World Discovery is Terrible, Prevents Monetization and Discourages Effort
hdorriker
VRChat world discovery is a uncontrolled catastrophe. It makes the Creator Economy much less useful than it should be for all but a tiny handful of world creators which is so narrow in scope that doesn't even include some of VRChat's most prominent world and game makers because it is too efficient at sending too many players to too small a number of worlds only.
I want to be very clear that this is NOT a staff problem.
All VRChat staff I've ever encountered are doing their best to counteract as much of the effects of VRChat's ineffective world discovery system as they can. I see them encourage creators to use Creator Economy features, support creators through CE, and I've always seen Creator Economy staff doing their best to make the system work despite the terrible state of world discovery.
It is time to do something about world discovery!
There are clear steps which VRChat needs to take to make Creator Economy make more sense for worlds to adopt:
- Stop sending players to the same top 3 or 4 worlds.
It's time to take the "classics" off the menu. The prominence of these worlds is no longer justifiable as countless other worlds which are better optimized, better looking and more feature rich have been released since and have been held back by this de-facto policy that massively preferences being "first" above absolutely everything else. Worlds should be allowed to rise and fall on their merits, and the tops of the lists should be a bubbling fountain of variety.
- Act against exploits that deter world creators and extract value from the community.
The current system is extremely favorable to minimal "turn-key" worlds and "games" which monetize "gameplay" for a week or two and then are deleted after their attention curve has waned. These worlds are usually promoted by their creator spinning-up dozens (or hundreds) of instances (each instance adds one to the total player population for a minute or two even if it's the same player), which promotes it up the Hot and Popular listings, until a critical mass of unsuspecting players (or atrocity tourists) are flowing through to sustain the position. This sharply demotes earnest worlds of any level of effort or quality made by active community members. Active community members are also much more likely to re-spend VRChat credits on other content within VRChat than someone who is obviously only here to grab some easy cash and dip. Allowing this kind of exploit to continue cultivates a sense of platform decline and general ruin that makes it that much harder to justify any effort, monetized or not.
- Add MUCH More World Favorite Slots for All Players.
Many players have simply stopped using the favorites system entirely because it fills up so quickly, and must be constantly purged of blanked entries due to the high rate of world deletion (You might want to look into why that is...). The current limit of 400 means that players can only give out 400 "thumbs up" to worlds they like. Which is it? Is this a utility to bookmark worlds for later viewing, or is it a metric? It's treated like a metric, but doesn't mean anything (anyone can farm favorites). This particular part of the system is very obsolete and desperately needs to be overhauled.
Log In
Geneec
Monetization seems important to keep VRChat going. The platform is important to us (the users) and we want to see it succeed and continue to exist. But I'm not going to get into the arguments for or against that particular piece of this canny, because I don't think its central to the issue @hdorriker started with, discoverability is not great.
VRChat is hiring a software engineering position for discovery, I actually tossed my hat in the ring for it, but I think they're looking for someone with more expertise with like youtube or amazon style "algorithm" recommendations than strictly search engine kind of people. When I was originally interested in the position, I did some thinking on how I might go about it.
My advice to VRChat is, start at metrics, get a good basis for what actually makes a world good or desirable. Get an idea of how the world performs to the cohort. Then, move on to social analysis; determining virtual friend circles, world-to-world interest circles, then you can do recommendations. Breaking the feedback loops (consistently/always recommended worlds) will take a lot of work and tweaking, this is a long-term job.
My write up was over the 2500 character budget, so I just tossed the rest in a text file as an attachment. It goes over ideas like world quality analysis, friend-circles, tagging or world-circles, and post-visit surveys/ratings.
DarkSwordsman
I guess to get it out of the way: I appreciate the sentiment, but I think this feedback post isn't exactly as constructive as it can be. It doesn't really point out what specific problems are happening besides maybe the artificial boosting which I wasn't even aware of, and world favorites. It somewhat promotes solutions for problems that are unclear.
I can definitely agree that world favorites needs to just be one category with no limit. Favoriting is different from visiting, and I think it could be a metric used to promote or verify worlds. I'd argue that I'd like to search my favorites, and that worlds should be able to have more than 5 tags (like 50 tags really).
Besides that, I guess I'm not really sure what else could be done for discoverability.
From my experience, some of the most successful worlds are only that way because of the way they are used. For a world to grow, it's best done through friends of friends or to run events there. That would be a much more natural way for a world to become popular.
Otherwise, what else can be done? I guess I don't really see an issue with the current system like the variety tab, which refreshes every time you open it.
Lastly, and maybe most controversial:
I don't think monetization should be a major feature or goal. To me, this is just like Roblox when it started to go heavy on monetization.
I'm afraid of changes that would more directly encourage monetization. It would reduce the effectiveness of people just making worlds for fun or out of passion.
So really, the fact that this post has an underlying goal of monetization kind of sours it for me.
I definitely think people should be able to monetize their work in various ways. But I also think that if your goal is monetization and that fundamentally makes you not want to make worlds... why even use VRChat?
