Brigading

A brigade starts with a call to action. Someone with an audience, often a forum moderator, an influencer or an organized channel on a chat platform, points followers at a target and implies what to do: downvote, leave a one-star review, mass-report an account, or pile into a comment thread. The participants are usually real accounts operated by real people, which is the main difference between brigading and a bot-driven attack.

The tactics vary by platform. On forum-style sites the tool is the vote, and brigades try to sink or elevate a post. On review sites the tool is the rating. On social networks the tools are replies, quote posts and abuse reports. A brigade against a company page might combine several of these in the same few hours.

Detection depends on timing and network structure rather than on content alone. Analysts look for a sudden spike in volume from accounts that had no prior interaction with the target, clusters of accounts that share a referring link or source community, near-identical phrasing, and a sharp mismatch between the volume of activity and the size of the underlying audience. A spike that traces back to a single upstream post is the strongest signal.

A hypothetical illustration: a regional retailer announces a policy change, a widely followed account posts a screenshot with a call to leave one-star reviews, and within a day the retailer's location listings collect hundreds of new one-star ratings from accounts that have never reviewed the business. The reviews rarely describe an actual visit.

The reverse also occurs. A group can brigade in support, pushing a post or a hashtag into visibility through coordinated upvotes and shares, which makes a fringe view look like a broad one.

Platforms name the behavior in their rules. Reddit's site rules, for example, prohibit organized vote manipulation and brigading between communities.

Brigading distorts the signals a comms team uses to decide whether a story is real. A wave of hostile posts can look like a groundswell of customer opinion when it is a few hundred coordinated accounts. Responding to it as if it were organic criticism can amplify it, while ignoring genuine criticism because it "looks like a brigade" is the opposite mistake.

The practical question is provenance: where did the wave start, how many independent accounts are in it, and is it moving into mainstream coverage? Tracking a story from its first appearance to its pickup by news outlets is what lets a team decide whether to respond, request platform enforcement, or wait. See narrative intelligence for how PeakMetrics frames that tracking.

Brigading is not the same as a bot attack. Brigades are mostly people; bots may join, but the defining feature is coordination. Detecting them requires looking at behavior and origin, which is the domain of bot detection and coordinated inauthentic behavior analysis.

It is also not the same as a pile-on that forms organically. When many people independently react to the same news, the volume can look identical, but there is no organizer. The distinction matters for response, since an organic pile-on reflects a real audience reaction and a brigade reflects an organized minority.

Finally, a brigade does not have to be hostile to be brigading. Coordinated positive activity that misrepresents the scale of support is the same mechanism, closer to astroturfing.