Roblox Discoverability Algorithms

[Sinne_J (Sinne)]

This post is based on my own information from experiences of releasing multiple games, feel free to add your experience of getting your games discovered and their path to success or doom because of the algorithm.

Currently, nearly every new game depends solely on getting recognized by the algorithm. This is mostly beyond the developer’s control, as there is little to no information on what we can do to be ‘liked’ by the algorithm. The only thing we can do, as developers, is simply make the best game we can and make sure it is enjoyable.

The problem as I understand it, is that currently a game is either on the algorithm or off the algorithm, with no in-between. It is either getting a very significant boost in traction, reaching thousands of players, or it is struggling to gain even 100 concurrent users. There are 2 games I would like to use as an example:

1. Sword Haven: I have already made a post about this as a ‘bug report’, but as Roblox staff said, and I agree it belongs in feature requests better. To summarize; the game’s performance seemed completely randomized from the very beginning, regardless of what we did. I believe it was just the game constantly getting on and off the algorithm, but struggling to stay on for more than a day. It should not be this inconsistent.

2. Yeet a Friend: This is a more positive one, but nevertheless a good example of just how much impact the algorithm can have. After releasing, for the first week we were just simply sponsoring the game, no marketing outside of Roblox sponsors. Absolutely no growth for that week; averaging a few hundred CCU. After that I do not need to explain what happened:
   

   

---

---

---

---

I would love to see the algorithm being more consistent. Let games grow organically, allow them a small traction boost to see how players behave in the game and how metrics adjust. Let the algorithm adjust accordingly and smoothly. It can be really disheartening and demotivating knowing that your game’s future depends so much on whether or not it will hit the algorithm and stay there. Using my game Sword Haven as an example, we were all euphoric and excited seeing it go from just a few hundred CCU to 4000+, then confused and heartbroken to oblivion when not even 1 day later it went below what it even had prior to the algorithm boost.

I would also like to mention that (from my experience), the algorithm was significantly better in the past. I have had 4 other games in the past that grew steadily and organically to thousands of concurrent users. The latest game I had that grew organically was released in early 2021. All other games in late 2021 and beyond have this weird on-and-off the algorithm pattern.

TL;DR: Experience growth is extremely inconsistent, and the algorithm should be adjusted to allow for more organic growth rather than praying and hoping your experienced will be noticed by the algorithm.

Please share your experiences of releasing & maintaining games and how they have performed, whether you have also experienced such abnormalities or not!

[MagmaBurnsV]

In conjunction to discoverability, there is a severe lacking of tooling for developers and players alike to utilize in game searching, making surfacing of specific games a player may be interested in extremely difficult.

The addition of Exact Match Search seemed like a step in the right direction for player querying, but unfortunately its design is abysmally useless as it requires direct knowledge of the syntax (Surround the search in quotations).

I think what needs to be added is proper game search options that is both obvious and easily accessible for players. These options could be exposed on a dedicated search page that can be accessed via a meatball menu and would include precise game search options for fine tuning results.

These aforementioned settings could contain, but not be limited to:

• Number of Active Players, Likes, Visits and Favorites
• Like Ratio
• Age Rating
• Server Size
• Creation & Updated Data
• Friends Played/Liked/Favorited
• By Users or Groups
• GENRE (please, please, please add them back)

I have a feeling Roblox is hesitant to change the overall algorithm because of how delicate it is and how significant a change could be, which is reasonable. But if Roblox were to add better search filtering, nothing of general search would be changed, just extended.

[OutlookG (Outlook)]

Feel free to add your opinions/concerns here

Roblox’s Algorithm Changes Are Killing Our Game

Introduction I run a team of 4 full time developers and a rotating team of part-time contractors called Ruddev Media. We’ve been responsible for many projects throughout the years but the project that our team has depended on over the last 4 years has been Bad Business. This is an FPS game that has 390 million visits, been featured in platform-wide events (Egg Hunt 2021, Ready Player Two), and was initially funded by Roblox in part of the Accelerator Program. Starting sometime in mid-late Apri…

[Sinne_J (Sinne)]

@MagmaBurnsV

I have a feeling Roblox is hesitant to change the overall algorithm because of how delicate it is and how significant a change could be, which is reasonable. But if Roblox were to add better search filtering, nothing of general search would be changed, just extended.

With all of the recent complaints, it is likely that it is being changed and indeed having significant effects, they are just taking every step in the wrong direction and continuously making the algorithm worse. From my personal experience it was A LOT better 1-2 years ago, games seemed to grow organically and steadily, and they also didn’t just very suddenly begin flatlining after doing so well.

I agree with your other points & suggestions, its absolutely crazy how there are literally 0 options for SEARCHING on a whole giant PLATFORM?! One of the most important aspects of the whole platform and it is completely neglected…

[planetville]

I agree with basically all the points on improving the algorithm.

However I just wanted to add that for your second example, there does seem to be a certain level of predictability on when the algorithm will “pick up the game”.

The results shown on the tweet seems to be consistent with the time your game took to be indexed by the algorithm.

[M1Jy (0idk)]

I would like to bring to attention that I am facing some difficulties with the visibility of my game, “Bum Showdown (Debate Game),” on Roblox’s discovery feature. Despite reuploading the game to the group, it does not appear in the search results. Furthermore, I have encountered another issue where the search bar displays irrelevant results when searching for random words. For instance, when searching for “sus,” Blox Fruits is shown instead. Similarly, searching for “priv” inexplicably leads to a game titled “[FREE ADMIN]”, pineapple gives for no reason “piggy” games… In the settings age was 13+, poor kids should break their psyche when trying to search for the game on the tablet and other devices?

[otterryy (tj_demon46)]

ye its insane i remember years back you actually use to be able to filter games by its genre u cant even do that anymore idk y roblox has no features like that. they want big games to stay big small games to stay small. if u take a step back, all the games on the front page are the only “quality” games. quick search of anything and u get a list of garbage after the first 5 results