Nearly Half of Apple Music Users in U.S. No Longer Using Free Trial

[navaira]

Its not going to just magically know what to recommend people out of the box.

Why?

I raised this point before: it's all in one app. It has access to my library, to my playlists, to my play counts, to my last played dates. Why doesn't it know what I like?

Spotify's Discover Weekly gives me new stuff every week. I like 90-95% of it. It only gave me two or three tracks (in last three weeks) that I already had but I didn't mind because they fit the playlist. And I don't even listen to Spotify all that much. Why does Spotify know what I like without me hearting stuff but AM doesn't?

 

[bking10]

exactly. 5000 and how did they find them. Apple has direct access to the full numbers not some tiny cut

you should take a statistics class. more than an ample sample size

 

[johngordon]

Why?

I raised this point before: it's all in one app. It has access to my library, to my playlists, to my play counts, to my last played dates. Why doesn't it know what I like?

Spotify's Discover Weekly gives me new stuff every week. I like 90-95% of it. It only gave me two or three tracks (in last three weeks) that I already had but I didn't mind because they fit the playlist. And I don't even listen to Spotify all that much. Why does Spotify know what I like without me hearting stuff but AM doesn't?

Fair enough with things like play counts and last played dates, but only up to a point.

As an extreme hypothetical example, suppose you don't really like much pop music, but Taylor Swift's 1989 album briefly became a guilty pleasure, and you played it to death when it came out, but have just gotten now sick of it. If AM went by that info, it might start recommending loads of Katy Perry and Little Mix for you, which you have no interest in.

But specifically saying "yeah, I really like this, and would love to hear more stuff like it" should, in theory, garner better results.

Which isn't to say using play counts and last played dates won't work in a lot of cases, but I can see the logic of not relying on it.

 

[navaira]

I've gone through my library hearting New Order, Pet Shop Boys, Suzanne Vega, Björk, Mylene Farmer, Amon Amarth, Tyr, Garbage, Army Of Lovers, Asgeir, The Cardigans... because For You kept on recommending me Michael Jackson, Janet Jackson and R&B in general. I like the Jackson siblings and I like R&B, but I like a lot of other stuff too.

Those are my For You playlists for today:



Is there a way to go to the beginning and re-do the pink bubbles without having to cancel membership and then pay for it?

*going through library and hearting more stuff*

 

[johngordon]

Click on your account, and there's an option to 'Choose Artists For You'.

Its been getting better for me recently - been hearting quite a bit of stuff on Zane's show, and stuff like New Order. So have had a pretty good indie one with a lot of stuff I don't have, and one of Arthur Baker produced tracks. And an Intro to Tegan and Sarah after I hearted and added one of there songs to my library. So starting to get a bit more off piste.

The Tegan and Sarah example is exactly what it should be doing - if I hear a track by someone new and like it, having a curated best of type playlist is going to be better than just adding all their albums and listening to them all. (At least a that stage).

But the Intros to.... artists that are already in people's libraries still seem a bit odd, although I guess out of the box it doesn't have a lot to go on. Maybe it would even be better starting with a clean slate and it telling you that stuff will start to appear in there when you heart stuff.

 

[navaira]

Thanks. Just did it and... it ran out of artists...?! I am now stuck with 10 bubbles and can't get any more. I clicked that I like dance, pop, alternative and electronic, but it still won't give me Pet Shop Boys, New Order, Depeche Mode or The Smiths. I got Bob Dylan though.

I'll try again.

Edit: Well that's a good selection



Edit 2: After three tries it just stops giving me new artists at some point, then when I click "More Artists" I get either one floaty bubble and all others disappear, or just a blank screen. I'm giving up. Somebody wake me up if they fix AM before September 30.

 

[Merode]

Thanks. Just did it and... it ran out of artists...?! I am now stuck with 10 bubbles and can't get any more. I clicked that I like dance, pop, alternative and electronic, but it still won't give me Pet Shop Boys, New Order, Depeche Mode or The Smiths. I got Bob Dylan though.

I'll try again.

Edit: Well that's a good selection

View attachment 576053

Edit 2: After three tries it just stops giving me new artists at some point, then when I click "More Artists" I get either one floaty bubble and all others disappear, or just a blank screen. I'm giving up. Somebody wake me up if they fix AM before September 30.

