The Twitter Discover Tab: For Suggestions That Aren’t Useful Or Entertaining, ‘Customized’ For You

Back in December Twitter unveiled a major redesign of, which not only included a complete overhaul of the user interface, but also a change to the menu navigation, and a brand-new feature: the Discover tab.

Using a fiendishly clever mathematical algorithm developed by the finest of minds, the Discover tab promised to let users “tap into a stream of useful and entertaining information, customized just for you”.

If only that were true. Because, from where I’m sitting, Discover not only isn’t customized, it also isn’t useful and it isn’t entertaining. Instead, it’s The Emperor’s New Clothes of Twitter features – by all accounts, not much more than a randomly-generated list of popular news stories and trending topics, and a waste of everybody’s time.

Twitter expanded on the role the Discover tab was supposed to play in their supporting documentation for the revamp.

When you use Discover, you’ll see results reflecting your interests—based on your current location, what you follow and what’s happening in the world. As you use Twitter more, Discover gets even better at serving up more content just for you.

Whatever you’re curious about, Discover will help you find out more.

Hey, that sounds great! Let’s compare that promise with a screenshot of the contents of the Stories section (which is the first thing you see) of my current Discover tab.

(Apologies – this is a big image.)

What, pray tell, dear Twitter, is customized about this?

Aside from a passing interest in Roger Federer’s win at the Australian Open, there is literally nothing in this list that has any appeal to me whatsoever.

I never participate in trending hashtags. I couldn’t give a monkeys about the Northern Lights or that Julian Assange is hosting a TV show. I don’t know who Andrew Lansley is. I couldn’t care less about a Blink 182 ‘sound off’, nor news about Freddie Mac or Garzón. You may care, but that isn’t the point. This is supposed to be for me.

Bottom line: this information has not been customized for my attention, as I have never shown any interest on Twitter in these subjects. Even the tennis result seems to have been generated independently, as I don’t tweet about tennis.

Evidently, this ‘algorithm’ is all something of a ruse.

Let’s clear something up: this isn’t an attack on knowledge. I’m all for learning something new. But this isn’t the purpose of the Discover tab. It’s supposed to be enlightening me with relevant information – not a bunch of random news items that are currently doing the rounds. If I wanted crowd-sourced news, I’d go to Reddit, where I can at least pick and choose the topics I care about. With the Discover tab, I’m not permitted any direct influence on what I’m supposed to be discovering. But that shouldn’t matter, as it’s supposed to be doing it all for me. I should be getting lots of wonderful goodness courtesy of some all-seeing Twitter algorithm. Automatically. Like science-backed magic.

Instead: I get this. And it’s terrible.

This criticism isn’t just limited to the Stories section of Discover. Activity has always been a terrible idea (and I think Twitter agreed, which is why they effectively buried it), and Who To Follow seems plagued by the same ropey algorithm that is powering Stories. I mean: it’s actually recommending that I follow Martha Stewart.

Of course, we’ve been here before. Twitter’s recommendation algorithms have always been, well, rubbish, with results almost always verging on the side of the ludicrous and indifference. You know, just to be safe. I’m not saying that the mathematics that powers these tools is a walk in the park, but couldn’t it be just a little more accurate? A litter more relevant?