Sudden drop in number of daily installs on Google Play Store

any news ?

 

Bad news. Our new store descriptions are no longer getting us listed in similar apps.

We may have to hire a shaman to investigate further ;-)

 

We have top shaman working on it right now.
Who?
Top ... shaman.

 

We also tried to update store descriptions of our game:
Zaptiye: the Ottoman Counter Assassin

Result: It got worse. Although it is a third-person action-adventure game, similar apps section displays bubble shooting games which are completely irrelevant.

I contacted the play store, they said the technical team will look into it.

 

I think it's time to come to terms with the fact that the Android marketplace might not be a viable business option for people who aren't able to pull massive audiences with high retention rates.

 

Yes, it's exactly that. If your game has an average user playing it for 5 hours across its lifetime, then your ad revenues would be high enough to start buying traffic with ads yourself. Anything with 10 mins playtime is from now on junk which Google wouldn't bother to spend resources on processing.

There is a valid argument though that there are huge swathes of users who are in principle incapable of holding attention on a game for longer than they can on a YouTube video: kids, low-IQ adults and many more (possibly as much as 50% of mobile audience, if not more)... Google is probably aware of these audience segments but has rationally decided that, since this audience doesn't have much (or any) disposable income, it can be largely ignored in calculations. Any games targeted at people who can't pay are generally irrelevant, and it doesn't matter how much traffic they get, if at all. Any such traffic will never convert into money for Google, or game developer. It will also never convert into sales of physical goods or services which AdMob advertisers offer, so any ad spend is wasted there, and, given proper controls, the marketer would not want to show any ads there at all. The only value this junk traffic may have is to create an impression of populated servers and active community in big multiplayer games. But indies don't make those.

You might also be interested in this article where the guy explains the new economics of mobile gaming:
http://blog.soomla.com/2018/08/15-types-hyper-casual-and-arbitrage-games.html?

 

I have just experiences a similar drop in downloads, but this was in late September. Have you experiences a 95% drop here as well? Do you think it relates to the previous drop?

 

Well the 'non-similar' apps displayed for my app have changed, however they are still non-similar (they are not even similar to each other). The last word in my App title is the word 'Plus'. The non-similar app listed for my app are :

FCMB Mobile Plus
Career Fair Plus
Gallery Plus
Daily Tarot Plus <= seriously!

Obviously got some pretty sophisticated algorithms at work there, so much for AI lol.

 

I think the new algorithm is reaping on data and testing placements, we should give it time.

My game was averaging 1300 DLs/day. With new algorithm back in June it went down to 600-700, but around September 20 it suddenly spiked and I've had around 3K/day for a week. Now it seems to be back under 1K/day, all this reminds me of the first months after launch.

My related downloads look alright though, so maybe there's a separate issue with that.

 

It has happened again with one of my games. Downloads fell from a consistent 1500 to 250 in less than a day. This is ridiculous. Anyone faced the issue again?

 

Seeing a lot of ups and downs now. But we're back up to perhaps 80% of the daily installs we had before The Event. 'Similar Games' stuff seems to be working a bit more normally these days too.

 

This is basically my experience right now. 3 months of losing half my downloads, then I had a record-breaking peak of 10 days at 3x my old figures, then down again to about 75% of "normal" downloads. I made zero changes to store page in this period.

 

I personally never recovered from 'the event'

 

Same here.

 

Same here.

 

Thanos snapped his fingers and half the developers relying on Google Play as a storefront all went broke.

 

I think Thanos would say, he actually fixed, what was once broken

But seriously, here we are, where technology fails, when everyone relies on same tech.
Now big brother is laughing in a face.