The charm of VRChat is that there isn't pressure to monetize. You can make whatever quality or level of world and just post it. Monetization would just be a benefit.
Otherwise you'd just get what is modern Roblox: A slop machine designed to take as much money from kids as possible.
hdorriker
DarkSwordsman I’m not sure how this can be more constructive. This is an issue many other world devs have brought up for years, VRChat staff is very aware of it but can only try to mitigate the effects manually to a certain extent understandably within what they are allowed to do. This issue needs to be brought to managerial attention, and there are currently no routes through which to do that.
The focus on CE here is because CE raises the stakes and is important to a lot of world devs. It is a part of the platform and I’m more interested in getting a dialogue started here that points towards a solution for the platform as it is, rather than as it may have been.
As for how this effects worlds made for fun or passion, I have a few of those, and I can tell you for certain this issue started before Creator Economy and also impacts worlds that aren’t monetized.
One of my worlds got impacted by a thematically similar, but almost empty patreon template world on release night in 2024. Whether or not it was intentional or coincidence is irrelevant to the effect it had (it was definitely a cheater). Were this another creator that actually put effort into their world and didnt bot their crap to the top of “new and noteworthy” or whatever, I would have held no hostility towards them and probably even reached out to talk shop. But this other world was getting thousands of visits and favorites per hour. Something far out of my league, but when I saw it, it was almost entirely empty, nothing but an almost featureless Unity terrain, and an incorrectly-configured driving prefab that wasn’t even drivable, and a patreon subscriber board.
It is unlikely that this world was actually getting thousands and thousands of real visitors and favorites per hour (it got like 100k visits in a few days) and it was obviously meant to get their patreon in front of players entrusting the world discovery system to guide them to something new, and they completely succeeded at this exploit undaunted. They never intended to make a real world for VRChat, they were simply using VRChat as a temporary billboard. And sure enough: today that player is gone. They got their cash and ran.
The current world discovery system is highly preferential towards this behavior which began prior to Creator Economy’s release, and continues because the broken world discovery system seems to be treated as sacrosanct instead of thrown out for its obvious failures.
「Cuddle」
World discoverability isn't as bad/dramatic as written in this canny. Discovery is actually in a pretty good place, more than it ever was. New worlds are pushed daily into the top rows of popular, active and more.
If you release a new world and people are interested in it, it WILL get boosted very hard into the New & Noteworthy tab and even the top rows of the popular tab. If your world did not get pushed through multiple tabs, I'm sorry to say, people were simply not interested, no matter how much effort you might have put into it, and I know, it doesn't feel good but people just did not enjoy it as much as you thought they would. Good worlds and worlds which people are interested in, enjoy being in, will stay in the popular Tab.
Imo, the new world boosting is kind of too high in some ways. It's actually actively killing new communities forming as they're getting continuously pushed down by new worlds pushed all the way to the first rows. If you don't have an active community who uses your world most of the day, it will basically start dying off directly once new worlds get pushed all the way up there, so if they would boost more worlds for a better discoverability, well, it will simply kill any new community trying to form, including yours, which is even worse as this is a community driven game.
As prismic mentioned below me, there is value showing where people are actively hanging out, it would not make sense to push worlds where people are not active in, interested in, etc. I understand that many long term players do not enjoy seeing the same worlds at the top, but many others including new players do, that's why they're popular to this day, if they didn't they wouldn't be there and would have dropped from those spots a long time ago.
Invertex
「Cuddle」
I do agree that OP's post wording is a little off the mark.
But in regard to what you're saying, even if a world is good and gets pushed to New & Noteworthy, that is generally a short-lived ranking that will be seen by the few who happen to be on during that boost, and that can only have lasting power if a world has reason to keep players coming back to it. This incentivizes strongly towards "hangout" worlds instead of cool experiences.
There needs to be some extra kinds of ranking systems, including a direct rating imo. And then tabs like "Hot (most active)", "Highest Rated (All Time/Past Year-Month-Week)" and maybe a most visited category. And then also provide a search menu to allow you to set parameters around that.).
「Cuddle」
Invertex
That boost actually lasts quite a bit and spans over multiple days, in which it also gets boosted into the 'Popular' Tab for all to see. If it's short lived, it simply means people were not interested in it, performance issues, world issues, or had no incentive to go back to it after visiting it.
Invertex
「Cuddle」 You're oversimplifying things. In practice it's not that simple, especially as more and more worlds are occupying that space and people become comfortable going to what shows up first in their lists as it can be annoying to go world searching and seeing bad world after bad world with no way to differentiate if it's not a currently "Popular" one.
There are tens of thousands of worlds and growing. And there are types of worlds that are good experiences, but not something people are going to be in a long time, that there might be only a few people checking out each day, assuming they could even know about it, simply due to design. Quality one-off experiences that you play and then move on from, there has to be a way to give those a way to be seen.
There is also the issue of world to player count ratio and herd mentality, making the issue worse as more and more quality worlds are made. Yes one world might be "better" and hence is where people are going, but it doesn't mean another world is significantly worse such that it should have limited visibility that then just results in people heading towards what they see is the most popular, as they HAVE NO OTHER METRIC to judge it by before checking it out.