Yeah, the bubbles screen is terrible. Bubbles tend to glitch, you may run out of them before picking enough artists/genres and they get stuck when they get big and you try to move them from side to side on smaller iPhones.

This app couldn't really be any more un-Apple.

I don't know why they won't allow me to just type in my favourtie artists/genres.

 

[SunbeamRapier]

Thanks for reminding me. I signed up but have never used the service so now i have cancelled the auto-renew. To be honest I don't know HOW to use the service. I don't wear earbuds so no interest in listening on my phone. I have an old iPod Classic connected to my stereo but it doesn't support Apple Music. duh. Oh and I only really listen to classical - and there are two good curated classical free to air stations in Sydney. And I have 700+ CDs ripped on my iPod but i don't even listen to that because iTunes lost my playlists and i can't be bothered to work out how to navigate the umpteenth reworked iTunes user interface. I don't think I am an Apple Music customer... More than that, Apple has separated me from my music. Kill iTunes please.

 

[Benjamin Frost]

Thanks for reminding me. I signed up but have never used the service so now i have cancelled the auto-renew. To be honest I don't know HOW to use the service. I don't wear earbuds so no interest in listening on my phone. I have an old iPod Classic connected to my stereo but it doesn't support Apple Music. duh. Oh and I only really listen to classical - and there are two good curated classical free to air stations in Sydney. And I have 700+ CDs ripped on my iPod but i don't even listen to that because iTunes lost my playlists and i can't be bothered to work out how to navigate the umpteenth reworked iTunes user interface. I don't think I am an Apple Music customer... More than that, Apple has separated me from my music. Kill iTunes please.

Oh for proper classical music categorisation in iTunes, let alone Apple Music.

I almost would like to apply to Apple to create a new job for them called Classical Music Specialist, in order to bring order to their store. Music should be ordered by composer, and also by types like chamber music, oratorios, etc. The omposer category in iTunes is hopeless, as it isn't used properly.

Considering that it would cost Apple loose change to transform iTunes into a well-organised source for classical music, it lowers my opinion of them considerably.

 

[rhoydotp]

5$ is less. For free is less. Less is better.

Yet, most of us are asking for more meanwhile paying less or not paying at all.

 

[Andrei90]

How does it compare to any other free trial for any other product or service you can think of?

If 20% of trial users become paying customers, is that typically above or below the average?

If only 20% of 11 million (which is only 1.2% of your total customers base) become paying customers, you can say goodbye to Apple Music. Spotify has 20+ million paying customers and Apple was targeting for at least 100+ million paying customers. So yeah, having only 2 million paying customers would be one of the biggest Apple failures ever, maybe even worse than the Apple Watch.

 

[Andrei90]

exactly. 5000 and how did they find them. Apple has direct access to the full numbers not some tiny cut

Apple numbers mean something different though. The guys from the survey tell us 50% of people using AM removed the auto renew option, that doesn't mean they stopped using the service. Apple on the other hand is telling us 80% are still using the service, nothing about auto renew there. Obviously the survey asked the wrong question. I'm using AM and I removed the autorenew option to not get charged at the end of the 3 months by mistake.

 

[williamsk0]

Can't say exactly what it is I dislike about Apple Music compared to Spotify... I don't like not being able to see my loved tracks or not being able to shuffle play all of my loved tracks. I don't like having to add music to my music before I can put it in a playlist (what does that even mean? adding it to "my music" doesn't mean I own it). In Spotify, if I want to save an album to listen to it later, I can just drag and drop it onto a playlist folder and boom - new playlist is created. Much simpler.

Probably over all, it's just that I have tons of time spent liking tracks and building playlists in Spotify and I don't want to do that all over again in Apple Music.

Oh yeah - and even when I'm listening in iTunes, I can't scrobble my listens to last.fm unless I add every single song to my music first. I have years and years of data on last.fm that I don't want to abandon, and Spotify natively scrobbles everything I listen to whether its from radio, a playlist, or something I just randomly queued up.

So anyway... I'm not planning to keep Apple Music past the free trial.

 

[alidixon]

They need to fix all the broken crap that borks users libraries and mismatches music and artwork. It's pretty sad that 60% of the people who tried don't want to renew. I am in that camp, because AM trainwrecked my music files.