 

same here, bad and hope in future will get back downloads

 

Same here. For those 3 weeks, I was getting 40,000 downloads a day ... more than I've ever had. And then it came crashing back down. Back to 1,000 downloads a day, which is not enough to pay rent! Too bad google is stamping out thousands of independent and creative people, all for $$$

 

Let's imagine the world in 50 years, when google's self-driving cars have replaced ALL cars on the road. You ask your Google self-driving car to take you to a new restaurant that serves delicious food. It drops you off, once again, at a monolithic, grey, chain-restaurant that feeds people pure crystallized blocks of MSG. You didn't even catch a glimpse of the small family-run restaurant that has been a local treasure for over 80 years, because the Google Car made sure you didn't drive past it or see it. It didn't even appear on the map. "Sir, we want to ensure that customers spend as much time and money in restaurants as possible, consistently", says your google car. Only 3 monolithic restaurants remain in the entire country, as thousands of start-up restaurants and small family-run restaurants go out of business. "Doesn't this take away opportunity from young people trying to start businesses, decrease diversity of restaurants, and concentrate power into the hands of a select few giant corporations?" you ask a Google employee. "Sir, we want to ensure that customers spend as much time and money in restaurants as possible, consistently", the Google employee robotically responds to you.

I fear the day when Google owns the lion's share of self-driving cars. A google future is not a bright future.

 

I fear the day when everyone is reliant on cars. Unfortunately, that has been every day for years.

Edit: What I mean is that the ignored cost of short term conveniences has been impacting us for a long time, and I think that other conveniences in that metaphor (eg: the car) have far more important impacts than which restaurant Google prioritises!

 

The new algorithm is a rollercoaster for sure.

 

Looks fun to me.

But doted prediction is unjustified. At most, I would expect anything between lower blue line, to halve distance to first gray horizontal line. Or at most to the line. But not higher than was initially.

You may expect pick around December.

 

That's not prediction, that's the google graph comparison vs last period, so that dotted line at the end is actually before the start.

Here's the full graph.



End of June is when the algorithm changed. The spike after that is a feature. The longer uptick at the end... i'm not sure, residual feature effect maybe? But then why it crashed immediately after a feature I can't tell you... Finally, dead again after that.

I'm hoping for a feature next week. I'll update the graph again in a month or so and you'll see more crazy i bet.

I feel like my game is sitting on some threshold line in the algorithm, but i have no idea how to get it consistently above that threshold.

 

Ok, I see match of previous graph November and latest May.
Just strange match tho, as appears like what, 6 month cycle? But nvm.

 

You can't really read into it, the first spike is the game launch, so new and updated apps feature i guess. The second spike is another feature.

So yeah, it's not really a cycle thing, more just me updating things and getting features as a result. It's odd that their editorial team likes my game and gives it features but their algorithm hates it.

Go figure.

 

Ah I see.

 

So.. I just noticed that all the 'related apps' on my list have similar keywords in their short descriptions, but I don't. I just changed mine to better match there's, I wonder if I'll creep back into the related apps lists for those games.

 

Yep definatelly worth trying.
As the result should be valid, as long algorithm don't change again.

 

The independent restaurant can still put up a good old real world sign you can see as you don't have much to do while your Google car is driving you around, and then since you went to the same restaurant 100 times already this year, you may respond to this advertisement and tell your car to stop at this restaurant you've never been to.

I'm guessing in this thread there is not a lot of independent advertisement for the games here, nor much social media promotion. Primarily depending on Google just delivering customers. When you depend entirely on Google doing all the work, you're always going to be at the mercy of Google's choices. Google's choices are always going to be in their own self interest, whatever they suspect will maximize their own profits in the short and long term. Google's self interest isn't necessarily the same as your self interest. It probably isn't in Google's self interest to promote your game as much as another game that may be generating Google more profits, so I am surprised there seems to be so much surprise in this thread when Google made these changes.

 

While I agree that relying on google is not the best solution, it's often not that transparent as to where your downloads are coming from and why. So while I understand that it's completely up to google I'm not just going to sit here and say

"30,000 downloads a day was nice, but now for what appears to be no reason it's 3,000, oh well."

no I'm going to ask, "What is the difference between now and then? What's the difference between my game and other games that appear to of similar quality with less downloads and less reviews? How can I leverage that free marking more consistently?"

