i've found plenty of interesting titles through discovery queues

Originally posted by LaurenceHazlewood:

I have lost count of the times I've gone through the entire list clicking on "Ignore" for every title.

There is your problem. I've tested it, and if you have TOO MANY titles ignored, Steam's algorithm starts throwing random stuff with few tags assigned at you. I used to have around a 1000 ignored titles and was constantly getting awful recommendations but then I cleared the list down to under 200. My recommendations became good again.
Why does that happen if you have too many ignored titles?
Example: if you ignore enough games with the tag "Strategy", Steam stops recommending you games like "Age of Empires", "Sins of a Solar Empire", "Supreme Commander"....

The more games you ignore, the more associated tags the algorithm assumes you DON'T want to see. Eventually, it's at loss because all tags end up included and then you get only weird, out of left field recommendations.

The discovery queue is based on many of the games we wishlisted to provide a better experience finding games of interest.

Originally posted by LaurenceHazlewood:

What is the point of the Discovery Queue?

Personalized advertising. There is no real value for users, especially when you look at the product classifications on Steam, which are completely out of control.

But when countless people here already provide free advertising for developers and publishers on their profiles or activity feeds by displaying screenshots, artworks or even showcasing games, it is not surprising that this mechanism is widely used...

It gives me good titles, but I don't run it very often. The more you go through the Discovery queue, the more likely it is that the queue has shown you good stuff. Games don't come back, so at one point the system has no choice anymore.

Also happens when people ignore a lot of titles, mind.

A lot of the games I played over the years, I've first found on Discovery Queue. I wouldn't say it's accuracy is very high, when it comes to my tastes, but I would say each queue has like 60% chance to have something, that will end up on my whishlist (that otherwise I wouldn't know of).

The discovery queue is great and I've found a lot of games there that I would otherwise not have discovered. Not every game release has promotion like Battlefield etc.

The Discovery queue mainly goes by the tags on the games you own and then if you block tags it tries to remove those games.

Problem that happens is that people are blocking far too many tags which will hit nearly every single popular game and they are left with the trash.

But the queue or even an AI wouldn't be able to tell if a game is good or not for you. It's just not possible.

Mileage may vary. As with every other tool it's best when used sparingly. At a certain point, some people have viewed so many titles (and some have ignored so many) that the system simply starts throwing things at the wall to see what sticks.

Originally posted by LaurenceHazlewood:

I have lost count of the times I've gone through the entire list clicking on "Ignore" for every title.

And that's probably the reason why you're getting bad results. You've told the system you don't like so many things it struggles finding something not having stuff you don't like.

And that's another misconception of how the queue works. You don't need to click on 'ignore' for every title that is shown to you. Once it's on your queue that's it, it's not going to show up again. If you don't interact with it you've already 'ignored' it.

Discovery queue has been great for me and I've found quite a few gems. It's a useful way to surface games and dlc.

100 pages of slop.. then one good game followed by.. 100 pages of slop..
and it loads like a constipated isp...

It's supposed to show you games based on the stuff you play.

Reckon you should prolly only hit ignore on stuff you don't want to see.

Example: I got strict adult filters and hentai excluded from search results and I see games recommended in the genres - or like the games - I play.