking.[2]

A focus group of practitioners convened as part of the study expressed surprise that factors often considered “red flags” were not substantiated by the study data as evidence of trafficking in the study data. This focus group included 11 law enforcement and prosecutor participants and two victim advocates. The group usually interpreted these indicators as signs of trafficking.

Not All Beliefs Are Wrong

The study affirmed that four key indicators are strongly predictive of sex trafficking (Table 2). The indicators of provider trustworthiness and obscured phone number surprised the practitioner and survivor focus groups, while advertising the provider’s ethnicity and language suggesting youth confirmed their prior beliefs.

Table 2 : Ad characteristics affirmed to be strong indicators of sex trafficking.
Indicator of TraffickingFinding
Ad language assuring potential clients of provider trustworthiness is more likely to represent a case of trafficking.Such language was over four times as likely to represent a trafficking case.
Obscuring the advertised phone number to avoid detection by a software program makes it more likely to be associated with trafficking.Such techniques increased the likelihood of trafficking by almost 12 times.
Offering an explicit choice of ethnicity or race is likely to be associated with trafficking.Stating the ethnicity of the individual advertised made it five times more likely to be trafficking related.
Suggestions that a service provider is under 18 make it more likely to represent a case of trafficking.Such language was four times more likely to indicate trafficking. But interpret with caution: researchers noted that nontrafficked sex workers also use this language regularly for marketing purposes.


The researchers found that the statistically significant predictors were the four identified in the table immediately above, rather than others that have been shared widely and anecdotally over the years but that are not significantly associated with trafficking.

Investigators have long used escort ads as evidence to identify potential trafficking victims. The important implication of that finding is that law enforcement can use ads much more effectively to generate leads or support investigations on trafficking while reducing false positive identifications.

Study Objectives and Design

The primary goal of the research was to help investigators and prosecutors better focus their limited resources on cases that involve trafficking rather than consensual sex work. This research adds to knowledge about how traffickers construct and use ads when they advertise a trafficking victim, as opposed to ads for non-trafficking sex work.

It had two objectives:

  1. Examine whether there are indicators that can differentiate escort ads related to sex trafficking from ads for consensual, non-trafficking sex work.
  2. Determine which indicators are most likely to predict whether the ad represents a trafficking case.

There were three main research activities:

  1. Analyze ads from closed cases across U.S. jurisdictions, supplemented by ads from web scraper archives.
  2. Interview investigators and crime analysts to learn how they use ads in trafficking investigations.
  3. Conduct focus groups with three groups of stakeholders:
    1. Sex trafficking survivors.
    2. Non-trafficked sex workers.
    3. Criminal justice and victim advocate professionals.

The researchers searched for evidence of sex trafficking in case narratives in cases investigated for trafficking or related activity, rather than relying on charging decisions, recognizing that prosecutors often pursue lesser charges to obtain a conviction, because of the high burden of proof in trafficking cases, or they may drop a trafficking charge altogether. Relatedly, police may make an arrest for prostitution only to realize later that the individual had a trafficker.

They flagged, or “coded,” the content in the police report and prosecution case file narratives that supported all elements of the federal definition of trafficking. Using the same process, they then coded ads for both trafficking cases and the cases determined not to be trafficking. The study team tested 27 potential indicators of trafficking in ads derived from their research and the literature. The final dataset included over 1,600 ads from commercial sex and massage cases in seven states, with the ads covering 35 states and one province in Canada.

Trafficking Enforcement Challenges Since 2018

In 2018, Congress enacted two laws primarily aimed at curbing online sex trafficking. The statutes, together known as FOSTA-SESTA, are the Allow States and Victims to Fight Online Sex Trafficking Act and the Stop Enabling Sex Traffickers Act. However, the research team reported limited progress since the reforms.

Although better understanding of ad trafficking indicators can help enforcement, the report also noted new challenges that have emerged since 2018, including:

The shutdown, by federal law, of the commercial website Backpage –Escort services and sex workers often used Backpage to sell their services. The law’s intent was to crack down on illicit commercial sex. But law enforcement officials told researchers that it was much easier to identify trafficking cases through Backpage than it is today. Shifts in commercial website advertising have also made trafficking victim identification much more difficult.

Moreover, Backpage and Craigslist regularly cooperated with subpoenas, but few of the remaining websites did. This results in fewer prosecutions, according to study focus groups.

Despite the fanfare after its passage, there has been “extraordinarily little” enforcement of FOSTA-SESTA, according to the U.S. Government Accountability Office.

Higher levels of street-based sex work –According to the focus groups’ practitioners and sex workers, much commercial sex solicitation moved to the street after Backpage shut down. Online advertising activity, however, has rebounded somewhat among more diverse, smaller sites.

Fewer prosecutions –Law enforcement has also received less cooperation for subpoenas, especially from websites hosted outside the jurisdiction of the United States due to the passage of FOSTA-SESTA. This has resulted in fewer trafficking prosecutions, according to focus group discussions.

Implications and Recommendations

By themselves, sex ads and indicators in ads are not sufficient evidence to determine if trafficking is occurring. Research on indicators can help increase lead generation, but law enforcement must investigate and find corroborating evidence to establish trafficking.

The researchers concluded that, while this study represents foundational work in the study of ad indicators in escort ads, these results require replication. They recommended that future research utilize a larger dataset to confirm these results.

Researchers suggested that incorporating the study findings in law enforcement activities might uncover more victims.

About This Article

The research described in this article was funded by NIJ award 2017-MU-CX-0005, awarded to the Justice Research and Statistics Association (now known as the Justice Information Resource Network). This article is based on the grantee report “Indicators of Sex Trafficking on Online Escort Ads” (2022), by Kristina Lugo-Graulich, Leah Meyer, Karen Souza; Susannah Tapp, Bailey Maryfield, and Lindsay Bostwick (2022).

Law enforcement often searches for trafficking operations when a suspiciously young-looking person appears on a web ad for escort services; however, sting operations that target activity tied to young-looking escort ad images are often unproductive, the study report noted. Researcher analysis of escort ads found that photos of suspiciously young-looking persons are three to four times more likely to be in a non-trafficking ad, or an ad of unknown origin, than in an ad for a trafficking operation.

Return to text.

Date Published: May 8, 2024