It has given me tons of duplicates too

 

[blasto2236]

I love the service, but 2 months in and it continues to randomly delete songs I have saved to my iPhone for offline listening. I have no service at work, and company Wifi blocks the iTunes Store (and thus AM), so I find it infuriating to use. I have downloaded and re-downloaded the same album 10 times only to have it disappear again a day or so later.

If they can fix this problem, and fine tune a few other things, I'll stay when the trial runs out next month. I'll wait for the final version of iOS 9, and if I'm still experiencing these issues I guess I'll have to look elsewhere for my music subscription. The problem is, I always preferred Beats Music, which is now gone. I don't really care for Spotify, and Deezer still isn't available in the US. Didn't care for Tidal either. Any suggestions?

 

[Dimwhit]

Fair enough with things like play counts and last played dates, but only up to a point.

As an extreme hypothetical example, suppose you don't really like much pop music, but Taylor Swift's 1989 album briefly became a guilty pleasure, and you played it to death when it came out, but have just gotten now sick of it. If AM went by that info, it might start recommending loads of Katy Perry and Little Mix for you, which you have no interest in.

But specifically saying "yeah, I really like this, and would love to hear more stuff like it" should, in theory, garner better results.

Which isn't to say using play counts and last played dates won't work in a lot of cases, but I can see the logic of not relying on it.

I'll go one further. I don't want Apple Music looking at play counts because my family shares our iTunes library and there's no way to differentiate users.. I don't want music recommended to me based on all the music my kids play.

 

[johngordon]

If only 20% of 11 million (which is only 1.2% of your total customers base) become paying customers, you can say goodbye to Apple Music. Spotify has 20+ million paying customers and Apple was targeting for at least 100+ million paying customers. So yeah, having only 2 million paying customers would be one of the biggest Apple failures ever, maybe even worse than the Apple Watch.

Not quite what I asked.

Only having 2 million paying subscribers by when? Presumably people can still sign up following 30 September, and seeing as there's a new iPhone coming out, an update to iOS and likely a fairly big push for AM, its likely that, come the end of the year, they'll have more than 2m subscribers.

Apple numbers mean something different though. The guys from the survey tell us 50% of people using AM removed the auto renew option, that doesn't mean they stopped using the service. Apple on the other hand is telling us 80% are still using the service, nothing about auto renew there. Obviously the survey asked the wrong question. I'm using AM and I removed the autorenew option to not get charged at the end of the 3 months by mistake.

Exactly - I can imagine a lot of people turning off auto renew, but won't make a decision until the trial runs out.

I think someone mentioned Tame Impala earlier - they're on Zane's show today.

 

[maxsix]

They released it too soon. Apple should've put another 3 months of work into it before releasing it so that people's first impression of it was favorable rather than experiencing it as a bug filled beta that leaves a long term negative impression.

You've hit the nail on the head.

Much like the messy, rushed launch of Apple's Buggy Maps, Apple seems to be almost desperate to get a strong foothold in the music space. Their obscenely overpriced buy of Beats, with it's high priced, rather dull headset that sells largely because young people worship the image, they worship obscenity laden hip hop and rap music that features booty shaking slutty women, reveals Apple's willingness to lower themselves into the cash filled gutter of current popularity.

Anything for the almighty dollar. Now aligned with the segment of society that spends big bucks on drugs, fast cars, and other excesses, Apple's raking in the cash while they can. It's good to be Apple

 

[flux73]

With statistics and sample sizes, I thought the total size of the population was irrelevant?

Assuming a random, normally distributed sample, isn't the equation for the margin of error:

The margin of error in a sample = 1 divided by the square root of the number of people in the sample

So if the sample size is 5000, the margin of error will be about 1.4%.

The total population could be 50,000 or 500,000,000 and it wouldn't change that.

It's not irrelevant. But it's also not the only factor. Other factors include what you're looking for, whether it's quantitative, semiquantitative, or qualititative, and also the prevalence of the what you're looking for.

For example, if your study question is just whether a single user dropped out of Apple Music, and the prevalence of the drop out were 99.9%, you wouldn't need a big sample size at all. Probably 10 users would be sufficient to detect whether or not there's a single user who dropped it. Now say, the prevalence is 0.1%, you'd need a MUCH bigger sample size to detect find that single user who dropped. If you only surveyed 10 users like before, the chances are high that you wouldn't find a person and therefore concluded that not a single user dropped Apple Music, which would be a false negative.