Hence the need for some sort of rating system and filtering options, and separate tabs for viewing based on those separate things other than just straight "Recent popularity".
I would suggest watching this video if you haven't yet and considering what is being said in detail: https://www.youtube.com/watch?v=iv9rg-YuyAg
Prismic247
I absolutely agree with the end goal of better world discoverability, but not necessarily all of the specific objectives given above.
- I believe there is value in showing where people are actively hanging out, so I wouldn't say "stop sending people there" outright. That said, it is certainly a feedback loop, where people go to the active places, which helps keep them active. Rather, keep a section for "Active" as it is, but have other sections that allow worlds to rise and fall, alongside other metrics than simply "how many people have been in the world over some span of time" to go off of. A system to rate worlds for example but it could greatly help in providing additional criteria to consider, and doesn't have to be an end-all be-all.
- No set of criteria is ungamable, and any changes would just mean a new paradigm. At the very least, I wouldn't consider any of the most popular worlds as low effort, or having gotten there through ill or unhealthy means. They are popular because they do provide value to at least enough people to warrant it, and continue to do so. It's not "bad" for something to be popular for a long time. But I do think that lumping all different kinds of worlds into a one-dimensional popularity contest isn't a fair way to slice it, since a visual exploration world will never compete with a club for staying power. Perhaps we need more explicit categories for worlds, with a degree of enforcement as well, like there are for worlds in the Game and Avatar categories? Categories for Clubs, Exploration, Story, Scenic, Sleep, etc.
- I do agree with this one, but alternatively even if we just had a way to rate worlds and see the list of them, and a way to at least see if we've been to a world before, would help a lot.
Ultimately in a dream world I would love a way to choose which categories show up on our discovery feed. Popular worlds are great for newcomers to find people and see what's good, but allow long time players to sort the categories, choose which ones shows up, have custom filters for tags and such like Steam lets you do with the dynamic games collection.
Invertex
Curious if this feedback was spurred on by what was talked about in the last Dev Update video haha. I expressed similar concerns on that video.
They did say they are looking into having a better system for this, but that they want to avoid it being easily exploitable and it pushing people towards making specific world types.
Unfortunately, the current system already is very exploitable and heavily pushes creators to just make "club" type worlds, spaces to hang out in, because the strongest metric for world visibility is active player count.
If you make a short, one-off experience, few people will see it since enough people if any aren't staying in it for long enough. It's treated as a bad thing, even though the intent of the world is to let them have a fun experience, not keep them there.
Same can go for worlds where people might want to pop into once in a while that might have progression systems too. It could be a really fun world, but with there being no proper ranking system, it will just sit in purgatory unless a large group is active.
We need some sort of ranking system at the least, and then the default world's menu being a mix of rising, currently active, highly-rated, frequently played and maybe a few other metrics.
Having a ranking system would also greatly aid world-testing, allowing people to more easily avoid worlds that have already been downvoted more than enough, able to better spend their time on other worlds that need attention.
hdorriker
Invertex it was spurred on by the “Success Stories”. The juxtaposition of encouragement and the reality of a system that prefers cheating was a bit Kafkaesque.
VRChat’s world discovery has been sending a message by picking empty billboards for patreons over higher effort worlds for years now. It’s even worse if you’re new to all of it.
Despite the efforts of staff to try to encourage earnest creators, the system currently does not.
Patroll
I'm completely discouraged by the current world discovery inside VRC as well and pretty much only bookmark cool worlds that popup on Twitter profiles of creators and photographers, which is not the most optimal. The world favourites limit is so bad that for the past years I've only been removing the top random worlds from my faves to free just enough space to add something for later, it's a dead feature to me (it's especially infuriating when the VRC website throws silent Too Many Requests responses when I try to remove a world fav and add a new one, it's very aggressive for no reason). Overall, it's just demotivating to see so many amazing worlds not getting the spotlight they deserve and just getting buried and I can't even save them for later for myself.
Maebbie
I absolutely agree with your post, the algo showing 4 to 5 year old worlds as "Trending" or "Hot" is very demoralizing for any ambitious creator. Imagine going on a site like reddit and the front page would show some 4 year old post.
A clean solution specifically would be an Exponential Regression Curve whose weight starts at 1 and goes to 0 over the course of 1 year and serve as a multiplier for the heat level in those categories.
I have made a very similar post 3 years ago with 200+ votes, which unfortunately had little impact on the algo. https://feedback.vrchat.com/feature-requests/p/1236-trendinghot-section-is-hurting-ambitious-worlds . I have personally benefited from this current algo and many who upvoted it back then have also.
evalka
The problems described in this post are exactly why I stepped down from world creation; it was so, so discouraging and growth/monetization was nearly impossible despite the heart I would put into and the praise my content would receive.
If VRChat wants to continue creating the unique experiences it advertises and being a place for creatives to share their hearts with the world, this post should be read, understood and acted on.