User-Generated Content (UGC)：决定规则的是来源，而非外观
 - Arfadia.jpg "User-Generated Content (UGC)：决定规则的是来源，而非外观
- Arfadia")
There is no widely accepted definition of user-generated content (UGC), so this entry uses the OECD's 2007 working definition. Under it, UGC is content made publicly available online, reflecting some creative effort, and created outside professional routines and practices.
That last test does most of the work. An unboxing video can come from a buyer who liked the product, a buyer sent a free sample, or a creator the brand paid and briefed. On screen, the three may look identical. Under disclosure, review, and reuse rules, they are different things. The rules follow where the content came from, not what it looks like.
This table applies the OECD's 2007 working definition of user-created content to example content. The pass and fail calls are this entry's analysis, not OECD rulings.
| OECD test | What it checks | Example that passes | Example that does not pass | Paid creator UGC-style content | Label / source |
|---|---|---|---|---|---|
| 1. Publication | The content is made publicly available over the internet. | A public product review | A private message to a friend | Passes easily | OECD (2007), VERIFIED; examples and verdict: Analysis |
| 2. Creative effort | The person adds something of their own, such as an opinion, a photo, or an answer. | A customer's opinion, a photo of the product in use, or an answer on a Q&A page | Content where the person adds nothing of their own | Passes easily | OECD (2007), VERIFIED; examples and verdict: Analysis |
| 3. Outside professional routines and practices | The content is not produced as part of a job or a commercial arrangement. | An unboxing video from a buyer who liked the product | An unboxing video from a creator the brand paid and briefed | Fails, because it is made inside a commercial arrangement | OECD (2007), VERIFIED; examples and verdict: Analysis |
| Format (not a test) | None of the three tests is about format. | A review, a photo, or a forum reply can pass | The same review, photo, or forum reply can fail, depending on who made it and why | Can look like UGC, but the look does not decide which rules apply | Analysis |
What Counts as User-Generated Content
The Three Tests in the OECD Working Definition
The OECD report Participative Web: User-Created Content uses the term user-created content (UCC) for what marketers call UGC, and proposes three working criteria instead of a fixed rule.
The first is publication. The content has to be made publicly available over the internet, so a private message to a friend does not qualify while a public product review does. The second is creative effort: the person adds something of their own, such as an opinion, a photo, or an answer. The third is creation outside professional routines and practices, meaning the content is not produced as part of a job or a commercial arrangement.
None of the three tests is about format. A review, a photo, and a forum reply can each pass or fail depending on who made it and why.
Where Commissioned "UGC-Style" Content Fits
Advertising briefs now use "UGC" to describe a look: handheld video, a natural voice, an everyday setting. Much of it is made by creators the brand pays or compensates. This entry treats it as a labeled subtype, commissioned or paid UGC-style content, rather than folding it into organic UGC.
The reasoning starts with the OECD report itself, which states that most user-created content activity is undertaken without the expectation of remuneration or profit (VERIFIED). The report is also reported to note that the line between amateur and commercial content is getting harder to hold (REPORTED). Paid creator content sits right on that line. In our analysis, it can pass the first two tests easily and still fail the third, because it is made inside a commercial arrangement.
The practical consequence is disclosure. In the United States, the FTC's Endorsement Guides, revised in 2023, are reported to require clear disclosure of a material connection between an endorser and an advertiser, with payment and free products as typical examples (REPORTED). So "UGC-style" describes the aesthetic. The payment decides the obligation.
Common Forms of UGC
The OECD report sorts user-created content by type, from text and images to audio and video, and by hosting platform, such as blogs, wikis, and social networks. In marketing work, the familiar forms are narrower: customer reviews and star ratings, photos and videos of a product in use, threads and answers on forums and Q&A pages, comments under blog or social posts, and longer pieces such as tutorials written by enthusiasts. Each can be organic, incentivized, or commissioned, and the form will not tell you which.
Comments and replies feed an engagement rate calculation, but a high count says nothing about whether they came from customers, staff, or purchased accounts. The FTC rule discussed below separately prohibits buying or selling fake indicators of social media influence, such as bot-generated followers or views. Counting UGC is easy. Knowing where it came from is not.
