Sudden drop in number of daily installs on Google Play Store

both ads and IAP.

 

For me it's a free puzzle game with only Rewarded Ads, no IAP.
Drop from ~300 daily downloads to ~10 !

 

Ok, going to reign this in a bit, while it is ok to discuss this topic, bear in mind this forum for Game Development and Unity in general. So, please no links to external sites or "campaigning" for change, calls for "action" or "spreading the word" or anything of that nature. Not over the top ranting, not repeated spamming the same comments. There are places on the internet for those things, not here. Here you can discuss how it is affecting you and discuss things you as a game developer can do to get better results. (which does not include social media campaigning). Those posts have been removed.

Some Notes:
1. It appears from the responses that people are getting that it isn't a problem to google, it is a change in their matching and other algorithms. So, don't expect any changes (at least until they change it again). Also, transparency for ranking and matching will never happen, it is not good for anyone. It is important to not expose them to avoid gaming, it is that way for all markets/search engines/aggregators.

2. I was curious about monetization to see if there was a pattern favoring IAP/Premium vs. ads only. It would suck, but it would be understandable, as Google stands to make more money that way. But from the few responses it doesn't seem to be the case.

3. Organics are nice and can be profitable, but remember they are largely out of your control and Google (and every other store) makes no promises or guarantees and is under no obligation to promote your game in any way. If you aren't advertising, and you have had a drop in installs... sounds like your wake up call. Advertising, marketing and promotion are critical to making (and sustaining) a successful game. Other than providing the best presentation you can on the store's platform, the store is not marketing/advertising for anyone other than the store itself.

4. Everything in tech and the tech market changes ALL THE TIME. Adjust. Writing angry letters to google won't make a difference, they are looking at a very different big picture than we do, they have different goals. The best way forward is to adapt. Update your games / listings. Increase your ad-spend now, while everyone else is figuring out what to do, take action.

---

Again, discussion and ways to move forward are what this thread (and the forums in general) are for. Ranting and pitchforks and torches are not. Let's keep it all grown up and stuff.

 

I think this is super-important and if this is not solved, Unity will also be hurt hard, so can you please let us post links here?

 

I will leave that one, no more. This isn’t a unity issue it a market place issue, and should be brought up through those channels. It really doesn’t affect unity and is nothing that unity has any say or influence on.

Edit: those links were just echo chambers, quoting posts from here and pointing back here, and mostly the same people involved. So no. Links from official sources are fine, but not blog/crossposting.

 

Can you please recommend some channels of advertising? All kinds of advertisements I found cost something around couple of bucks per install. My game generates around 3 cents per install. How can I advertize and be profitable without organic traffic?

 

why my installs are decreasing ?

 

I see that I am not only one who has this problem. I have 3 accounts in different countries and in all of them I got the same problem. This is just example one of my games from 3k to 40 installs per day. I thought I something missed up, I need to update or change in my games. But I see a lot of developers has the same problem. My revenue in all games falls down by 90%...

 

When you look at the "Similar Apps" category, it really does look like a drunken mess of app picks. I could only guess that a monkey would do a far better job taking a handful of darts, and tossing them at any random apps on the store, and the resulting algorithm would be superior to this one. It is a hot mess.

I have no idea what goal they have in mind with this change, but it really does look like a glitch when you really dig in deeper... I just can't even imagine this being an intentional change.

I just hope they're scrambling to fix it, so this nightmare can be over.

 

It really depends on the game and audience. For example yt influencers work fantastic on one of our games, yet not on others. Facebook will get you installs, but often have poor retention. One of the best methods is spread it around a little and see works best for your game, then ramp up the ones that work. Podcasts can be a good route if you can find ones that match your game. I’m no expert by any means, but we have someone on our team who is. That is the best solution, find someone who knows their way around that area.

 

I get that your games were not affected, so hurrah to that, but could you at least wait with the preaching until this thing will be somewhat sorted out, because to someone who's been tracking this issue and talking with other devs for a few days now, you seem to be quite detached from what's happening and don't have much empathy.

