Forum
Hi Gorge!
You say that her music was taken off Spotify during her peak, but Apple Music wasn't even existing during 1989's big months. The platform started in 6/30/2015 and she also blocked her albums there at some point.
By the end of 2015, their number of users was still 20 times lower than the current count of Spotify users, on that context 500 million streams equals to 10 billion current Spotify streams. It's always key to keep market statistics in mind before doing assumptions! Extrapolation ratios are never perfect but that's the best we have to avoid biased data. Plus considering Apple Music public is more urban oriented than Spotify public someone like Drake would be more penalized if we start going that way!
What you're suggesting is mathematically impossible. According to your formula, for every stream on Spotify, an artist receives around 1.6 streams on other comprehensive audio streaming devices. Following your formula, since Taylor received 0 streams on Spotify from later 2014 to mid-2016, she also received no streams on Apple Music. It does not add up. We're now almost 4 years after the released of 1989 and the album is still easily charting within the top 100 daily albums. How am I supposed to belive an album like Unapolegetic from Rihanna has more than 3 times more streams than 1989 on Apple Music despite being bigger on every metric and being way more promoted with a better chart run?
Hi again Gorge!
As I said, no formula is perfect. The point is to not enter into biased / manually manipulated data that would be way more flawed still than fixed formulas used for all.
Your example precisely show that it is all the more hazardous to get into home made figures. You fail to mention the market once again and you suggest 1989 got zillions of streams on Apple Music while the platform wasn't even existing back then. If you say that 1989 should be higher because it was taken off from Spotify (while it was basically taken off from every relevant streaming service at the same time), then why don't you argue than her older albums should be lower? They got streams on a time that had Spotify dominating much more the industry, so the ratio for their year would be lower.
The whole point / validity of the formula is because of the continuous and fast growth of streams. Basically, streams that happened in 2014 are irrelevant already. Even big hits from that era got way, way more streams since then that back in the day. That's why the current ratio (1,6) is used for songs of every year, because it is the ratio that is valid now, meaning now that the market is so much bigger. By the end of 2014, a song like Coldplay's The Scientist was still under 80 million streams, by now it stands on 540 million. Back in 2014, barely 1 million daily streams was enough to be #1 worldwide on Spotify and as previously mentioned and there was no Apple Music. 1989 was unavailable from streaming platforms to fuel its pure sales, you can't expect that people now uses manually adjusted formulas in order to inflate its streams and retrieve what they could have been!
Hi! I was just curious: at what do you estimate 1989's Asian sales? You say N/A, but then you say 2.7M overall so how many did you count from 1989? Have a good day
Hi Albus!
1989 was on 445,000 units in Asia (minus Japan) as of the date of the article! 🙂
Looking at your statistics from different artists your formula shows that 1989 (which we both know to be one of the most successful album of the decade) has less comprehensive audio streams than objectively far less successful albums. Your formula can't be applied to Taylor, you're basically saying she received no comprehensive audio streams what so ever for almost 2 years since she wasn't on Spotify. You're saying 2014 streams were insignificant, but I can hardly see how it's insignificant when you even have albums like 'Demi' by Demi Lovato - an album released in 2013 that barely made year-end charts that doesn't even represent a fifth of 1989's CSPC - with 2 times more comprehensive audio streams than 1989, the most streamed female album of all time on YouTube and the second biggest female album of the last 6 years after 25. You don't see the problem here?
Hi Foxie!
You posted the exact same message about a month ago under another account and I already told you that 1) 1989 was also unavailable on remaining streaming platforms 2) the market was terrible back then in comparison to today. You mention Demi, her streams have been mostly achieved after 2014, as for every track. Even monster hits from that time like Happy got more streams since 2015 to date than during their initial run. I already pointed out a few days ago that a track like Bohemian Rhapsody, who's popularity has been big since day 1 of Spotify, reached 50 million in late 2014 only - and added over 450 million since. So yes, streams back then were insignificant at the time and expecting an album to be among the most streamed albums ever up to now thanks to streams from these years is delusional.
I'm not suggesting it's the most streamed album of all time with comprehensive audio streams at all. I'm saying you're clearly undermining its comprehensive audio streams. What is delusional is saying the album received 0 audio streams from its release up until June 2017 like your formula suggests. 1989 was put on Apple Music in 2015 and her entire discography sans 1989 was available on Google Play for streaming during the same time period. How do not see that your formula is inapplicable for Taylor is beyond me.
