Moderation — is the review of ads, websites, and campaign settings against the advertising platform’s rules before the ads start being shown to users.
Every social network makes sure that ads do not mislead users, violate the law, or annoy people. If an ad fails the review, the campaign is rejected, and in the case of serious violations, the account may be blocked immediately.
Who reviews ads and how
The review almost always takes place in two stages:
-
Automated moderation (bots and algorithms). The first and fastest filter. Neural networks and bots scan the text for prohibited words and phrases (stop words such as «lose weight in 3 days» or «easy money»), recognize objects in images and videos, and also follow the link to check the website itself (landing page).
-
Manual moderation (human moderators). Some ads are additionally reviewed by platform employees. Such a review may be required if the automated system cannot clearly assess the ad or if questionable elements are found in the campaign. It is much harder to fool a person than a bot.
What the platform checks
It is important to remember: moderators evaluate not only the image in the feed, but the entire chain:
-
Creative: text, headlines, visuals (image or video), and emojis.
-
Landing page: the website the link leads to. Moderators look for aggressive countdown timers, fake reviews, promises of instant results, intrusive pop-ups, and missing required information (privacy policy, contact details).
-
Targeting settings: whether the audience matches the content of the ad (for example, age-restricted content must not be shown to minors).
What is important to remember
-
Post-moderation (background review). An ad is not reviewed once and for all. Even if it successfully passes the initial review and is already running, the platform’s bots regularly rescan the campaign while it is active. If users start reporting the creative or the algorithm checks the landing page again, a rejection or ban may occur while the budget is actively being spent.
-
Account trust and appeals. The strictness of the review depends on the profile’s reputation: a fresh, newly created account is scrutinized much more closely, while warmed-up profiles with a history of legitimate ad spend pass the filters more easily. In addition, a rejection — is not always final: if the ban was issued because of a bot error, the ad can often be restored through support or a request for another manual review.
Moderation in «white-hat» and «gray-hat» traffic arbitrage
In «white-hat» verticals (regular online stores, services, online schools), moderation — is a standard routine: it is enough to follow the network’s rules, avoid making unrealistic promises, and make minor corrections in time if an ad is rejected because of a typo or excessive use of capital letters.
In «gray-hat» verticals (gambling, crypto, nutra), moderation turns into a constant struggle with algorithms. Webmasters use neutral creatives without direct hints, hide the actual websites using cloaking (bots are shown a white informational website, while users see the offer), or redirect traffic through apps and Telegram channels.