1. Algorithm wasn't just changed, it got broken. That is an issue, not the change itself. Similar app listings are broken, recommendations are broken, category listings are not being properly populated (or are empty in some regions). Those are serious issues with serious impact to a big number of developers.

2. There is no pattern, because things are broken. Games of all sizes, monetizations, target APIs, countries and genres are affected.

3. Suggesting that indies, who are losing their minds about maybe going bankrupt to go for paid advertising. Classy and great timing! Not every game needs paid advertising, not every game can be profitable with it either.

4. Partially agree, because I don't subscribe to causing panic, wild speculations or anger, but adapting to broken state of things is also not the greatest course of action. There will be time for adaptation, when this is going to be over and the store will back to normal (maybe with new algorithm) state. Although, this is a decent time to boost above the competition, I guess.

 

I do not see any channel or group when I click on this link.

 

There is something I don't understand... People is saying that the"similar apps" algorithm changed (or it's broken), and it's the root cause of the downloads drop. My downloads dropped 50%, and some developers lost 90%... I feel it's not possible that 90% of downloads come from "similar apps" section for an app... Don't you?

 

@pep_dj - it does seem very far fetched that the similar apps section would be pushing 90% of traffic that some people lost as a result of that change. Really not sure at this point what to conclude regarding that, but it is very clear that organic keyword searches are only a small % of the drive for installs. There are multiple factors, and the similar apps section was seemingly a big one.

I did speak to a rep, and he said he's still keeping me in the loop as to what google says about this. As of current, they are still "investigating" to see if this is a bug. We have no clear-cut answers yet, so before further panic ensues, let's wait and see what they have to say.

Hang tight everyone... the answer is still not definitive just yet.

 

I am not preaching, this forum is about developing and helping out. FUD, hyperbole and freaking out doesn't solve anything or anyone. My games don't seem to have changed, and neither does anyone or other companies that I know (and the folks I know at google are unaware of anything odd, one of whom is in analytics). I am not saying it isn't happening to some developers/apps, but it's really only a handful of folks here, (several who registered in the last couple of days presumably from following links), and a couple threads on reddit which are echo chambers of only a few people. Even searching for information on pretty much just brings up this thread. There may be other factors that the handful of developers who are experiencing share, that is why I was asking questions. Google may be doing some weeding/pruning, or it may be that a derivative effect. Who knows.

But right now, hyperbole/ranting/rallying/echoes is not something for this forum. So no more of that, or we'll just shut this down as noise. Offering suggestions on improving your presence/marketing is a good thing and encouraged.

Updates from google and official sources as to the state of things, and of updates to your situation are all good. "Let's spread the word" type stuff is not.

You used the word "paid", I did not. Paid advertising is only a small part of marketing and promotion, and not the most effective.

 

Hello,

Error or not, in a game of 80k-90k daily installs down to <5k in 2 days from June 20th.

 

@zombiegorilla - I will politely correct you in your statement saying that "it doens't seem to be happening to a lot of people."

I've spoken to lots of developers, affected and unaffected. I also study the appstore charts like no other, and I can tell you that the appstore rankings are a huge mess, as if they've been turned upside down. It's impacted a LOT more developers than are chiming in on the threads. Just because we don't see a tens of hundreds of threads or comments on the topic, doesn't mean a sea of developers aren't impacted.

The charts feature apps now in the top rankings with poor ratings, low installs, and vary so widely, that it literally makes no logical sense. I know this, because I have studied my categories very carefully, and do so on a regular basis.

I've seen certain categories on the appstore totally turned upside down with completely different apps in the top rankings, trending sections, similar apps section etc. So it's definitely a widespread thing. I don't want anyone to think that this isn't a minimal issue.

I also understand and respect your statement regarding hyperbole, and FUD, and keeping things in perspective. We don't have answers yet. But as soon as I hear anything, I will post it up here.