Foxie,
It's because one more time you focus on the period that interests you rather than looking at the whole picture. Why don't you mention that from 2010 to November 2014 she achieved streams almost exclusively in Spotify which was far and away the strongest platform (Apple Music wasn't existing back then) and that these streams are weighted with current (obviously too high) standards? In the same way, she got heavy streams on Spotify in June / July 2017 after the long removal, that surge was much more relevant than Apple streams in 2015 since its number of users is over 10 times bigger than Apple Music back then. At the end of the day you have multiple years for which the formula was strongly favorable to Taylor up to the end of 2014, then a blank half year in 2015, then 2 years with unfavorable results, then 1 year with favorable results in 2017, and now it's all 'sorted'. What's beyond me is how you can focus on lost streams from mid-2015 to mid-2017 on a low-sized platform (Apple Music averaged 15m subscribers from 06/15 to 06/17) and 'forget' that on a 150m users platform it got a massive boost lately and that for several years it got streams on Spotify while not on Apple Music. It's the whole meaning of an average, to settle a number that averages every period. Pointing out the lone period where an artist did better than suggested by the average and say "hey, you must inflate the figure to make that up!" makes no sense.
i like the new layout of the articles. Flows much more naturally.
Btw, do you guys update the figures in the Data Collector area after every updated article? Like if you do update all of Taylor's numbers or Drake's streaming numbers, do the respective tables in data collector get updated as well? I wouldn't mind helping with that in case help is needed 🙂
Awesome update, I noticed you removed the section on the streaming charts that included total audio streams, do you still factor in Apple Music and other streaming services?
Hi Pat!
Yes it is updated after every article (we are updating it tonight with new Taylor's results). Anthony does it with automated processes that he prepared so it requires no manual changes luckily for us! 🙂
Hi Tommy!
Yes of course, they are still accounting for. The column was kinda redundant since it was an application of the ratio mentioned in the introduction, so we decided to remove it to make tables easier to digest. The formula is printed already and EAS results use it!
Hi Emiliano!
Well, actually they went up from 149 million to 153 million 🙂
There is two things that have been removed: non-pure sales of 1989 in Mexico since their certs include YouTube views and pre-release downloads of promotional singles that got dropped in anticipation of the Complete My Album functionality on iTunes. Now download figures fall perfectly in line with the last Billboard update for her career total!
I'm a huge fan of your method, Can you please do Nicki & Aretha. I really wanna see if Nicki has sold how much she brags about i her songs
Hi Lucas!
As I already mentioned, these "sales" are wishful thinkings from some fans. People think you have to pay 20 yuan to buy Reputation or 1989, it's false. You have to pay for it only if you aren't a premium user. Once you got a Green Diamond subscription, you can download it for free. That premium account costs only 10 yuans per month, so people willing to pay 20 for an album are premium users already. That's why Tencent grosses so few from their platform (in 2016 QQ Music grossed less than Deezer) and that's also why these supposed miraculous Chinese sales were nowhere to be seen on IFPI reports.
Chinese market is booming right now, something I said would happen many years ago. Still, it's still at the level of Brazil, behind markets like Australia and South Korea as of the last IFPI report. Some people speak as if it's the new US - it could be in the future, but it definitely isn't. Plus, their entire increase is made on the back of streaming subscription, they have little to no purchasing culture and Tencent own financial reports reflect that, mentioning gifting sales rather than sales. Expecting albums to be downloaded legally there more than in the US is pure delusion. You can see here that their premium account opens the right to paid-downloads stuff. As long as the IPFI doesn't treat this as sales it would be pure inflation to act as if they are legitimate purchases.
BTW, if I use the initial shipment figure of the IFPI (4,5m) along with 10/1 ratio for downloads and 1500/1 ratio for YouTube views, Reputation ends on 8,2m, 100k away from HDD's total. This shows that figures are in line with official data and that HDD too hasn't consider these supposed Chinese sales. Taylor's success there will be reflected at some point, it's only time to get more information about their streams. One shouldn't expect miracles though, that will only bring minor additions as long as China hasn't got into the very top markets globally.
Not sure I should care about it but for the sake of accuracy, here is some debunking on some of the wild delusional claims made on this typical trash-topic from ATRL.
QQ Music are pretty good in certification but the problem is Chartmaster discredit them because music is cheaper in China. He always ignore the fact purchasing power is not same in every country. The bias is real
Well, a typical ATRL member can't write a sentence without mentioning "bias" since that's their own motor, so that comment isn't surprising. I won't even comment the terrible confusion between a streaming platform (QQ Music) and an audit organization (that would issue official certifications). So the exclusion of free to subscribers 'sales' (see previous comment) is now a "bias against poor purchasing countries". As I lived in Kinshasa, I would say the real bias here is to rate China as a poor country, but let's move on this nonsense to get into more figures-related data.
