Actions speak louder than words. I’m not getting my hopes up until I see change.
This is very interesting. Hopefully it helps quality games get more recognition and attention!
its like spotify and facebook when u first join i think people will fill it out
and if u think about it, as long as i can remember, roblox has NERVER asked for your preferences they even removed the option to filter games, so i think people would like it
This is so exciting to hear!! Great roadmap and love the thoughts dump, appreciate the clarity and communication!
As with some other comments I would also agree that there should be a way to “reset” audience impression data or postpone the initial scan, as I personally have old projects which are very alpha and don’t have much content but I want to develop.
And please bring back the genres feature, it made it so much easier to find specific games.
Awesome! Let’s hope there will be a future where a game with a good idea, quality and effort put into will automatically become popular without any external promotion.
I remember already searching with quotes a year ago and it worked back then, so it may not be new, but it is always a great thing to have, especially if someone wants to look up your game and it isn’t listed within the results without using quotes.
I had 7k ccu and upates did nothing. We had a steadily decreasing player-count. I am a believer that my game will be filtered out of the algorithm regardless “to keep the platform fresh and exciting”. It’s incredibly difficult to keep a sustainable experience on this platform.
Ad purchases with the new ads manager have been ineffective for my new releases. 20 robux per play was an outrageous spend to reach an initial audience.
LifeTogether is a new hit. If games like that can have a sustained playercount and simple initial discovery I am happy.
I don’t know what to think about an algorithm which keeps recommending games not enabled on phones to a mobile player.
Personally, I really think a play-later sort should be added. I see tons of good looking games all the time, but rarely have the time to play them. So I favourite them all on site so I can find them again when I have time to play.
This is not ideal for two three reasons.
A) I had to sacrifice my favourites sort for this
B) It is linear and previously favourited games may become hidden behind new ones.
C) The favourites sort is currently located at the very bottom of the page and not convenient to access.
The addition of a play-later sort would
A) allow me to use the favourites sort as intended
B) quickly find games that caught my interest in the past to play when I do get the time
A play-later sort should have/be:
A) quick and easy to add games to and remove games from
B) have an option to sort by most recent
C) have an option to sort by random (so older ones can be shown instead of only the most recent)
D) optional algorithmic filter to allow the algorithm to predict which from the list one would be most likely to enjoy (last minute suggestion and the least important)
It’s been a bit so I’d like to share my current thoughts…
When I think of communication, I think of someone responding to something and that other person responding back. Most communication on Discovery has not been like this.
What usually happens is that Roblox makes a statement, we respond, and then don’t hear back. Exactly like this thread, there are no staff responses to what people are saying. Or, if we hear back, usually the response is to the entire community, not a specific idea from a specific person.
Roblox staff should respond to the people specifically and what they are saying specifically. Respectfully, the forum has accounts and replying for a reason. Please use it.
The AMA seems to be the best way to do this. And so I think that’s the hiccup for not responding right now in this thread. But, that can’t be said for those other years and threads.
I’ve read through this post multiple times now. As a seasoned developer on roblox whose been making games on the platform for a living for multiple years now, it feels like its getting harder to actually become successful by making quality games on the platform.
Despite everything you’ve mentioned here regarding retention, engagement and monetization, it doesn’t seem like it’s helping good quality games to be shown to the appropriate audience.
Here is my front page:
This doesn’t look like roblox is promoting good quality and enjoyable games as they claimed
(Note; roblox claims they curate what games are shown to their appropriate audience, yet, I would never even think of playing any of these games)
Now let’s take into consideration our game RoVille, which has been in development for many years. In the past, we’ve had a decent amount of traffic coming from roblox’s algorithm, but recently, the algo seems to be killing off our game:
As you can see, our game is getting less and less new users over the past few months, despite constant updates. Most of our playerbase nowadays (around 70%) is made up of returning users, because roblox just isn’t showing our game to new potential users as much anymore.
You can see this even more clearly on a longer term graph:
All the while, our like ratio has been going up, and engagement, retention and monetization stats remain constantly above average throughout the same period:
As a developer, it sucks to see that a game which players clearly enjoy, judging by metrics and overall feedback, is getting slowly killed of by the algorithm, despite constant updates from the developers.
Roblox’s discovery still has a long way to go, and as a passionate developer, I can only hope that this improves in the future, for the sake of the developers and players alike that make Roblox what it is today.
Frankly, I don’t see why Roblox should be promoting any of the popular games, more obviously when AdoptMe… MM2… etc are all so popular already. It’s natural they make money from it, but it’s almost like they are getting a bigger cut lol
Either way, discovery is VERY limited, that Roblox does not give players a more flexible option, like a drop down menu, or a side menu, for choosing the games by genre or tags… etc … as it would help players.
Only options given are:
Continue - no brainer
Recommended Games - by Roblox
Sponsored - People promoting their games
Favourites - Players saved games
I guess the “Discover” Tab is more spread out & promotional using stats, but again… mainly promotes the TOP games… not by genre’s.
Up & Coming there does probably the best advertising for new games, but the rest are successful games already.
Though the search helps find certain games more precisely since they fixed it… it still is not perfect… but guess we’ll give it time.
I could be wrong, but a dropdown option might help discovery a little, more to players/fans liking.
Hi @Reditect, thanks for sharing. Rest assured that the team is reading all your feedback!!
There is currently a large range of specific ideas / questions that people have about Discovery surfaces and we’ll try to find a good way to address them.
We appreciate your patience as we improve our processes for working together with the community
Thank you for providing us with your feedback! We appreciate it and are reviewing your questions and comments. We will address them in future AMAs and posts.
Is there any reason why we have Skyscraper ads inside of the message inbox now?
I feel that this Skyscraper AD should not be there at all, since there is one banner at the top.
Not sure how to crosspost. Relevant - Roblox’s Algorithm Changes Are Killing Our Game - #348 by BuildIntoGames
Thanks for the transparency! It’s really what we want to know. Looking forward to the follow up posts
The change to the search algorithm has killed off my game. Even if I search the exact name of my game, if I don’t use exact quotations, no one will find my game. The story game revolves around a niche meme, so 99% of the players had searched up the exact name of the meme to find my game, but now only irrelevant results pop up.
~250 searches to zero
I’ve had a varied experience of the discoverability of my games on the platform. I used to have 2 games in the top 10 returned results when searching by name. That changed approx. 18 months ago and noticed a distinct drop in users.
Since this update however, we have doubled our CCU purely from the games being featured in the Recommended list.
I can also say that the titles presented to myself in Recommended when I am on the Home page are so much more varied recently. I enjoy discovering the new content seeing what everyone else has made on the platform.
I welcome the new changes and say thank you for listening to us smaller devs and allowing us to be showcased to the player base.
Our new game, Ultimate Dodgeball, is not getting picked up by the recommendation algorithm. We cannot recreate any of the examples provided above.
We have so far sponsored and reached nearly 4 million impressions (over a course of two weeks), and have been constantly making iterative updates and changes to improve onboarding and the overall experience of the game. Despite all this, we have not seen any Home Recommendations whatsoever.
I don’t understand what further changes we have to make to our workflow, or what we’re doing wrong. It’s demotivating seeing no home recommendations despite our continued efforts. I already have a game, Chaos at the Bistro which consistently gets home recommendations, despite seemingly below average session time and moderately average monetization statistics. Ultimate Dodgeball has been doing better in these terms so far, and yet we still see no change.
Any help is appreciated. We’ve been trapped at this hurdle for a long time.