One more thing to add is sponsoring the game may also lead to a decrease in discovery. Therefore, I stopped sponsoring my game, and I saw the discovery coming back up.
Care to elaborate? I’ve found sponsoring a massive waste to drive visits in the first place, so it would make it extra funny if it’s also tanking discovery.
That’s odd,
shouldn’t it be the exact opposite?
That’s just plain weird and shouldn’t work like that from how its supposed to work.
Actually it IS relevant because players choose what to play depending on quality they deem good or bad.
Some people play games purely for being terrible like most obby games and tycoons which are minimal effort.
You can see a clear bias towards certain genre of games just by looking at the front page so to dismiss this as if it’s irrelevant to the subject is stupidity.
It is clear that certain demographics require different experiences which also means the playercount for said demographic will vary wildly.
And why should games with better metric be promoted while the ones with lesser metrics shouldn’t?
that just causes the whole issue of a set of games dominating the front page again and leads to less playtime on newer experiences by smaller devs and studios.
What I see here is people wanting their own games to do good and others to be of lesser metrics which doesn’t solve the still apparent issue of ROBLOX not having a good discovery page system.
What is likely is that either your demographic is too small in comparison to other genre demographics or the System itself it messing with your games or maybe even both or none of these.
All you can do is speak out and hope ROBLOX tweaks that system to make it more fair.
Just don’t put other’s stuff down simply because you value your own more because ROBLOX likely values it in a different light from your perspective.
Here are my daily visits since January 1st. On March 1st I decided to try regular biweekly Friday updates. You can see how good that was for my playerbase. After the algorithm change, ~May 1st, I simply wasnt getting the players, and despite regular updates since, Im not getting players, even despite school being out.
Where is the bottom for my game? Is it just destined to get to zero? This game is a big chunk of what pays my bills.
It’s like ROBLOX is putting on a cap on total impression or something. The home discovery is climbing back to normal when I stop sponsoring, but yeah sponsoring 5k a day wasn’t worth it at all. There were more “search + friends + others” than discovery new players which were not supposed to be. The game was always having more discovery new users than the others combined.
Super demoralizing to keep working on my game when stuff like this is happening even when school is out. As a comparison, from last year’s summer to this year:
I am so grateful to have a game I made on my own reach these stats but I find it hard to be motivated to work on updates when it feels like my game is only dropping in players no matter what I do.
Whoever is struggling with the algo changes, you are not alone !!
I love when the algorithm promotes me scam games
for all of u small devs, just stop making big projects and jump on trends, this is how the algorithm works now unfortunately and I don’t think It’s gonna change any time soon
ROBLOX’s response to this thread was just “skill issue get better”
I was thinking on starting to work on a big project recently but if this is how the algorithm is gonna stay then I’m just not gonna do anything
The simplest solution is to reintroduce the Friends Activity section. Almost all of these issues can be attributed to this once instance.
I give Roblox the benefit of the doubt because you guys have all the data to make these kinds of changes, but I can’t imagine this hurts the platform in any way.
A lot of small games depend on this, and it’s a fantastic way for players to find new games.
Our team recently ran a 28-day long, 100,000 Robux/day sponsor campaign, and here are the results:
(There are multiple weekends where we sponsored an extra 100K on updates in May, but no difference is seen)
How can this possibly be acceptable? Yes on some days the sponsors are more competitive & so the impressions get split between more games. But from June 10th (when we started the 28-day campaign) there is a clear and obvious gradual decline in impressions.
Why would the impressions from the same amount of spending on the same campaign be cut down in HALF over 3-4 weeks? This effect on impressions is of course directly translated into new users:
Sponsors are already hardly worth it and developers constantly complain about it. Now they are half of what they used to be for us, why would we possibly continue sponsoring?
To add on, I have made made a couple of other reports on discovery issues in the past:
These issues mentioned by 100s of developers, small and big, need to be taken way more seriously. Please Roblox.
This issue is still happening and the lack of a fix is very insulting to us as developers.
We’re following the TOS by not scamming users with fake advertisements of Robux, yet WE are the ones being demoted compared to these games??
