Pressfront / Trace / What is attention intelligence?
Explainer · Updated September 2026
Two questions about the same story. How fast is it moving, and where did it first appear in public. The first is weather. The second is forensics. Attention intelligence is a tool that answers both on one screen and tells you how sure it is about each.
Attention intelligence is weather plus forensics. The weather half measures how fast a story is moving right now: volume against its own recent baseline, velocity over the last six hours against the six before that, which platforms it has jumped to, and who is carrying it. The forensics half works backwards to the earliest public appearance it can verify, and reports how confident it is and what it could not see.
The load-bearing word is public. A tool that never logs in, follows, or touches a private account can only ever report the earliest public appearance, so that is what it should print, in those words. Every number carries a label: measured straight from a platform, derived arithmetic over things measured, and modeled a judgment with the rule shown next to it.
It is not prediction. Nothing here tells you what will go viral, and the honest version of this product says so on the page rather than in a footnote.
The two halves, on one real result
The Project Anchor gravity trace, 2026-09-16This is one real result, not a mock. The full trace, including the run that got it wrong first, is on the story page.
The private-account rule. We never log in, follow anyone, request access to a locked account, or use scraped private data. A private account is invisible to us and stays that way. So what you see below is the earliest public appearance we could verify, never a proven origin. If a story started in a group chat, a DM or a locked profile, the first public post is the node we can see, and we say so rather than dressing it up as the source.
Run one yourself: Trace a story. How the sweep and the confidence number are built: the methodology card.
Weather is the half people expect. Attention score against the story’s own baseline, velocity over the last six hours compared to the six before it, how many platforms it has reached, who the biggest carrier is, and whether mainstream news has picked it up yet.
The thing that makes it useful rather than decorative is the baseline. "Mentions are up" is meaningless without knowing what normal looks like for this subject, and a plain count of posts is the reason so many monitoring dashboards get ignored after the second week. Velocity against a baseline is a signal. Volume on its own is a chart.
Breakout is the modeled part, shown as High, Medium or Low with the three rules that produced it printed next to it. It stays a label rather than a percentage until there are enough recorded outcomes to back a percentage honestly.
Forensics runs the other direction. Expand the query into the phrasings people actually used, sweep day by day across every platform reachable, drop what is off-topic, link the posts that quote or repeat each other, then take the earliest post with a timestamp you can trust.
Two rules make this defensible rather than defamatory. The result says earliest public appearance we could verify, never "origin" and never "started by". And a post that came after another with matching rare wording is described as posted after, with matching wording, never as "copied". Those are product rules that happen to also be legal ones, and both of them cost us punchier copy on purpose.
Three kinds of number end up on the same screen, and mixing them is how monitoring tools lose their credibility.
This is the part I care about most. A dashboard that prints a modeled guess in the same typeface as a platform timestamp is not being confident, it is hiding which of its numbers you are allowed to rely on.
Pressfront does two other jobs. Getting published puts an accurate description of you on the news and syndication sites Google indexes and AI answer engines read. Ads put a message in front of people who have not heard of you. Both are things you say on purpose.
Attention intelligence is the other direction: what is already being said, how fast, and where it first surfaced. It feeds the other two rather than replacing them. A narrative forming about your category is a genuine reason to publish into it, and knowing the fuse length tells you whether you have a week or an afternoon.
Attention intelligence measures how fast a story is moving right now and works backwards to the earliest public appearance of it that can be verified. Weather plus forensics. The weather half is volume against a baseline, velocity, platform jumps and the biggest amplifier. The forensics half is the earliest verifiable public post, a confidence number, and a printed list of what the sweep could not reach. Every figure is labeled measured, derived or modeled.
No. Trace describes what has already happened and how fast it is moving, and it does not forecast which stories will break out. Breakout appears as High, Medium or Low with its rules shown, and it stays a label rather than a probability until enough outcomes have been recorded to support one. Anyone selling a virality prediction is selling a model of the future that nobody has validated.
Measured means the number came straight from a platform, such as a post timestamp or a follower count. Derived means it is arithmetic over measured things, such as a velocity figure or the number of hours between two posts. Modeled means it is a judgment we made, and every modeled item shows the rule that produced it so you can disagree with the rule rather than the number.
No, and ours never will. We do not log in, follow, request access to a locked account, or use scraped private data. That is a hard rule, not a current limitation. It means every result reports the earliest public appearance rather than an origin, which is also the node that matters in practice: a rumour inside a locked group chat is not yet a crisis, the moment it is screenshotted into the open is.
Deciding whether something needs a response, and how quickly. Knowing that a claim about your company first appeared publicly eleven hours ago on one account with 2,000 followers leads to a different decision than knowing it has been sitting in a forum for six weeks and just reached an account with three million. Both of those are answerable. Which one will go viral tomorrow is not.
Free trace, no account for the first one. Weather and forensics on one screen.
Honest fine print. Trace reports the earliest public appearance we could verify inside the window we searched, which is not the same thing as a proven origin. We only read public posts and public metrics, so deleted posts, screenshots laundered out of private accounts, and anything that started in a DM or a locked group are invisible to us, and every result names the sources that came back empty. Confidence is a number we publish together with the gaps behind it, never on its own. Trace describes what has already happened and how fast it is moving right now; it does not predict what will go viral, and no honest tool does.