Not every account is a real person. Bots, sock puppet accounts, and paid personas flood platforms to push narratives, inflate numbers, and deceive. Spotting them protects your analysis from manipulation and is the first step in exposing coordinated inauthentic behavior.
What you'll need
- The account's public profile, history, and follower graph
- Social Media analysis tools
- Reverse image search from the Data Acquisition category for profile photos
Steps
- Read the profile signals. Check the creation date, a generic or missing bio, a suspicious handle (name plus long number strings), and whether the display name matches the activity.
- Reverse-search the avatar. Run the profile photo through reverse image search; a stock image, a stolen photo, or an AI-generated face is a strong tell.
- Analyse the behaviour. Look at posting volume and timing — round-the-clock activity, extreme frequency, or bursts synchronised with other accounts suggest automation.
- Inspect the content mix. Genuine accounts vary; fakes often repost, share near-identical phrasing, or push a single narrow agenda relentlessly.
- Map the network. Check who follows and interacts with the account. Clusters created together, following only each other, point to a coordinated operation rather than one lone bot.
Common pitfalls
- A real person who looks "bot-like." Highly online, single-issue users are easily mislabelled — no single signal is proof.
- Judging one account in isolation. The strongest evidence is the pattern across a network, not any one profile.
- Trusting follower counts. Followers and engagement are cheaply bought and mean little on their own.
Verify your result
You can call an account inauthentic when several independent signals converge — a reused photo, automated timing, repetitive content, and a suspicious network — and you can articulate the pattern, not just a hunch.