Valve has been trying to find unique and creative ways to improve discoverability on Steam, and has been using a system called Steam Labs to do so. Today Valve has released two more experiments for users to try out.
Experiment 5 is called Deep Dive, which is an official built-in version of the popular Steam Diving Bell application. In this you pick a single game and Steam pulls up a list of 9 games. Three of the games are marked as "very similar", three as "somewhat similar", and three as "similar gems", which are games that are highly rated but with few reviews. For example, I tried it using the game Remember Me. My "very similar" titles were VA-11 Hall-A, Transistor, and Cyberpunk 2077. For "somewhat similar" I ended up with Fallout 4, The Elder Scrolls V: Skyrim, and Wolfenstein: Youngblood. Finally, my similar gems were Fumiko!, Technobabylon, and Copy Kitty.
Experiment 6 is called Community Recommendations. This filters games by showing recent user reviews for them that people have found helpful. It allows you to get an idea of what people are currently playing and how they feel about it very quickly. In addition to this, you can adjust the filter some so you can pick reviews during a certain timeframe, having a specific amount of playtime, and also filter games by tags. Using this I quickly saw positive reviews for games I've never heard of before, such as Loco Parentis and The Itch.
In addition to this, Valve has considered experiment 2, the Interactive Recommender, to be enough of a success that they've now promoted it to the front of the Store page. This is like the regular game recommender, only you can adjust certain factors, such as weighing by popularity, release date, filtering and excluding tags, and excluding wishlisted games.
Both Deep Dive and Community Recommendations are available right now on Steam.
How do you feel about these Steam Lab experiments? Going to play around with them? Find any new gems while you've done so? Let us know in the comments below!