Two posts can look identical and still carry different disclosure, review, permission, and link rules. What decides it is who made the content and what they got in exchange.
| Who made it | Disclosure | Review rules (US) | Permission before reuse | Links on your own site | Label / source |
|---|---|---|---|---|---|
| Customer acting on their own (organic UGC) | Disclosure depends on a material connection, and none is described for a customer posting on their own. | Genuine reviews are allowed. The business may not use groundless threats or intimidation to remove negative reviews, and may not claim a review section shows all or most reviews while negative ones are suppressed. | A public post is not a license. Ask first and record where and for how long it may appear. | Mark user-placed links with rel="ugc". Site owners might remove the attribute for members who have made valuable contributions over time. | OECD (2007), VERIFIED; FTC 16 CFR Part 465 (2024), VERIFIED; Google Search Central, VERIFIED; permission default: Analysis |
| Customer given a free product or other incentive | May have a material connection. Free products are a typical example. | Incentives are allowed only when they are not conditioned on a particular sentiment, whether stated or implied. A coupon for any honest review is different from a coupon only for five stars. | Ask first and record the agreed use, including where and for how long. | Mark user-placed links with rel="ugc". | FTC Endorsement Guides (2023), REPORTED; FTC 16 CFR Part 465 (2024), VERIFIED; an allowed incentive can still raise the disclosure question: Analysis |
| Employee | May have a material connection the audience would not expect. | Undisclosed insider reviews are prohibited. | Ask first and record the agreed use. | Not covered separately in this entry. | FTC Endorsement Guides (2023), REPORTED; FTC 16 CFR Part 465 (2024), VERIFIED |
| Contest or hashtag challenge entrant | If entrants can win something, the material connection question applies, even when the posts look spontaneous. | Not covered separately in this entry. | Ask first and record the agreed use. | Not covered separately in this entry. | FTC Endorsement Guides (2023), REPORTED |
| Paid or briefed creator (commissioned UGC-style content) | Has a material connection, which requires clear disclosure. Payment is a typical example. Disclosure tells the audience what kind of content it is. It does not make the content organic. | Buying positive or negative reviews is prohibited. | Record what was agreed, including where and for how long the content may appear. | Not covered separately in this entry. | OECD (2007), VERIFIED, fails the third test: Analysis; FTC Endorsement Guides (2023), REPORTED; FTC 16 CFR Part 465 (2024), VERIFIED |
| AI model posting as an invented customer | This is not user-generated content in any real sense. | Reviews attributed to people who do not exist are prohibited. The FTC's announcement names AI-generated fake reviews as an example. | There is no real creator to ask. | Not covered separately in this entry. | FTC 16 CFR Part 465 and FTC press release (2024), VERIFIED; no creator to ask: Analysis |
| Real customer, content edited with AI | This entry states no legal disclosure rule for it. In one study, labeled AI-edited reviews landed in between on consumer trust and perceived authenticity. | A harder case. This entry names no specific rule for real reviews edited with AI. | Ask first and record the agreed use. | Not covered separately in this entry. | Drozd and Söilen (2026), VERIFIED; convenience sample of 370, Latvia |
Same Content, Different Obligations
Once the origin is clear, the obligations follow. Each check below depends on who made the content, what was given in exchange, and whether a machine made or edited it.
Disclosure When There's a Material Connection
Under the Endorsement Guides (REPORTED, 2023 revision), the question is whether the poster has a connection to the brand that the audience would not expect and that could change how much weight they give the post. A paid creator has one. So may an employee, a customer given a free product for posting, or a contest entrant chasing a prize. A viral marketing push built on a hashtag challenge can produce posts that look spontaneous, but if entrants can win something, the same question applies.
Disclosure does not turn paid content into organic UGC. It tells the audience which kind they are looking at.
What Review Rules Prohibit
In the United States, the FTC's Trade Regulation Rule on the Use of Consumer Reviews and Testimonials, 16 CFR Part 465, was published in the Federal Register on 22 August 2024 and took effect on 21 October 2024. According to the FTC's 14 August 2024 announcement, it prohibits fake reviews and testimonials, including those attributed to people who do not exist, along with buying positive or negative reviews, undisclosed insider reviews, company-controlled review sites presented as independent, and buying or selling fake indicators of social media influence.
The incentive point is easy to misread. The rule does not ban every incentive. It bars offering compensation or other incentives conditioned on the review expressing a particular sentiment, whether that condition is stated outright or implied. A coupon for any honest review is a different act from a coupon only for five stars. In our analysis, an incentive within the rule can still raise the disclosure question above.
Then-FTC Chair Lina M. Khan explained the rationale this way: "Fake reviews not only waste people's time and money, but also pollute the marketplace and divert business away from honest competitors."
