What is a people graph?

A people graph is a record of the people who show up in the public conversations around your category, what they engaged with, and what you have already said to them. It is built from public activity: who liked, reposted, quoted, or replied to a post, who asked the question, who answered it.

What goes into a people graph?

Three things. People, with their handle per platform and whatever they say about themselves publicly. Posts, with who engaged and how. Your own history, meaning every reply, mention, and note you have already made about that person. The third part is the one most tools skip and the one that matters daily, because it is what stops you answering the same person twice in a week with the same suggestion. Each engager on a competitor's launch post is a fact about the market, and a hundred of them together describe who your category is actually for.

Why keep a people graph instead of a contact list?

Because the use is different. A contact list exists to be contacted. A people graph exists to tell you where to be and what to say. Reading it tells you which communities your buyers post in, which accounts they listen to, and which complaint keeps repeating. None of that requires messaging anyone, and messaging strangers who liked a post is the fastest way to get muted. Reddit's self promotion guidance describes the alternative: participate where these people already are. That choice of place is a presence, and the graph is how you pick one.

How is a people graph used in marketing?

Start from a post that did well in your category, pull the people who engaged, and look for the pattern. Their shared communities become your listening targets. Their repeated complaints become your posts. Their questions become the page you write. Tagging keeps those groups legible over time, as described on the people graph feature page.

Questions

Is a people graph the same as scraping a list?

No. A list is pulled to be sold or mailed. A graph is read to understand a market. The distinction is what you do next: publishing where those people already gather is marketing, while messaging them privately because they liked something is not, and platforms treat it accordingly.

Where does the data in a people graph come from?

From public engagement on public posts: likes, reposts, quotes, replies, and comments that anyone can see by opening the post. Nothing private, nothing bought. Coverage varies by platform, since some show every account that engaged and others show only a sample.

What do I do with someone once they are in the graph?

Usually nothing directly. Watch what they post, answer them in public when they ask something you can answer, and use the aggregate to choose where you spend your week. The individual row matters only when that person is already in a conversation you belong in.