 

Obviously a fair criticism here.

 

Could this be the reason? https://android-developers.googleblog.com/2017/12/improving-app-security-and-performance.html

Google:

• August 1, 2018 – All new apps required to target API level 26 (Android 8.0) or higher.
• November 1, 2018 – Updates to existing apps are now required to target API level 26 or higher.
• 2019 and beyond – Each year the targetSdkVersion requirement will increase. A year after every Android

Basically they could be penalizing (de-ranking) existing apps that target Sdk below 26.

P.S. I am not much into mobile development, so far I have only released 2 Android games. One of them is a preview demo (haven't updated it in years, installs have dropped quite alot) and the other is a simple platformer (never actually picked up).

 

This is interesting findings if is true.
But my question would be, did it happen in past, for previous versions?

 

As far as I know this is the first time Google enforces a minimum SDK version.
Most likely it's a response to a General Data Protection Regulation (GDPR).

From the article:

...For example, only apps with a targetSdkVersion of 23 (the API level of Android 6.0) or higher give the user full control over what private data – such as contacts or location – the app can access via runtime permissions.

So potentially the apps that target lower sdk will remain in store but will not be shown in the countries that are under GDPR (Basically whole EU).

 

That's a good theory, but definitely not the cause in my case; my game's targetSdkVersion is 28.

 

I think the Similar Apps Section is broken again, My app is a board game, but the similar apps none are game, and there are even 2 watch apps for Wear OS...

 

It's still early days but here's the results from changing my short description on the 9th this month. there seems to be a direct correlation in terms of timing of the short description change. I mentioned that we're in line for a feature but I've checked and cannot find us featured anywhere and google said we're scheduled for the 16th, so I don't think it's that.

I've also looked and found our game in some other shooting games 'related apps' sections now.

It looks like related apps is the key and google seem to place some weight on the short description. I'm not exactly sure how the weighting works but my strategy was to try and match 'keywords' of other titles short descriptions while still being relevant to my own app and being appealing to users at the same time.

 

Just makes me think ...

• Would it be google weighting thing?
• Or maybe people like short descriptions, which are more appealing to them?

 

My thinking is that it must be a google thing because I don't think the short description appears anywhere except in the related apps list (as far as i can tell) and it wouldn't appear there if it relied on people responding to it as it wouldn't be in the list for people to respond to. Also, our short description before was still appealing in my opinion, but just didn't match other popular shooter short descriptions.

 

We see a similar behaviour. But we still have no idea what we can do.
We are not shown in the similar apps category.
It works for the last 20 days. But now it stops working.

 

It's been nearly half a year. Stop throwing good money after bad.

 

Let's move to other platforms! Luckily, games made with unity can be built for them as well.
Your game(s) can be too good for mobile, but still, could earn nothing in mobile. I have a third-person shooter which i think can be a good pc game with some good effort. I will go after that. Waiting for google's mercy won't get me anywhere.

 

Hi, anyone huge drop install in December. My game was drop 90% over the previous month.

 

It seems to happen randomly. I'm pretty sure you will have fallen off some 'similar apps' threshold in the google algorithm. Every time my app dips this is why. It gets fixed again for a period either after a feature or if I optimise my games name and short description to better fit other 'related apps'.

 

I think what Google wants is a constant cycle of new, big earning games or games as services, and that you've managed to find a way to game that system. If nothing else, it's good advice for people going forth that there is a potential solution for the time being.

 

Who knows what google are trying to optimise for. From the looks of it they are trying to surface games that the audience would play rather than games that are similar in terms of mechanics and genre.

 

In your experience, how soon does it take for the short description changes to show in the download stats?

 

Almost immediately. Like 1 day maybe 2 max. so if nothing changes then keep trying new things.

 

Thanks! I was thinking about trying 'Store listing experiments" with various short descriptions, but I'm not sure if that would affect the "related apps" situation. What is your opinion?