In the United Kingdom, the clearest public signal comes from the Competition and Markets Authority. On 25 July 2025, the CMA said its website review had found that 54 of more than 100 businesses could be failing to comply with its fake reviews guidance (VERIFIED, UK, CMA case page). This is possible non-compliance, not a ruling of infringement. Announcing Amazon's undertakings in June 2025, CMA Chief Executive Sarah Cardell said star ratings and reviews strongly influence shoppers' choices.
For scale, a study by Alma Economics for the UK Department for Business and Trade, published 25 April 2023, estimated that 11% to 15% of reviews in three common product categories were fake, based on a model applied to 2.1 million reviews across nine UK e-commerce platforms (VERIFIED, UK, 2023; a modeled estimate range for those categories, not a global rate). In the same study's online experiment with 4,900 UK online shoppers, participants were "3.1% more likely" to buy a product that carried well-written fake reviews (VERIFIED, UK, 2023; simulated shopping task in which participants did not spend their own money).
AI-Generated and AI-Edited UGC
Origin includes machines. The FTC rule covers reviews that misrepresent that they come from someone who does not exist, and the FTC's announcement names AI-generated fake reviews as an example. A review written by a model and posted under an invented customer's name is not user-generated in any real sense, and in the US it falls inside the rule.
Harder cases involve real customers' content edited with AI. A 2026 study by Dariia Drozd and Klaus Solberg Söilen in the Journal of Theoretical and Applied Electronic Commerce Research kept the same review text across three conditions: no AI information, an AI-edited background labeled as such, and an AI-generated image labeled as such. Reviews in the AI-generated condition scored lowest on consumer trust and perceived authenticity, and the AI-assisted condition landed in between (VERIFIED). The authors write that "the key issue is not AI disclosure alone, but how AI-related cues shape perceived authenticity," which in turn affects trust.
Two limits apply. The study used a convenience sample of 370 digital marketplace users in Latvia, so it is not representative. And the labels and the images changed together, so the effect of the label on its own cannot be separated from the effect of the image.
Permission Before Reuse
A public post is not a license. A brand that puts a customer's photo or video on its website, in an email, or in an ad is using someone else's work. In our analysis, the safe default is to ask first and record what was agreed, including where and for how long the content may appear. This entry does not describe the copyright law of any particular country.
Reuse also changes who is speaking. Reposting a customer's video to extend a brand account's organic reach puts it in the brand's voice. When UGC feeds a wider content marketing plan, the permission record should travel with it, so no later team assumes a tagged post is free to use.
UGC on Your Own Site: Spam, Moderation, and Links
User-Generated Spam and Moderation
Google's spam policies define user-generated spam as spammy content added to a site by users through a channel intended for user content, and they note that site owners are often unaware of it. Examples include spammy forum posts, blog comment spam, and spammy accounts on open hosting services.
Hosting UGC is not a problem in itself. The same document's site reputation abuse policy lists sites designed to allow user-generated content, such as forums and comment sections, among examples that are not considered abuse. The risk is leaving those channels unmoderated.
Moderation has a legal edge in the US. The FTC rule prohibits unfounded or groundless legal threats, physical threats, or intimidation used to remove negative reviews, and prohibits claiming a review section shows all or most reviews submitted when negative ones were suppressed. Removing spam is moderation. Hiding honest criticism is not.
Marking Links With rel="ugc"
Google recommends marking links in user-generated content, such as comments and forum posts, with the rel="ugc" value (Google Search Central, page last updated 10 December 2025). Google also says owners might remove the attribute from links posted by members who have made valuable contributions over time. So a forum can treat a new account's link differently from a long-standing contributor's. Again, who added the link matters more than how it looks.
A Pre-Publishing Check for UGC
Before UGC goes onto a product page, into an email, or into an ad, five questions cover most of this entry. Treat them as an analytical checklist, not legal advice.
First, who made it: a customer acting on their own, an employee, a paid or gifted creator, or a model? Second, was anything given in exchange, and if so, was it tied to what the content had to say? An incentive conditioned on sentiment is where the FTC rule draws its line. Third, does the audience need a disclosure? Any material connection points toward yes. Fourth, is there permission for this specific use, recorded somewhere a colleague can find it? Fifth, if the content sits on your own site, is it moderated for spam without suppressing honest negative reviews, and are user-placed links marked with rel="ugc"?
For UK-facing businesses, the CMA's July 2025 letters add a sixth: are measures in place to prevent fake reviews? The CMA says businesses must have them.