Statistical power analysis is about calculating what your minimum sample size needs to be, in order to be fairly confident about your results. You can never be 100% confident unless you have 100% of the data. Which Apple does have, and MusicWatch does not.

To your first point though, population size is relevant. Let's take another extreme example. Let's say you have a group of 10 people, and let's say your study question is again whether or not a single one of them stopped using AM. If you selected 5 people, and none of them dropped AM, would you conclude that no one stopped using AM? No, you'd survey all 10, right? The conclusion becomes even less reliable if your asking what % of people dropped AM. After 5 people, if one of them dropped it, could you reliably conclude 20%?

Now say you surveyed 1000 people. If you surveyed 500 (50%) of them and found a rate of 20%, you'd be more confident because your sampling is bigger. And unless your selection of 500 people were somehow biased, you could feel better about your rate of dropout. However, if you were to survey only 10 of them (1%), I don't think you'd feel as confident about your results, right? Now, just extrapolate that out to 10 million people...

And therein lies the problem. 5000 people COULD be sufficient if your sample is truly random and representative of the overall population. But it also depends on what question you're trying to answer. MusicWatch is projecting an exact %, suggesting they think that the remaining 10,995,000 people are precisely like those 5000. The problem, as I pointed out before, is that this could NOT be a randomized study. Study participation is voluntary, which automatically introduces selection bias. Therefore, the probability is very high that the study participants are not representative of the overall population of AM users. Apple's subsequent release of data showed that it wasn't.

 

[CE3]

I signed up for the trial a couple weeks ago but I haven't used it much. I've been subscribed to streaming music subscriptions since around 01 or 02 (first it was MusicMatch until it got bought out by Yahoo Music, and then Rhapsody when they bought up Yahoo). I might give Apple Music another try, but the UI seems a bit cluttered and when I get in my car I just want an easy way to play and search for songs and albums and add them to my library. The Rhapsody interface is better organized + it's easier to stick with a service I've been using for years and already have a big library on.

 

[virginiaplain0]

I have had trouble with my playlists ever since I signed up for the trial. When I shuffle music in my playlists, sometimes the same songs repeat over and over. I've tried fixing it by turning off the repeat, which automatically turns on for some reason. I've also tried manually adding each song from a playlist to "Up Next", but nothing seems to work. I have enjoyed adding songs to my playlists without having to pay for the songs, but I am not enjoying how Apple Music has messed up playing my songs. I might just stop this.

 

[flur]

So yeah, having only 2 million paying customers would be one of the biggest Apple failures ever, maybe even worse than the Apple Watch.

Why do people keep on with this? The Apple Watch is statistically one of Apple's most successful new product launches EVER. Pick an ACTUAL failure, like Maps, to compare with, m'kay?

 

[SunbeamRapier]

Oh for proper classical music categorisation in iTunes, let alone Apple Music.

I almost would like to apply to Apple to create a new job for them called Classical Music Specialist, in order to bring order to their store. Music should be ordered by composer, and also by types like chamber music, oratorios, etc. The omposer category in iTunes is hopeless, as it isn't used properly.

Considering that it would cost Apple loose change to transform iTunes into a well-organised source for classical music, it lowers my opinion of them considerably.

Yes, EXACTLY!

The problem with Apple is that their idea of "culture" is Lady GaGa. Steve Jobs was many things, but he was not a cultured man and, as you say, iTunes is absolutely useless for classical music. My uncle, a Bach fan. abandoned iTunes altogether years ago and used another app to arrange his music on his iPod.

Of course it would be possible to write a decent music app for classical music but the hoo polloi are gravitating to streaming - which is ideal for the kind of disposable music that the younger generation has always preferred, and if the app was any good you can be sure that Apple would just copy it (sort of) and you would lose your investment.

I have too much music for Apple Match, duplicate tracks I can't find or remove and I have lost all my playlists in various moves from one Mac or hard drive to another. I have just given up listening to music - I just realised one day I never turn it on.

