Daily Briefing, Trending Radar and Tracking
Three ways Feedpik reads your sources for you — what each one answers, the thresholds that decide when they have something to say, and when to use which.
Three features sit below your reading views in the sidebar. They look similar and answer completely different questions.
| The question it answers | |
|---|---|
| Daily Briefing | What happened today, across everything I follow? |
| Trending Radar | What are my sources suddenly all talking about? |
| Tracking | Has anything happened to the thing I care about? |
They form a range. The briefing assumes you do not know what to ask. Tracking assumes you know exactly. The radar sits between them: you have not asked about anything, but something is worth your attention anyway.
Daily Briefing
Takes a day of your articles, groups them into topics, and writes each topic up across its sources — so three outlets covering the same development become one paragraph, not three headlines.
It only synthesises a topic covered by at least two articles, and covers up to six topics. A day when nothing overlapped produces a short briefing, which is the honest answer rather than a padded one.
Each topic carries a tone — positive, negative or neutral — read from the coverage rather than asserted.
There is also a weekly version, which uses seven days instead of one.
A briefing costs one AI credit, regardless of how many topics it synthesises. If the model is unavailable and the briefing degrades to no synthesis, it costs nothing.
Use it as the thing you read first: what happened, in one place, before you decide what to open.
Trending Radar
Watches your sources for a story that several of them picked up at the same time — the signal being not that something is popular, but that your sources converged on it.
A topic appears only when all of these are true within the last 48 hours:
- at least 3 mentions
- from at least 2 different sources
- at least twice the volume of the preceding 48 hours
That last one is what makes it a radar rather than a list of what you read most. A subject your sources always cover never trends; a subject that just doubled does.
An empty radar is normal and correct. Most days, nothing your sources follow accelerates like that. A radar with something in it every day would be one with the thresholds set to meaninglessness.
Each topic shows how many mentions, how many distinct sources, the ratio against the previous window, and who is covering it — enough to judge whether it deserves your time without opening anything.
Tracking
For when you know what you care about: a company, a ticker, a competitor, a technology.
Name it once — with aliases, since a company appears as "Nvidia", "NVDA" and sometimes both — and Feedpik watches your sources for it. What you get:
Volume against a baseline. Not "12 mentions" but "12 today, usually 3". A number without its baseline says nothing.
Sentiment, and the change in it. Whether coverage turned in the last few days is often the point, more than whether it is positive today.
Who reported it first, and how many hours ahead of the next source. Useful for learning which of your sources actually leads on a subject and which follow.
Which sources are carrying it, so you can see whether a story is broad or one outlet's preoccupation.
You can also have a watch alert you rather than waiting for you to look — to Telegram, Discord, email, or in the reader. A watch can fire on a spike, or specifically on coverage turning negative, which is the version people usually want for a competitor or a holding.
Which to use
Add tracking when you would be annoyed to hear something late. A company you hold, a competitor, a technology your work depends on. It costs nothing to leave running and only speaks when something moves.
Read the briefing when you have ten minutes and no particular question.
Check the radar when you want to know if you are missing something — but do not treat an empty one as a fault. It is the most common state, and it means what it says.
Common questions
The radar is always empty — expected. The thresholds are deliberately strict, and few sources make convergence rarer. More sources produce more radar hits, because the signal is agreement between them.
My briefing is very short — you had a quiet day, or few sources. It only writes up topics covered from more than one angle, and does not invent the rest.
Tracking shows nothing for a term I know was in the news — it searches your sources, not the web. If none of your subscriptions covered it, there is nothing to find. Add a source that covers that beat.
A tracked term matches the wrong articles — terms match as substrings, so a short one collides. Add aliases for the forms you want rather than shortening the term to catch them all.
Last updated August 6, 2026