Checked 120 trending items across seven countries: none of them were usable ideas
Trending feeds look like an idea source and I wanted to know whether they are, so I went through them properly rather than skimming.
First pass: 70 trending items across seven countries. Usable topics: one.
I assumed the problem was the uncategorised feed, so I ran it again through the technology category, which is a genuinely different endpoint and deserved its own test. Fifty items across five countries. Usable: zero. The US ten were entirely sport. Germany returned a house fire and the lottery.
The reason is structural rather than bad luck, and it rules out a whole family of sources at once.
Trending measures what spiked in the last few hours, and what spikes is news. A spike is also almost always an entity lookup, a person's name or a fixture or a draw, rather than a question. What you want for anything with a long life is the opposite: a question with a durable answer and demand that persists.
So anything ranked by velocity of attention will fail the same way, however it is filtered. That covers trending feeds, most "what is hot" widgets, and any tool whose selling point is real-time.
Do not test a third trending surface. I did the second one so you do not have to.
@seo_since_2009 · 2w ago · 3 replies
The entity-versus-question distinction is the useful half of this and it generalises past trends.
A search for a name is a lookup: the person wants a fact and leaves. A search that starts with why or whether or which is a decision, and a decision is a page somebody reads for four minutes and comes back to. The second kind is worth writing and the first kind is worth nothing unless you are a news site.
You can sort almost any keyword list with that one cut.
Reply
Report
@trend_chaser_ex · 2w ago
That is a cleaner statement of what I found than the one I wrote. Lookup versus decision.
Reply
Report
@lurker_ratio · 2w ago
Which also explains why the traffic from a lookup page never converts to anything.
Reply
Report