If it's true, and you are saying analytics is unaware of any recent changes, then I tend to think that this is truly a catastrophic bug, and they will have fixes in place. Otherwise, I am truly at a loss for answers.

 

1. For me it's not "handful of folks" but almost everyone I know in gamedev. I don't know anyone from Google though.

 

It applies to me too - a steady 2k installs/day for the last year dropped to under 100 over the course of a couple of days last week. Ad revenue down comparably. Note that this isn't with a re-skin game, it's a former google play front page feature that's incredibly niche. Typically 90% of installs in USA and Russia, and those numbers do seem to be holding percentage-wise.

 

Just to be clear, google analytics has thousands of employees. I was on campus the other day for lunch with him and brought it up (we were mostly talking about early access). He hadn’t heard anything about it. Someone may know about it, but if it were a crippling problem or massive impact, he would have probably heard. But they update the features and algorithms very often. And they carefully and deeply track patterns and data closely (that is pretty much all they are). It’s a complete guess on my part, but I would think they are very aware of what is happening and either it is working as planned or the data is expanding to normalize over time. Ideally the latter.

Also, to clarify, when I say handful, I am referring to those who are posting about it. And if there are thousands who are impacted who aren't voicing themselves, that still isn't a lot. Consider that on Play, there are well over 2,000 new games released a day. A day. That is what I meant by the echo chamber (of social media). If you have a hundred people posting and reading each others comments, it may seem way more dramatic than it is. Still I'm not saying it's not a real or potential problem, but if were hitting everyone equally, or even a large percentage of developers, the voices would be in closer to the being in the millions, not hundreds or thousands.

I get that it must be frustrating, and not denying it may be happening. But it is clearly not massively wide spread. It might be useful for those affected to try to find a common denominator, especially if you have multiple games that have been affected differently. Country, release date, last updated, languages, platform support, device support, something like that. I'm leaning toward the idea that whatever changed is intentional by Google (or will normalize). Getting out ahead of if it will make a difference, blog posts won't.

 

I don't think there is anything consistent with who has been affected. I think those that are less effected probably have larger portfolios, it'd make sense that more app from this developer is more prominent and converts better than a random and somewhat broken looking similar apps.

With my two games, MM2 beats MM1 in almost every metric you can measure. But MM1 is the one that benefited from this recent change and MM2 tanked. The biggest and most obvious difference is that MM1 was published by adult swim and they have an established game catalog where as MM2 was self published by us and we don't have a game catalog...

Could it be that more apps by developer has benefited from broken similar apps?

 

@zombiegorilla: In terms of the # of developers affected, that has yet to be seen. We are talking about something that happened days ago. I think we will agree to disagree on that. As I said, I closely watch charts, rankings, trends, and I see categories turned upside down, apps that were once ranked in the top 10 or 20 have fallen off, and new replacements have come in that all vary in ratings, installs etc. The app suggestions in the similar apps section are totally whack and make no sense. Perhaps again, attrition and time will yield better results.

I hope you are right in that if this change is intentional, that the results will take time to propagate, and the dust will eventually settle. I know one thing is for sure, your tune would be very different if you had in fact, been impacted by this. Appreciate your input though re: google. That's very interesting that they are seemingly oblivious to what appears to be a big problem. You also mentioned they have thousands of employees, so that makes me think that news has not spread about this issue yet.

@Antony-Blackett: I have spoken with over 10 app developers in total, many who are prolific app developers, and even the ones with large portfolios have been hit very hard with this issue. Having less or more apps doesn't seem to be a determining factor either.

 

Yeah, I guess if you think of it like a web where many of the usual connections have been severed and your apps are trapped on an island with no connections to them through that web then you'll suffer regardless of how many apps you have on that island.

But it's all just speculation really. I'll keep prodding the contacts I have at google. I'm not sure what else we can do otherwise. Without knowing if it's a bug or not I'm not sure it's wise to go changing things like store pages and marketing strategy when it was working fine before. There's a chance you'll just break what was working when it's eventually fixed. And while this issue if it is an issue is in effect, you'll have no idea which change may or may not have had an impact when it's restored to normal.