Probably the solution for Apple is to create a design centre in Europe somewhere and give them all the cultural stuff to do. To fix classical music you need an agreement with all the recording labels so that your indexing system works - otherwise Mozart, W.A. Mozart, Wolfgang Amadeus Mozart, Mozart, W.A. etc etc mean that you can't even really sort properly.

And it would be nice to be able to create playlists which select arias and miss the recitatives in operas in some cases - and in Bach Cantatas perhaps.

Oh to use an application which does not call an overture a "song".

 

[johngordon]

It's not irrelevant. But it's also not the only factor. Other factors include what you're looking for, whether it's quantitative, semiquantitative, or qualititative, and also the prevalence of the what you're looking for.

For example, if your study question is just whether a single user dropped out of Apple Music, and the prevalence of the drop out were 99.9%, you wouldn't need a big sample size at all. Probably 10 users would be sufficient to detect whether or not there's a single user who dropped it. Now say, the prevalence is 0.1%, you'd need a MUCH bigger sample size to detect find that single user who dropped. If you only surveyed 10 users like before, the chances are high that you wouldn't find a person and therefore concluded that not a single user dropped Apple Music, which would be a false negative.

Statistical power analysis is about calculating what your minimum sample size needs to be, in order to be fairly confident about your results. You can never be 100% confident unless you have 100% of the data. Which Apple does have, and MusicWatch does not.

To your first point though, population size is relevant. Let's take another extreme example. Let's say you have a group of 10 people, and let's say your study question is again whether or not a single one of them stopped using AM. If you selected 5 people, and none of them dropped AM, would you conclude that no one stopped using AM? No, you'd survey all 10, right? The conclusion becomes even less reliable if your asking what % of people dropped AM. After 5 people, if one of them dropped it, could you reliably conclude 20%?

Now say you surveyed 1000 people. If you surveyed 500 (50%) of them and found a rate of 20%, you'd be more confident because your sampling is bigger. And unless your selection of 500 people were somehow biased, you could feel better about your rate of dropout. However, if you were to survey only 10 of them (1%), I don't think you'd feel as confident about your results, right? Now, just extrapolate that out to 10 million people...

And therein lies the problem. 5000 people COULD be sufficient if your sample is truly random and representative of the overall population. But it also depends on what question you're trying to answer. MusicWatch is projecting an exact %, suggesting they think that the remaining 10,995,000 people are precisely like those 5000. The problem, as I pointed out before, is that this could NOT be a randomized study. Study participation is voluntary, which automatically introduces selection bias. Therefore, the probability is very high that the study participants are not representative of the overall population of AM users. Apple's subsequent release of data showed that it wasn't.

There is an assumption that the sample group is random, normalised, or whatever the statistical term is.

Assuming that it is, then isn't it still the case that a sample size of 5000 will have a margin of error of around 1.5%, no matter what size the total population is?

Or is that formula wrong?

 

[flux73]

There is an assumption that the sample group is random, normalised, or whatever the statistical term is.

Assuming that it is, then isn't it still the case that a sample size of 5000 will have a margin of error of around 1.5%, no matter what size the total population is?

Or is that formula wrong?

Assuming that it is, yes the formula is (sort of) correct. But it's not that sample size doesn't matter. Actually, it assumes an infinite or at least sufficiently large population. For 11 million people, that's big enough that it works. The formula does not apply with a small population. The best example of how this works is a flip of a coin. If you flip it only 10 times, you could easily get 20% heads and 80% tails. The margin of error is huge and sample size very much matters. If you flip it 100 times, you're much more likely to get closer to 50%. 1000 times makes it even more likely you get 50%. After awhile, flipping it more times won't change your results by that much. So sample size starts to matter less.

But you also have to assume zero bias and that the sample group is completely representative. Margin of error doesn't correct for bias. Given that Apple later released figures that showed that the MusicMatch figures were way off, there was clearly bias introduced. My assumption is that the survey is voluntary. And if it is, there's no way one can argue that bias is unlikely. It's for the same reason why customer feedback surveys tend to be negative. People take surveys when they feel like there's something to complain about.

Also statistically, we were kind of talking about different things here. I was referring to statistical power which applies to binary hypotheses, and does depend on sample size. The question I have is whether this study has the power to detect a problem with Apple maintaining subscribers. MusicMatch's data indicates a problem to most people. I don't think their sample size was big enough to have confidence in saying that it's problem.