Officially bigger than Lemonade or Anti. Hahahahahahahhaha
The whole irony is that this message comes from the same member as the previous one. Typical fanatics behavior: if something backs your claims, say it's "official", if it goes against your wishful thinking claim it's biased. Standard. I quote this to point out something very important: Chartmasters is NOT official and doesn't pretend to be. Official sources are limited to their commitments which, among other things, imply they have a hard time fixing their own errors (see the OCC). We are very happy to be free to go after the most accurate data rather than representing some legal entity.
First of all, let's stop pretending Chartmasters is the ultimate truth. It's not. It's a random chart blog created biased stans. Their stats are inaccurate. They use estimations and formulas they created - formulas used by no one outside of a few stans.
I'll pass on the typical fanatics-copyrighted bias claim and mention formulas. Each and every country uses formulas right now, streaming made it inevitable. The fact we use ours too is precisely why our method is relevant - we add figures from every market with a consistent formula, instead of adding apples and oranges that reflect the formula of each market rather than the true success on them. That consistent formula for all is exactly what prevents us from https://chartmasters.org/2018/03/rubbish-alert-the-ed-sheerans-divide-sold-under-11-million-thing/&sa=U&ved=0ahUKEwj88fnfqaHeAhWPsKQKHaj7DvIQFggFMAA&client=internal-uds-cse&cx=partner-pub-3868619518247857:4666236348&usg=AOvVaw3TrL4w1O1Qq0Iw7hfbFW0 q" target="_blank" rel="noopener nofollow">this kind of rubbish.
1. Chartmasters want us to believe that in the last year, Debut sold 90K, Fearless sold 135K, Speak Now sold 90K, Red sold 100K, but 1989 only sold 40K? :rofl: Obviously, he's not dumb. He's just re-evaluating his previous estimations for her past albums and saw they were off. It just goes to show how innacurate this can be.
Well, I would love everywhere source to be "how inaccurate" for 50k-ish adjustments. Now, if we scratch the surface and look at the updates, the story is different:
- 1989 grew near 200k but climbed only 40k because of the fix of Mexican certs including streams. While it was an error back in the day, Chartmasters is precisely the first place that clarified that Mexican certs were fueled by streaming. Various boards, including ATRL, though for long (and some still do) that albums were selling like hot cakes there while sales can't even gross $40 million now at more than $10 per album on average.
- Others got a 30/50k "bonus" on top of their ongoing catalog sales. Why? Last year, Billboard listed her album soundscan sales. There is fanatics trying to drag online day in and day out but none took a minute to add sales of each album and compare to the career to date total mentioned. There is indeed a gap of half a million (30,9m to 31,4m). Even assuming that the 7 albums rounded to the closest 0,1m were 40k higher it's still not enough to make up for the gap. Apart from listed albums, Swift only released iTunes Live from SoHo (which got added). So the question is where were these 300-ish thousands missing? I already got similar gaps in the past and one frequent reason is that Soundscan treats Karaoke albums apart, some of which sell very well (last time I saw a Karaoke chart with numbers, the top 15 sold 87k on average). I checked Discogs and saw that all Swift albums up to 1989 got a Karaoke version with relevant number of owners, hence the increase of about 50k each that also fills in the gap with Soundscan and makes totals accurate. It's not an evidence of inaccurate numbers, instead one more evidence that our numbers are far and away the must trustable ones since we apply fixes based on factual data were even fans haven't even notice an issue.
2. Same thing can be said about her Christmas EP, but instead, he decreased its CSPC. He had it at 1.57M last year, now he has it at 1.54M :rip:
What's the issue? Here too Discogs / YouTube show the Christmas EP was less popular than expected elsewhere and its album tracks sold less proportionally than they got streamed, why wouldn't we acknowledge it?
3. Her debut album had physical singles. His previous update had it at 6K CSPC, now he has it at 0. Why? He also reduced Fearless' singles physical sales (from 48K CSPC to 23K CSPC) and 1989's singles physical sales (from 12K CSPC last year to 9K CSPC). Again, why? Did he once again change his formula?