For example, I was running a store front experiment that I can no longer trust the results of as my overall store conversion rate has gone from 15% before this issue, to 30.5% but that's a result or far less organic traffic arriving on my store page and so a larger proportion of that traffic will be looking specifically for my app.


Interestingly, Major Mayhem 1 that benefited with extra installs from this issue is now slowly normalising back to where it was before. The store page conversions are also down, so it seems that even apps that may have looked like big winners initially may soon suffer long term problems as well.

 

In the same situation, downloads have dropped a lot, and some other developers I know have the same problem.At present, no solution has been found. Contact Google, saying that it may be the cause of Android vitals. However, our product crash rate and ANR rate are actually very low values.

and how can I join the Discord developer group，I opened it without any content

 

If I'm not mistaken, ZombieGorilla work for a big company, that and the adult swim comment make me made my own custom conspiracy theory.

 

The old link was broken. Here's the working one:
https://discord.gg/5Hny2Xy

 

I received a new answer:
"Hi,

Thanks for your reply.

We received more information this morning that this situation is the result of a change in the ranking algorithm (suggested in my previous message but now known for sure). Our engineering teams are currently investigating this and working on a solution, though we do not have an ETA to offer you at this time. I apologize for the inconvenience.

Please let me know if you have other questions.



Regards,
Pete
Google Play Developer Support"

 

They know about the problem and are already working on its solution.

 

Just my 2 cents : we dropped by 80% installs on 20th june too.

 

I used this form
https://support.google.com/googleplay/android-developer/contact/publishing

 

Is someone here from US (physically and account, not only account registered) whos installs also dropped significantly?

 

I've got another reply from Google Play Developer Support. Although the reply might sound frustrating and disconcerting, it seems to be more the opinion of the support person, rather than actual feedback from the teams working at the Play Store algorithms.

Below is the continuation of the conversation started before and listed in this previous post.


Google Play Developer Support new reply today:

Hi Alexandru,

Thank you for your additional information.

We really appreciate your effort on 'Similar App' investigation. We also understand and appreciate your concern about apps displaying in the Similar Apps section.

This behavior is normal and to be expected. To ensure the Similar Apps section promotes discoverability and shows new or different apps to users, the apps are algorithmically selected and change regularly. Also, if we always showed the app the user just came from, the number of apps that could be displayed in Similar Apps would sharply decrease, and the feature wouldn't be as useful to users.

Over time, the apps in their Similar Apps section will change. This is to keep Similar Apps results fresh for users and to give other apps an opportunity to succeed on Google Play. Unfortunately, due to the constantly changing nature of Similar Apps and Similar Apps algorithms, we don't have the tools to track the other apps that might have your app appearing on their Similar Apps section or other apps appearing on your Similar Apps section. I apologize for any inconvenience this may cause.

I really appreciate your thoughtful feedback about Similar Apps, and I'll pass it along to our product team. I hope this helps! Please let me know if I can assist you further.

Regards,
Luna


My reply:

Hi Luna,

Thank you for your reply and for passing the feedback regarding "Similar Apps" to the product team.

What I would mention though, is that the current algorithm behind "Similar Apps" doesn't promote discoverability, because it only shows the same narrow selection of games everywhere, at the expense of all the others. For example: there are thousands of free flight games on the store, but currently only the same six (6) free flight games show up in the "Similar Apps" section, while the rest of them are completely absent from these sections. This is the opposite of promoting discoverability, which means that the current algorithm is dysfunctional.

Thank you again, and if the product team needs more evidence to prove that the "Similar Apps" section is indeed malfunctioning, then I'm more than happy to collect such evidence and provide it to them.

Best wishes,
Alex



Anybody knows exactly what a "Product Team" is at Google and what it does?

 

They only send "politically correct" responses... they will never tell us what is going on...

