Facebook Marketplace statistics, with the definitions that matter
Quick answer
Meta reported 430 million items and 44 million vehicles listed monthly on Marketplace globally in July 2026. These are listing figures, not counts of buyers or completed sales. For fraud context, government reports and surveys measure broader categories; they cannot establish the percentage of Marketplace listings that are scams.
Download the statistics source ledger (CSV). It contains ten records with units, scope, dates, definitions, limitations and original links. Sources checked on 17 October 2026. No figures below come from a Sift survey or a test of marketplace transactions.
Current Marketplace figures
| Figure | Population and period | Interpretation |
|---|---|---|
| 430 million items monthly | Marketplace globally; reported July 2026 | Listings, not unique buyers or completed transactions |
| 44 million vehicles monthly | Marketplace globally; same announcement | Do not add to the item figure: overlap is unspecified |
| One in three | Young adults on Facebook in the US using Marketplace daily | A restricted population, not all US adults or global users |
Source: Meta’s July 2026 announcement. Meta does not provide detailed counting rules in that announcement. In particular, it does not explain relistings, the boundary between items and vehicles, or the age range behind “young adults.” Keep those limits attached to the numbers rather than filling them with assumptions.
Historical figures need historical labels
Marketplace’s launch announcement is dated 3 October 2016. The more-than-450-million monthly figure in that article refers to visitors to buy-and-sell groups. It describes activity that preceded Marketplace, and belongs in a history section rather than a current Marketplace user chart. Meta’s launch announcement.
A number can be accurately copied and still be misleading. “Users,” “monthly visitors,” “listings” and “daily active people” are different measures. An article that combines them into a single growth line implies continuity that the source data does not establish. If you need a global active-user figure, seek a release that explicitly defines that measure and period.
Fraud context: three different kinds of evidence
| Measure | Result | Scope and source |
|---|---|---|
| Reported scam losses originating on social media | US$2.1 billion | 2025 FTC Consumer Sentinel reports; all social platforms |
| Merchandise fraud reports / victims / losses | 2,596 / 2,222 / CA$11.7 million | Canada, 2025, CAFC merchandise category |
| Adults who ever experienced an unrefunded counterfeit or undelivered online purchase | 36% | US adults in Pew’s April 2025 survey |
| Same survey experience within the past year | 12% | Prior 12 months relative to the survey, not calendar-year 2025 |
The FTC analysis concerns social media as the initial contact method. It includes fraud beyond shopping and acknowledges a collection gap during the 2025 government shutdown. It cannot attribute the entire total to Facebook Marketplace.
The CAFC table distinguishes reports, victims and reported money lost. Keep the Canadian dollar currency explicit. Its merchandise category is broader than any single platform; the three columns are not interchangeable.
Pew surveyed 9,397 US adults from 14–20 April 2025. These are estimates from respondents’ experiences, not a count of police reports. “Ever” and “past year” describe different time windows. The combined counterfeit-or-non-delivery measure includes overlapping responses, so adding its component percentages would double count some people.
A citation worksheet
Use this six-part sentence before putting any number in a chart:
Publisher reported measure and unit, covering population/geography, during measurement period, using definition, with limitation.
For example: “The CAFC recorded 2,596 merchandise fraud reports in Canada for 2025; the category is not specific to Marketplace.” This tells a reader what the number supports without implying a risk probability.
| Proposed calculation | Missing information | Better use |
|---|---|---|
| Social-media losses divided by Marketplace users | Matching platform, population and period | Discuss aggregate reported harm |
| Monthly listings multiplied by twelve | Deduplication and turnover rules | Quote the monthly figure as published |
| Survey percentage multiplied by online purchases | Person-based survey versus transaction denominator | Describe adults’ reported experiences |
| Canadian losses plus US losses | Common currency, collection method and scope | Present separate country rows |
| Ten local listings turned into a citywide price average | Representative sampling and confirmed sales | Use them as a clearly labelled personal comparison |
These are methodological checks, not claims that a small sample is useless. A buyer can learn from a few close comparables. The problem starts when a convenience sample is described as an entire market.
What these figures help a buyer do
Large listing totals do not tell you whether the exact model you want is available within a practical pickup distance. Broad fraud totals do not authenticate a particular seller. Use the statistics to understand the scale and limits of the evidence, then make the purchase decision with item-level information.
Record model, condition, included accessories, asking price, location and date for each plausible listing. Keep sale prices distinct from asking prices. An item disappearing from search is not proof that it sold for the displayed amount. Our fair-price guide gives a way to build that comparison, and the scam guide covers warning signs during the conversation.
How this source ledger was prepared
We selected primary company announcements, government reporting and an original survey that answer common questions about Marketplace scale or shopping risk. We transcribed published numbers without projecting totals, filling missing values, or estimating a platform-wide scam rate. The CSV’s value column stores proportions as decimals; its definition column preserves the original population and wording in summary form.
The source date identifies the publication or displayed modification date; the reviewed date is our check date. Where the exact publication day was not established, the ledger uses the published month. A missing denominator is described explicitly rather than recorded as zero.
When using this page later, reopen the original source. Product statistics may be superseded, and reporting agencies may revise datasets. Cite the publisher for the statistic and this ledger only for the way the sources have been organized.
Questions people ask
How many people use Facebook Marketplace?
The sources reviewed here do not establish a current global monthly active user count. Meta’s July 2026 announcement reports listings and a specific US young-adult daily-use measure. Neither can be converted into a worldwide buyer count.
What percentage of Facebook Marketplace listings are scams?
These sources do not provide that rate. A valid estimate would need scam listings and all comparable listings from the same population and period. Social-media loss totals and surveys of online shoppers do not supply that denominator.
Can I reuse the statistics in a report?
Yes. Use the downloadable source ledger, retain each figure’s geography, period and definition, and cite the primary publisher. The review date records when we checked the page; it does not change the period measured.
When did Facebook Marketplace launch?
Meta announced Marketplace on 3 October 2016. Its launch article described more than 450 million monthly visitors to buy-and-sell groups, a different measure from Marketplace use.
Sources
- 01Meta Connecting Real People on Facebook
- 02Meta Introducing Marketplace
- 03FTC Reported losses to scams on social media
- 04CAFC Top 10 frauds in 2025
- 05Pew Research Center Online shopping scam experiences