On Thursday, August 21st 1997 Oasis released their third album Be Here Now. As befitted their status as easily the biggest rock band in the world at that time, it was a record which saw a phenomenal level of demand. Just how phenomenal is reflected in its documented sales – 356,000 copies on its first day and 696,000 copies by the weekend. The highest single-week sale achieved by any album in chart history ever. And all in just three days rather than six (or seven). This was the all-time peak of the CD album, a genuine high water mark in consumer demand.
This week the Number One album is a collection of songs from the 1930s and 1940s sung by a 73-year-old man whose career began back in the 1960s. Bob Dylan’s Shadows In The Night sold just 22,031 copies to beat the rest of the market to the top of the charts this week. Meanwhile, lower down the eighth biggest-selling album of the week is a 30-year-old release by a long-defunct rock band. 8,890 people bought Brothers In Arms by Dire Straits last week, most of them because Google Play was essentially giving away downloads at 99p a time.
Whilst releases by the biggest names of the moment can still move product, both Ed Sheeran and Sam Smith sold over one million albums each during the course of 2014 after all, nobody is immune to the winds of change. When Take That stormed to Number One with III last December its first-week sale of 144,358 was impressive. But in contrast to the 518,601 sold by its predecessor Progress when it was first released just over four years earlier, it was a negligible figure. Their fans hadn’t gone away, the recent Number One single the group landed had proved that. They had just moved on to other ways to appreciate their idols’ brand new music,
Last week in the UK, the total size of the album market was 1,204,723. This week five years ago there were 2,034,416 purchases. That’s almost half the market that has evaporated and vanished in the blink of an eye. If you know someone who used to run a record shop you may now also understand why they don’t do it any more.
Put simply, everything we used to know is wrong. Whilst the idea of a physically packaged CD album will never totally die (just as vinyl sales stull chug along to a hardcore of aficionados) a generation is growing up that will never ‘own’ the music they consume. If I want to buy a chart CD I either have to order online or travel miles out of my way to one of the last remaining HMV outlets (where I am confronted with a display showing me what is “trending”).
Yet there is one section of the market for music consumption that is booming. And it hardly needs a nice graphic supplied by the Official Charts Company to explain what that is.
As I’ve written before, the new digital age has further diluted the whole picture of what precisely makes “an album”. A collection of songs by a single artist under a common title, fine. Yet for many acts this collection of songs is something that can be tweaked, remixed, added to and repackaged in an ever-growing array of Deluxe, Special, Version 2.0, Single, Double and Bonus editions meaning that in many cases the album you bought on the day of release may only bear a passing resemblance to the one you buy for a relative a few months down the line at Christmas. The album as a work of art is also dying medium. It is all about the collection of songs and exactly which ones people are consuming at any one time.
Hence the announcement today that the Official UK Album chart is to encompass streaming data in a similar manner to its bigger brother the singles chart did last summer. Whilst streams of individual songs will continue to register for the singles chart, collected streams of tracks from the same album will now be totalled up and combined with downloaded and physical purchases, albeit in a slightly more convoluted manner than we are used to.
This is mainly to avoid a problem that befell the Billboard 200, the chart of album sales in America which introduced streaming at the end of last year and by common assent got it wrong. The problem is that many albums are defined by one or two super-hits, the core tracks that have become big hit singles and which people return to again and again. This is an unrepresentative skew and uncorrected will end up with an album chart that starts to mirror the singles chart – as they are essentially both tracking the same thing.
Hence the brand new look album chart will log streams according to a specific set of criteria. Only the 12 most popular tracks from “the standard version of the album” will count. Streams of the two largest (presumed to be the hit singles) will be down-weighted in line with the average of the rest with the new totals then combined with sales data on a 1:1000 ratio. In other words, an album will have to receive 1000 weighted streams of any combination of its tracks to register the equivalent of one purchased sale.
The fact that there is now a formula to take into account will have a few people scratching their heads, brought up as we have been on a modern chart era where all sales are tracked and all count equally. What is not widely appreciated is that this is a fairly recent innovation. Prior to the 1990s, the chart survey represented a smaller subset of the market and sales were logged in terms of weighted "panel sales" according to where they had been made and in what shop. All that has now changed is that the Official Charts Company are being transparent and open about the methodology used to calculate sales in this brand new market.
Last summer taught us that the effect of this new data on the chart listings is actually quite limited. The stuff that streams well is the stuff that also sells well which should hardly come as a huge surprise. This is a process of evolution rather than revolution and is simply a case of the methods of chart compilation being ahead of the curve and anticipating what is sure to be a growing trend.
Those who still sit coveting their extensive CD and LP collections will doubtless once more see it as the end of days and a further nail in the coffin of whatever traditions they believe the music industry should cling to. But technology marches on. The way people consume music is changing and thus the way we track what is popular must change as well. My four-year-old daughter enjoys me selecting a CD to listen to from the large cupboard in the living room. But I know for a fact she is unlikely ever to purchase one for herself. If future music historians are to want a record of what her generation enjoyed listening to, they will need a music chart which properly reflects the way they listen. The new album chart is a crucial step along that road.
If you want to take away one positive, consider this: it will now be in the interest of everyone, artist and label, for all tracks from an album to be streamed and listened to as much as possible. Perhaps more than ever the idea of "all killer, no filler" will start to penetrate the thinking of the industry. No more padding out an album with half-finished material that is there to bolster the running order. Every single track has to count - because put bluntly every single track now really does count. And that can only be a good thing surely.
Streams of album tracks will be incorporated as from February 23rd 2015, the first such chart published on March 1st. Number One will be Ed bloody Sheeran again won’t it? I just know it.
ADDENDUM: Just to demonstrate that there are always unasked questions, talk of logging the "12 most popular tracks" from an album prompted one reader to wonder how that affects catalogue albums that have less than 12 tracks on them. Chart fan and my personal fact-checker Ben Cook wrote to the Official Charts Company to raise that very point. They told him:
Where an album has fewer than 12 tracks the two with the most streams are levelled to the average of all remaining tracks, and the sum of all tracks divided by 1000 equals the albums streaming figure
The actual analysis for albums with fewer tracks is quite extensive and complicated - we tested this with a large sample of albums and as an example;
· The last Foo Fighters album was only 8 tracks long
· The average across the 8 tracks is higher than that (on average) across a 12 track album, which means the top two get levelled “less”, therefore the sum of the 8 is not dissimilar to that of a 12+ track album