Well, that's the point - her debut album had no physical singles. In the past, we used to apply market shares for each country for singles issued in that transition period from 2005 to 2012. Now we got Discogs so we go through each and every release. Some singles, like You Belong to Me or Bad Blood haven't follow market shares simply because they weren't released at all in countries like the UK and Germany. Here too nobody ever bothered checking her release history of physical singles. We did and got hard facts to commit into sales figures. Are we supposed to be inaccurate just because we took the time to check every detail rather than copy/pasting something?
4. I get that he included audio streaming in his analysis, but why didn't he tell us the stats? Normally, they're always included in the streaming masters.
How is that even an issue ? It was already explained – comprehensive audio streams were results of already mentioned formula, so they were flooding users with numbers for no reason. Tables are much easier to read without them and results aren’t impacted at all. The same format has been used on articles of Drake, Lennon, Lovato too. Acting as we are trying to hide something while the section starts with the formula is a truly weak argument.
5. As always, his numbers for Asia are incredibly off. And I'm not even talking about China here... Just in general his estimations for Asia are awful. I'm very curious as to how he comes up with these numbers. In his last update, he had Taylor at 5.4M albums sold in Asia, now he barely has her at 3M He literally cut her Asian sales in half with no explanation.
They are no doubt very off from usual ATRL wishful thinking. The previous update had 2,705k, thus she gained over 300k there since the last update. A typo happened a year ago with totals and Japanese figure being doubled on totals page, but that was fixed long ago and detailed breakdowns have never been lowered. We have got numbers for virtually every relevant Asian market for all their large albums, anyone claiming 3m up to date to be “way off” will have a hard time proving it since the margin of error is close to 0.
But anyway Rep must be in the top 3 of albums released in 2017?
yes. it's top 2 for sure behind Divide
It’s not. Evolve by Imagine Dragons was slightly ahead its number already when they were published. Dua Lipa’s eponymous album has over 5,4 million from audio streams alone it’s safely ahead. I haven’t check in detail, but Maroon 5’s Red Pill Blues is ahead too or will be very soon, same for Kendrick Lamar’s Damn. There is several more albums just behind that are pacing faster right now too like The Greatest Showman. Then there is Drake, Sam Smith and many more with 4m+ that can top Reputation in the long run. Saying “it’s top 2 for sure” without even looking at competitors makes really no sense.
People acting like the biggest female album of the past 3 years is a flop, I’ve seen it all
« The past 3 years » includes 25. Then, ANTI is ahead as of now, so is Dua Lipa (haven’t bother checking Zara Larsson). Dangerous Woman is at most 200k behind. It’s good to make claims it’s better when they are accurate.
As a general note, this comment isn’t (at all) about Taylor Swift. The whole point is that accuracy matters. Whining because a figure related to your favorite artist won’t help and calling anything you don’t like biased just because you are biased won’t make you look credible, quite the opposite. There is no doubt ways of improvement for every figure posted online (including official ones), trying helping looking at accurate receipts, identifying flaws, bringing something new to the table, pointing out blocked youtube videos that got views elsewhere, etc, but spreading nonsense is nothing else than that, spreading nonsense.
In the same way, there is nothing negative about Taylor in this article. To put it into its context, it was the second CSPC article ever posted, just after Rihanna. Back then, I selected them because they perfectly reflected that the whole “album seller” vs “singles seller” was nonsense, validating the CSPC approach. A pair of years later, it’s clear for all that it’s needed to check each format to gauge the success of an album, but it wasn’t when we first published this concept. When I say that GGGB did better than every album by Taylor Swift, I’m not saying that she is a flop, instead I’m barely pointing out that pure album sales aren’t the be-all end-all of charts and sales, something that is even truer in 2018. It’s key to really understand the meaning of things rather than sticking to a fanatical perspective.
A last point is that the comments section is opened to everyone - we basically validate every comment that isn't a raw insult. Feel free to raise your concerns here and get explanations. If I use an ATRL ton I would say some may be afraid to see their ignorance exposed so they prefer trolling on an external board. It isn't the way to go. Let's be factual and improve things together rather than falling into childish behaviors!
The whole idea that China is a big huge market for paid users, are so delusional. They only goes for the free streaming ones. And foreign artists rarely got super huge there. Which is why not many top superstars were willing to tour there until recently.
Considering she released 6 studio albium as opposed to Rihanna 8, and also considering that Rihanna released like a gazillion orphan singles. I think Taylor's result is more impressive as of right now.
Yeah but Rihanna released 7 albums in 8 years that's probably why her album sales are this low compared to taylor's. She used to give more importance to singles.