 

I have information that the Belarusian developer "gismart" has a huge increase in installations.

 

Are we talking about a titanic in which only the first class are saved?

 

Good to know that it's just an issue.
And yes, there're thousands (my guess) of developers who don't use discord or unity forums. My first post here, and yes, we have the same problem with x5 installs drop. =)

 

I noticed from last 48 hours the fall rate is not increasing and games are at 20 % download of previous average ..

 

There are many apps which are listed almost no matter what you are searching or currently seeing, games which are as unrelated as they could be ... the related stuff from my point of view never worked as it should ... it was just a promotion for the favorites. Same for new apps ... we have a new apps list which is full of games released 10 years ago =)) ... like COC etc ... and even the apps shown there as new are not indie ... are apps backed by millions of installs from cross promotion by their companies like EA, Ubisoft, Gameloft etc ...

Maybe now there are some small developers who got some gains ... but I'm allmoust 100% sure that the big transfer of installs from people who lost their installs is going to the big sharks ...

 

On being constructive, for apps which lost their installs what was your User Acquisition=> Acquisition reports=>Conversion Rate Benchmarks compared to the 50% mark ? Maybe only the under performing apps lost traffic because of the algo change

 

my conversion rate was slightly above 50% (green bar) before this happening..
now it falls to a concerning 10% long red bar

 

SAME FOR ME i lost 40 games on donloads and dont think ill get back, maybe in time when rank again some of games

 

lost more than 90% of installs

 

We're seeing the same problem across 12 of our apps. I added the details in a Reddit post:
https://www.reddit.com/r/androiddev...ificant_drop_in_downloads_since_june_20_many/

 

We saw 80% drop in downloads from 20K to 4K

 

We are developers, we know what it is to suffer, my games are still in the sewer and while "nobody knows anything about the matter" I am looking for plan B, C, D ... I invite everyone to do the same , look for alternatives to our future.
I wish her much strength.

 

Just adding to the discussion, reading all the pages here and the countless threads on Reddit. This doesn't seem like an isolated issue with only a few developers...

One person pointed out that users might not come here, and they are right. I hardly use this forum... Unless I have a problem.

Our game purchases and ad-rev just about follow the same curves downwards. Here are 2 graphs of our installs for 2 of our games. The curves are almost identical in shape, even with the download numbers being so different.

 

No, my app that increased downloads was performing worse in all metrics to the one that lost downloads.

Revenue, store page conversion, vitals, downloads per day, review volume per day, review scores, my reply rate to reviews. All metrics that you see on the dashboard.

I’m positive it’s a bug. I have a support ticket with google, they said they are investigating and will update me when they know more.

You’re all going to have to be patient, keep doing what you were doing before until we hear more information and stop speculating, it doesn’t help anyone.

 

My graph is also identical to those.

 

What is strange is how when I go to a related app, you would think my app would be near the top on it's related apps, nope. Heck, its not even on the see more page. All the related apps seem to be the same, and you can click back and forth one to another on them on related. It seems that the rest of us live outside of this bubble...

 

A new email came from Google Play Developer Support. The new reply doesn't seem to come from the previous support person I've been in contact with (Luna), and in it they required me to prove that the "Similar Apps" section is indeed broken.

Below is the exchange, the previous part being available in this post.


The new reply from Google Play Developer Support:

Hi Alexandru,

Thank you for your email.

As mentioned, the apps in Similar Apps section will change from time to time due to the constantly changing nature of Similar Apps algorithms. This is to keep Similar Apps results fresh for users and to promote app discoverability on Google Play. You may provide evidence to prove why you think the "Similar Apps" section is malfunctioning, and we would love to investigate further.

Currently, our relevant team is still working on to resolve the daily install drop. Your patience will be much appreciated.



My (lengthy) reply to them:

Hi,

Thank you for your reply.

I'm assuming you're not the person I spoke to previously (Luna), so I will reiterate some of the evidence I showed her with further explanations and details included.

