I’m interested to see how this experiment goes, but from the announcement so far, I’m happy that removing CCUs is part of the experiment. I’ll explain why:
Completely ignoring how this could be a double-sided blade for surfacing bad faith actors (or, the above is also very valid), discovery has only ever benefitted those with top statistics for a long many years now and it’s made it horrible for new experiences to break through the surface without a significant financing effort on our part, to which the existing ad system is a bad gamble. This feels like it’d make experiences rather equal and rely less on “algorithm luck”.
Discovery has never felt like discovery, and the only experiences that get surfaced to me personally are those with already large communities and don’t need any further help from Roblox - but are still everpresent and always in my face. It’s difficult to find new experiences with the lack of developer-curated sorts (e.g. via genres, tags or otherwise), and lack of discovery is also a contributor to why our experiences can’t reach audiences even during peak update times.
We are currently lucky enough to have a longstanding and fiercely dedicated community spanning over a decade that, during peak update times (we update infrequently but massively when we do), our existing playerbase showing up to play and inviting some friends is what helps us through. We otherwise barely get any visibility in discovery that we otherwise had back when genres were up (which is what helped me find the game in the first place and eventually become one of its developers as well).
I wish the experiment well. A better discovery driven by both algorithms and developers will help us all win here, and the first step I feel would be to let all experiences have a chance to appear - rather than only letting discovery feel like it’s tailored towards top earners only.
ETA: This may be articulated a bit poorly. This is mainly in regards to the whole update. As for the removal of CCUs, I may hold the unpopular opinion of “I like the idea to experiment with that”, but time will tell how this will truly behave for experiences dependent on player-to-player interaction. As much as it indicates that there’s enough of a presence in sessions making the friction to joining easier, it also only rewards… obviously only popular experiences. No CCUs means equal chances. I think the precious balance is to figure out how to make it apparent that there are players playing.
There already has been quite a bit of backlash from the removal of the friends playing category and the addition of the little label that hides the game’s info.
Removing CCU from games is a huge deal. What if you are looking to play a social hangout and you end up joining no one in a recommended game given to you via the algorithm?
Little factors like this decide whether or not a player will click on the game to play. I wouldn’t go out of my way to play games with no one; what stops young players from doing the same? Most people play games with high concurrent numbers, which are known to be made of high quality and shown to retain players due to being really engaging and fun for their audience. Can you say the same for other low-quality, rushed games without looking at the CCU?
I understand this is a countermeasure to help assist experiences with no players to gain impressions, but it’s an interesting method to bring some sort of first impression to games without looking at how many people are playing.
I’m glad this is only an experiment, but please do not remove CCUs from Home tiles. They can be greatly helpful for sorting out experiences that are active from those that are not at a quick glance.
I do not want to have to hover over games to see their CCU count either, it is unnecessary and should not be made part of the final product.
This is COMPLETELY backwards. The CCU is extremely useful and removing it out of nowhere would make the homescreen 50x worse. Why does every UX “improvement” to the homescreen just get worse than the last?
The only thing differentiating games from their clones is CCU numbers. Removing them will make it much harder for players to discern which game is the legitimate one.
Firstly, thank you for communicating this in advance!!! This really helps with figuring out why various KPIs are suddenly nosediving or skyrocketing with seemingly no major game changes. More communication like this moving forward please! This is a great start.
Does this mean there will be less experiences shown on the home page? E.g. 6 experiences down to 3-4?
Are we still keeping the “Friends” section? I often want to see what my various friends are doing regardless of the context of a given experience. If that’s removed I’d be joining my friends less often / would have to explicitly search for them, which would be annoying as a user.
If the above questions could be answered it would clear up some of the confusion I think!
I don’t really agree here - you can quickly view the game’s CCUs on its details page, or from the creator dashboard. Can be slightly annoying yes but it’s a non-issue imo.
Regarding the issue of players joining a game with no players in it - I don’t really see that as an issue. If I’m looking at an experience, I’m probably interested enough to join it regardless. Seeing a low CCU can actually make a user not join, when they otherwise would’ve had a fun time in an experience that happens to have only 80 CCU.
Bring. back. genres.
For real, bring them back - looking for a new game in a specific genre is so hard with how the front page is set up, the weird Whatever made up algorithm based faux-genre sorting doesn’t even work properly (at least on the site, I can’t vouch for the mobile app or desktop app since I won’t bother touching those with a 10 mile stick).
Clearly genres are still supported by the platform’s APIs since they’re still around in a few places for no reason.
Bring back genres how they used to be, or better yet - like how other game distribution platforms/metaverse platforms have it, where you can search by defined genres and tag so people can find what they want and not what the platform thinks they want. (See: Steam)
Almost the entire developer community hates this. What are the analytics you’ve collected on the impact of this so far? There is no transparency here for the community to understand intention and impact. This is simply inconvenient to us.
Thanks for the transparency here, super excited to see what comes of these experiments. I think offering up thumbnail views of games will create a more ‘premium’ looking product for a lot of us… now I just need to stop delaying and actually update Bad Business’ thumbnails!
Curious how the friend activity is going to differ from the previous “Friend Activity” sort … unless this is just backtracking to a better system? I’d be okay with that.
And reduction in CCUs on display is interesting. I think it currently is being used as a quick quality-test, although like/dislike ratio and the new verification badge system could maybe be more accurate metrics for users to see.
Edit: I want to mention that I hope that these notifications of experiments happen more in the future, as well as following them up with data so we can see how further decisions are being made.
The previous “Friends Visiting” sort (removed a few months back) did not respect this privacy setting and was the main reason why I always refused to accept friend requests from anyone. I don’t want anyone to be able to figure out where I currently am; please make sure this respects our privacy.
One question, why? Developers spend a lot of time making a good icon for their game, and now its going to be tossed aside and replaced by a random snapshot? I’m aware that developers will be able to change these to whatever they want but currently thumbnails are often used to showcase updates or gameplay, while an icon serves a completely different role.
What if you want to have a video? Being forced to push it back will just result in less players seeing it, which only makes it harder to show what the game is like. This would also be especially annoying for browser users, as the video automatically plays itself when the thumbnail switches.
And also steals that attention from others. Developers will have to work twice as hard to make an interesting thumbnail due to games taking up much more space. I also want to point out that with this increased size, the amount of games displayed on a users continue playing page will drop from six individual games, to three! That is far too small a number for anyone who plays more then two games regularly.
Wasn’t this always a feature? I can clearly remember seeing pictures of my friends appearing under games they play. What makes this any different?
What? How are we supposed to know its popular if we can’t even see how many people are playing? This change just makes it even harder to know if a game is popular, and seeing at how roblox is a multiplayer game, I would argue that seeing how many people are actually playing before joining a server is extremely important, especially with several multiplayer games requiring more then one user to play properly. It’s like if fortnite or apex decided to hide their players, so you join a server and nobody is there, you just end up leaving because it’s pointless to wait around. And lets not forget that the popular page is literally sorted by active players, which just makes this change make even less sense in the first place. I’m extremely curious who prompted this change, since this is the first online game I’ve seen that intentionally wants to hide player counts.
Have you forgetten the backlash recieved from hiding players in the last update? You should be focusing on improving features like this, not outright removing them. I can’t think of a single person that would want this. Who would even benefit from this? It’s definitely not the developers and not a single player would want this.
Does removing CCU mean changes to the algorithm or is it just hidden to the users?
P.S. to other devs: they say you can still see it when hovering over, so it shouldn’t be a big deal for us devs who want to check CCU quickly. (Perhaps a bit more annoying though)