I’m sorry, but unless you are lying about what stats influence the algorithm there is no way these games should be beating us right now. We have over DOUBLE their playtime, yet they have DOUBLE the users then us? How does that even begin to make sense.
This is genuinely insulting and feels like we are being put down for creating a better game then our competitors, yet they get promoted even though they make less money and are activly breaking TOS by scamming and lying. That isn’t how it works in the real world, if someone makes a better product, their competitors die, why isn’t that happening here?
Fix this and revert to the prior algorithm.
I’m pretty sure they’re trying to give a chance to smaller games but still i dont think those icons (the ones in the screenshot you attached) should be allowed
This algorithm change will effectively kill my game. A game that has been trucking along fine for 12 years until this May.
2 weeks before this change I had an average of 2116 new PC players. These past 2 weeks I got 737 new PC players per day, with the number still reducing.
The majority of my income comes from this game, and its being actively and quickly killed. Updates that would regularly bring up the playercount do next to nothing, all I can do is watch as my game and income withers away faster than I could have ever predicted.
A game cant survive if it doesn’t get new players.
What am I supposed to do now?
I’ve only come across this today and now just fully understood what caused our average player count to take a sudden deep dive starting from May, this had caused me struggles and tremendous stress for the past few months since our whole year plan has been completely ruined and had a lot of difficulties regularly paying our contractors due to our revenue being heavily affected by this new change, to tell you the truth it had decreased our revenue by 50% since May.
As a small niche game that has been regularly receiving updates since 2015, this new algorithm will definitely kill a lot of older well-established games, I hope ROBLOX considers change that will not hurt games like ours while still being able to help newer games receive their new visitors and recognition.
Here’s the data for our experience, Wings of Glory.
For anyone interested, we have since doubled our E2E and improved other metrics. Our new users shot up, which is great. We now have:
- Top 1% Day 1, Day 7, Day 30 Retention
- Top 1% Day 1, Day 7, Day 30 Stickiness
- High avg playtime
- High AARPU
- High E2E
- Literally all of our stats except CVR are high above baseline
Despite this, our impressions fell to a new low yesterday.
This is what a new account sees:
On Desktop, we are 6 categories down. This placement really matters. We are effectively being punished for our genre.
On mobile, I scrolled the home page for a few minutes. I did not see Pet Sim X AT ALL. I had several people verify this. We we’re always the 2nd-3rd game recommended. Now we seem to be shadow banned?
During a call with a Roblox engineer about these algo changes, the #1 recommended game on our home page was a Pet Sim X clone. It was literally our game, stolen, with an identical icon. Pet Sim X wasn’t showing at all.
Hoping things get better.
I don’t think it’s fair to have the same game on every day for the entire existence of roblox…
One of the drawbacks of Roblox algorithms is that they initially promote someone’s game for the discovery page, but eventually ban it after several months/weeks…
That’s not the point, it’s the fact that Roblox made a change which consequently affected a large proportion of games, some which are well known and huge.
They propose that these games try to improve their metrics to fix this but despite doing so nothing comes out of it, so either Roblox is lying or they themselves do not have a handle on this.
Nobody is arguing whether a game should be promoted everyday just because it has been there to begin with.
The majority of developers present here are those whose games have enjoyed long-standing success or had a big boom in the number of players in a particular month… Since they are no longer earning significant revenue through DevEx as they did previously, they have chosen to express their grievances here. They wonder why their game from the period of 2018-2020 is not prominently featured on the front page, like Brookhaven or similar titles.
P.S: as it turns out some people don’t know how the game-matching algorithm works.
I disagree.
My game was seeing major success right before these changes.
The second these algorithm changes came through, my impressions dropped by 90% and my players, over a few months, dropped from 16k ccu to 500. Even though my stats (playtime & retention) are WAY higher now than they were during the peak. Monetization, obviously, is the only thing that has gone down.
Impressions are everything.
Do these factors have identical weight? Could you explain to us if one or two of these factors is actually quite a lot more important than the others?
If my game is on the upper benchmark of all these metrics except Retention, would it be discovered at the same rate as if it had all upper benchmark metrics except for, say, Relevance?