Thank you for replying! I hope the algorithm didn’t mess me up, it feels like it did and maybe partly because of the fact I’m using reserved servers, maybe it messes up some of the signals. Some features I found already break due to the useage of reserves serves such as ReferredByPlayerId being 0 even if you were directed to an experience by your friend that’s in a reserved server. So I wouldn’t be surprised if more things are affected by it.
Hi @LootRPG , benchmarks are purely for informational purposes and do not affect the algorithm / impressions. They’re just there to show you what people who are playing your game are also playing.
Ths issue we’re describing is that this information was missing for some period of time (only for analytics reporting purposes), so during that time, we weren’t able to show the similar experiences data.
Thanks for explaining - would you be able to file dev forum bugs for these issues you’ve found?
That will help us to ensure that the correct teams are tracking them.
I have but it got no staff acknowledgement. Here is a youtube video if you’d like to forward it to the relevant staff (you can see the video is cut because its originally a gyazo capture)
Thanks could you link the bug you filed?
1 Like
By the way, do you think there is a chance that because of issues like this the algorithm has been unfavourable towards my game? Maybe some metrics like the co plays time not registering or something?
Thanks for sharing the bug report - will forward!
My guess is this won’t be significantly impactful on algo (totally agree that we should still fix bugs though).
1 Like
Is there some sort of Logic to the extreme movements of the impressions? Are analytic values being recalculated every hour? lol
(Game is “Monster Survival”)
Hi @Antiael, you can look at the distribution of your users in the Demographics tab.
I’d expect that if there are cyclical trends it’ll line up with when these users are concurrently online.
You can also filter by certain demographics on that tab, e.g. 18+ if you were to look the particular users from the 18+ segment winning thumbnail.
I don’t know what timezone your screenshot is in, but reminder that 4/18 was Good Friday so many people had the day off, and it’s unlikely that next weekend will perform as well if the experience is overall stable (aka not growing).
Hey all, now that it’s been 3 weeks since we launched the new Home Recommendation analytics, I’d love to hear any feedback you have on what actions you’ve taken as a result of these charts and how we can improve them.
If you have questions, we’re happy to answer as well.
All my games have had a 60-70% drop in CCU since the outage from about 5 hours ago (at the time of writing this). The status page mentions that the issue has been resolved but my CCUs on multiple games still haven’t recovered which is very concerning. Why did this happen and when should I expect them to go back to normal?
Here’s links to some of the affected games: Be A Blob - Roblox
Find the Keycards 🗝️ - Roblox
1% Win Glass Bridge Obby! - Roblox
Big & Tiny Size Obby [2 Player] - Roblox
4 Likes
I have seen something similar as well. Anyone else?! Like a 20% drop in all games since yesterday crash. Anyone else???
3 Likes
Someone already asked this but it wasn’t answered or missed. (Unless I missed something), but do all the signals have an equal weight for the algo? Or is their weight the order in that they are listed in the dashboard: Qptr, 7d playtime, 7d play days, 7 day co-play, spend days, spent?
1 Like
Easter holidays are over. (Dropped about 20% from Monday to Tuesday)
This site https://romonitorstats.com/ shows the total roblox CCU and normally when the roblox CCU is lower, your impressions will be lower too as it’s relative.
Yesterday’s CCU drops have nothing to do with Easter holidays. There was an outage which resulted in a change to how the algorithm distributes traffic to games which is why global CCUs are still the same.
Some games received an upwards spike in home rec impressions while others (like mine and many others) had their impressions reduced to fractions of what they otherwise would have been.
4 Likes
That’s interesting. I didn’t know that. If that is the reason, they should probably be a bit more transparent with the changes…
It doesn’t seem intentional, something probably went wrong or was accidentally changed during the outage
1 Like