Attached to this email are included some images which depict the current status of the "Similar Apps" sections for several flight simulator games. The games were chosen semi-randomly, but with the condition of being free, having more than 1 million downloads each, and good ratings, but in fact no matter which flight games would've been chosen, it wouldn't make much difference in proving the argument below.

There are six images, one for each of the following reference flight simulator games which were chosen for this demonstration:
- Turboprop Flight Simulator 3D https://play.google.com/store/apps/details?id=com.AXgamesoft.TurbopropFS
- Extreme Landings https://play.google.com/store/apps/details?id=it.rortos.extremelandings
- Flight Simulator 2017 FlyWings Free https://play.google.com/store/apps/...s.googleplay.flywings2017flightsimulator.free
- Avion Flight Simulator 2016 https://play.google.com/store/apps/details?id=com.teapotgames.avionflightsimulator
- X-Plane 10 Flight Simulator https://play.google.com/store/apps/details?id=com.laminarresearch.x_plane10
- Flight Simulator: Fly Plane 3D https://play.google.com/store/apps/details?id=com.i6.FlightSimulatorAirplane3D

Each image displays the following:
- In the top-left corner is the icon and name of the reference game;
- In the left-side there is a green rectangle in which is depicted the previous state of the "Similar Apps" section as it was weeks ago (retrieved from the Google Cache);
- In the right-side there is a green rectangle in which is depicted the current state of the "Similar Apps" section as it was captured today or yesterday;

In each image, in the rectangle which depicts the current state of the "Similar Apps" section, are also displayed the following details:
- Some of the apps are highlighted with red and have some numbers or letters inscribed inside;
- The games highlighted with solid red, are games which are strongly similar to the reference game (they are free flight simulators or free aircraft-related action games);
- Each of the strongly similar game has a red identification number, which is consistent across all the screenshots (example: the game marked with 3 in a screenshot, will also be marked with 3 in another screenshot);
- The apps highlighted with a dotted red square, are apps which are somewhat similar to the reference game (they are either paid flight simulators, or cartoony games barely relevant, or non-games with a flight theme);
- Each of the somewhat similar app has a red identification letter, which is consistent across all the screenshots (example: the app marked with C in a screenshot, will also be marked with C in another screenshot);
- The games and apps which are not highlighted with anything, are not similar at all to the reference game (they're not flight related, they're not even simulators most of them).

The following observations can be made by examining the provided images:
Observation 1: There are very few apps displayed in the "Similar Apps" sections currently (especially compared with how many they were before);
Observation 2: Most of the apps displayed in the "Similar Apps" sections currently, are not even similar to the games they're supposed to be alike (whereas before, almost all of them were);
Observation 3: The few games and apps that are indeed similar, are the same narrow set of apps displayed in all the "Similar Apps" sections (there are only six strongly similar games, and four somewhat similar apps, and nothing else as far as I could found on the Play Store, and I looked at many "Similar Apps" sections on many flight-related games).

ARGUMENT:
Premise 1: The main purpose of the "Similar Apps" section is to display similar apps as compared to the selected one;
Premise 2: Most of the apps currently shown in the "Similar Apps" section are dissimilar as compared to the selected one (from Observation 2);
Premise 3: The other purpose of the "Similar Apps" section is to keep results fresh for users and to promote app discoverability on Google Play (as stated in your reply);
Premise 4: The apps that are indeed similar are few (from Observation 1) and are the same ones (not-fresh) across all the "Similar Apps" sections (from Observation 3);
Conclusion: The current state of the "Similar Apps" section doesn't fulfill the purpose to display primarily similar apps, nor the purpose of keeping results fresh and promote app discoverability, therefore the algorithm which populates the "Similar Apps" section is indeed malfunctioning.

Thank you for your attention and patience (if you managed to read all of the above and to reach this far).

Best wishes,
Alex



Below this post are 4 of the 6 images that I attached to my reply (only 4 because of the forum's limits).