30
Jan 15

Sclerotica

FT9 comments • 581 views

Before I start writing about Number Ones again, a quick bit of stattery around the current state of the charts. This is an extremely wonkish post, so reader beware.

The official Top 100 Streaming chart was launched 30 weeks ago. Let’s see what’s up with it.

Of the 100 songs in the streaming chart.

44 have been in it for 30 weeks (i.e. since the first ever chart)
38 have been in it between 10 and 29 weeks.
18 have been in it for fewer than 9 weeks.

Compare this to the official chart – which incorporates the streaming chart, of course.

33 in it for 30+ weeks
36 in it between 10 and 29 weeks.
31 in it for fewer than 9 weeks.

I should have done that as a graphic or something. Basically, streaming charts are slower. As streaming grows as a proportion of “sales”, it’s likely to slow the charts down further, as the die-off of hits gets longer.

How about just looking at the Top 40?

For the streaming chart using the same definitions as before, you get 15 out of 40 sluggards (vs 7 in the main chart), 10 out of 40 newbies (vs 14 in the main chart), and 15 in the middle (vs 19). Again, a big skew towards long-stayers in the streaming chart – maybe even bigger than looking at the full 100. And streaming is going to grow in importance.

None of this is reflected in the Number Ones, because record label marketing teams are usually very good at getting hits to a high initial placing, so there’s a high turnover of Number Ones that’s completely atypical of the sludgey inertia below.

There’s an argument that a slow chart reflects listener behaviour and taste better than a faster one. That’s true, but only up to a point, and I think the streaming chart goes well past that point. There is an average wear-out time for any given hit – the number of weeks people want to keep hearing it. There is also a distribution curve for each hit – not everyone is introduced to it at once, and new people keep on discovering it.

So imagine you have two people who both like Hozier’s “Take Me To Church”. One heard it in the first week of release. The other in the 14th. And let’s say that for a Hozier fan, 15 weeks is about the die-off time before even they are tired of hearing it. What it means is that the first Hozier fan is getting tired of “Take Me To Church” just at the time the second fan is getting enthused about it.

Record shops – and even online shops – have always favoured the first person, explicitly (in the days when rack space was at a premium) and implicitly (in the days when homepage real estate is). Even radio playlists – which the UK charts have never included – have some commitment to refreshing themselves. Streaming services do no such thing – the music is there, free, all the time, catering for fans all along the distribution curve. That sounds very fair. Except it means that everyone at the front of the distribution curve – who hear songs early – now have to wait for a song to work its way along the curve before it goes away. Eventually, enough of the people who will ever like Hozier will have heard Hozier and got tired of Hozier, and Hozier will drop out of the charts. But that might not be for a very long time.

Short version: streaming is a subtle flip in the way the charts work. Sales charts have always been geared towards reflecting people who pay attention to new music and come to a song quickly. Streaming charts are equally geared towards reflecting people who don’t pay attention to new music and come to a record late. There is nothing wrong with those people! But the result is a slower-moving chart and hits whose lifespan is a function of their slowest fans, with chart stays extended beyond most individual fans’ tolerance. The question of what a weekly chart is FOR – or at least, who its audience is – becomes more and more open.

(This may have a bizarre – and surely unintentional – outcome. The behaviour familiar to a generation of British pop fans – listening to the Top 40 to hear the latest music – is now actually penalised. If you listen every week to the charts, you will run through your tolerance for individual hits with the bulk of their lifespan still to run. A slow moving chart rewards occasional dips more than loyal following.)

Comments

  1. 1
    James BC on 30 Jan 2015 #

    Good piece. There’s an obvious unfairness about streaming: if I buy Uptown Funk and listen to it fifty times a week for a year, the chart still only counts one sale in the week I bought it. But if I stream it, the chart registers the streams all year.

    I would suggest only allowing a particular streaming account to give chart points to a particular track for one week – say the week where it makes its 10th play of the song. You’d increase the weight given to streams in the week they counted to make up for them not counting at any other time – maybe even make that 10th stream equivalent to a whole sale.

    The broad question is: exactly what do we want the charts to measure? No one has a very specific answer to this, but the current streaming policy is too close to a dreaded airplay chart for my liking.

  2. 2
    Pete Baran on 1 Feb 2015 #

    There needs to be a bit of a look at streaming behaviour, particularly around playlists. If I add a track to my “current pop” Spotify playlist (which has about 250 tracks on it) it is likely to last a year, possibly only being played once a week. But I wonder how many people have a top 20 playlist, and if it gets at least one play a week.

    This was always an airplay issue I wondered about. Did playing the number one as part of the chart countdown count towards airplay?

  3. 3
    Auntie Beryl on 1 Feb 2015 #

    I listen to a stream of Rotterdam Termination Source’s ageless “Poing” on Spotify via my laptop at home: this play counts towards the chart.

    I download “Poing” from Spotify onto my phone to listen to whilst in transit, as wifi coverage is poor around my way: these plays do *not* count towards the chart.

    So the figure contributing to the chart isn’t even bulk dumped data from Spotify et al. It’s a subset. Worra swizz.

  4. 4
    Tommy Mack on 1 Feb 2015 #

    Poing for next Xmas #1. Only 326 days left: Get streaming.

  5. 5
    mapman132 on 2 Feb 2015 #

    Welcome to the world of a Billboard Hot 100 follower :P

  6. 6
    Tom on 11 Feb 2015 #

    An update, two weeks on. The number of perpetuals in the Streaming Chart is down from 44 to 39. They’re thinning out!

  7. 7
    weej on 24 Feb 2015 #

    Just wanted to note this here and see if it seems prophetic or naive or whatever at the end of 2015.

    * Uptown Funk had seven weeks at number one – the most since Bleeding Love in 2007
    * Love Me Like You Do has three weeks now – one more and we’ll go into March with only one new #1 for the year, which hasn’t happened since 1993.
    * Will this continue to mean a massive slowdown in #1s like we had in 2007, or is it a case of ‘small sample size so far’? In theory the slow turnover of the chart and diminishing interest in chart placings should lead to this eventually, but now?

  8. 8
    Tom on 24 Feb 2015 #

    I’d be a bit wary of judging – last year Happy and Rather Be kept things slow (albeit not THIS slow) and it sped up again. It might be a case of labels not having much on their schedules at the start of the year – which itself feels like a holdover from old, retail-era, industry practice.

    You would expect a gradual transition to a US style slow chart, though.

  9. 9
    Tom on 22 May 2015 #

    So, 46 weeks since the streaming chart was introduced. 13 songs remain from that first top 100, those being:

    Ed Sheeran – Thinking Out Loud
    Sia – Chandelier
    Sam Smith – Stay With Me
    George Ezra – Budapest
    Clean Bandit – Rather Be
    John Legend – All Of Me
    Mr Probz – Waves
    Ed Sheeran – Don’t
    Ed Sheeran – Sing
    Ed Sheeran – I See Fire
    Ella Henderson – Ghost
    Ed Sheeran – Photograph
    Pharrell Williams – Happy

Add your comment

(Register to guarantee your comments don't get marked as spam.)


Required

Required (Your email address will not be published)

Top of page