Two posts that look the same can produce different answers. That is why the questions are worth asking.
Frequently Asked Questions
Is paid creator content user-generated content?
Not in the strict sense. The OECD working definition requires creation outside professional routines and practices. Paid content in a UGC style is treated here as a labeled subtype, commissioned or paid UGC-style content, and in the US a material connection to the brand is reported to require disclosure under the FTC's Endorsement Guides.
Can a business offer an incentive for reviews in the US?
The FTC's 2024 rule prohibits incentives conditioned on a review expressing a particular sentiment, whether the condition is stated or implied. By that wording it does not ban every incentive, but an incentivized review can still raise a disclosure question.
Are AI-generated reviews covered by the FTC rule?
Yes, when they misrepresent that they come from someone who does not exist. The FTC's announcement names AI-generated fake reviews as an example of what the rule prohibits.
Does UGC help SEO?
No official Google documentation reviewed for this entry describes UGC as a ranking boost. Google does document how to handle it: mark user-placed links with rel="ugc" and keep user channels free of user-generated spam.
Can a brand repost a customer's public post?
Public visibility is not the same as permission. The safer practice is to ask the creator and record the agreed use before reposting, especially in advertising.
How common are fake reviews?
The sources used here contain no global figure. For the UK, a 2023 study commissioned by the Department for Business and Trade estimated that 11% to 15% of reviews in three product categories were fake, a modeled range based on 2.1 million reviews.
Related Terms
- Influencer Marketing: paid or gifted creator partnerships, where disclosure applies by default.
- Community Management: moderating the forums, comments, and groups where UGC lives.
- Brand Awareness: a goal UGC programs may serve, measured separately from the content.
Sources
- OECD (2007). Participative Web: User-Created Content, DSTI/ICCP/IE(2006)7/FINAL. Working definition and remuneration statement VERIFIED; observation on the blurring line with commercial content REPORTED. https://www.oecd.org/content/dam/oecd/en/publications/reports/2007/09/participative-web-and-user-created-content_g1gh826c/9789264037472-en.pdf.
- Google Search Central. Qualify your outbound links to Google (last updated 10 December 2025). Vendor documentation, VERIFIED. https://developers.google.com/search/docs/crawling-indexing/qualify-outbound-links.
- Google Search Central. Spam policies for Google web search (last updated 28 August 2026). Vendor documentation, VERIFIED. https://developers.google.com/search/docs/essentials/spam-policies.
- Federal Register, 22 August 2024. Trade Regulation Rule on the Use of Consumer Reviews and Testimonials, 16 CFR Part 465. US, VERIFIED. https://www.federalregister.gov/documents/2024/08/22/2024-18519/trade-regulation-rule-on-the-use-of-consumer-reviews-and-testimonials.
- Federal Trade Commission press release, 14 August 2024. Federal Trade Commission Announces Final Rule Banning Fake Reviews and Testimonials. US, VERIFIED. https://www.ftc.gov/news-events/news/press-releases/2024/08/federal-trade-commission-announces-final-rule-banning-fake-reviews-testimonials
- Federal Trade Commission. Guides Concerning the Use of Endorsements and Testimonials in Advertising, 16 CFR Part 255, 2023 revision. US, REPORTED. https://www.ftc.gov/legal-library/browse/federal-register-notices/16-cfr-part-255-guides-concerning-use-endorsements-testimonials-advertising.
- UK Competition and Markets Authority. Online consumer reviews case page, gov.uk, update of 25 July 2025, and CMA announcement on Amazon's undertakings, 6 June 2025. UK, VERIFIED. https://www.gov.uk/cma-cases/online-consumer-reviews. https://www.gov.uk/government/news/amazon-gives-undertakings-to-cma-to-curb-fake-reviews.
- UK Department for Business and Trade (2023). Fake online reviews research: executive summary, prepared by Alma Economics, 25 April 2023. UK, data before 2024, VERIFIED. https://www.gov.uk/government/publications/investigating-the-prevalence-and-impact-of-fake-reviews/fake-online-reviews-research-executive-summary.
- Drozd, D., and Söilen, K. S. (2026). AI Labels, Perceived Authenticity, and Consumer Trust in User-Generated Reviews. Journal of Theoretical and Applied Electronic Commerce Research, 21(5), 154. https://www.mdpi.com/0718-1876/21/5/154. Convenience sample, Latvia, VERIFIED.
