How Long Should a Playlist Be?

I’ve been pondering recently as to what makes a good playlist length. My personal view previously was ‘at least CD length’ and whilst that sounds on the surface reasonable , i.e 10-15 songs minimum, 74 minutes of playing time, I thought i’d look at this in a bit more detail to see where the evidence really was.

1. Minimum length - Covering Shuffle & Freemium
Both Spotify and Deezer have free tier users which shuffle both albums and playlists and won’t let you play in order on mobile. To shuffle a minimum number of tracks are needed.

Spotify will add tracks onto a playlist if there are less than 15 songs.
On Deezer this is increased to 30.

So we have our starting point. On this basis playlists on Spotify should be at least 15 songs long and Deezer should be 30 in order to stop non-curated songs being added to the playlist.

2. Ideal Length- Spotify’s Approach
On Spotify each user can access a ‘Discover’ playlist, refreshed weekly, which contains a 30 song list of unheard or similar music to what you've previously listened to, tailored via algorithms to a users tastes. This is a major driver of plays. The more popular a song, the more algorithms it will populate and so on.

Spotifys research has shown that plays on the Discover playlist peak on the refresh day of Monday and tend to tail off more each day after then.

My discover list lands at 2hours (30 tracks), which is enough for my phone to take offline for most users. Part of the thinking is that Spotify have kept the length like this so that it’s just enough songs to listen to in a week before refresh but also not too much data for syncing offline to a 16GB phone.

Their other playlists seem to range from 50-70 songs (New Music Friday, All new All Now), to increased numbers on some, but not all, mood playlists. I.e Your Morning Coffee (85), Yoga & Meditation (83),  #The Office Stereo (60), Afternoon Acoustic (121)

Some initially seem curiously shorter but contain longer tracks like Natural Concentration (16 songs but 2.5 hrs playing time),  Chill Chapter (24, 2hr) and so on.

Mood playlists could be played in shops/bars or home more often, where wifi may be more readily available, so we could surmise there’s not as much demand to sync to offline on these.

3. The Expert View
In discussions with industry friends at various music companies, and colleagues, i asked them what they felt the ideal length of playlist was.

"The top 10 playlists on Spotify (discounting top 100/40 lists) have an average of 58.3 tracks and are updated at least weekly. 50 tracks at 3.5m is about 3hrs of music which is feels (no data for this) like a number that would sensible be how someone might break up their day. The longer playlists are the acoustic/sunday/coffee ones. If we're looking at this like it's radio, then it would make sense to have the music for times when we people would be more likely to not 'change the station' be longer." - Brittney Bean, Songdrop


“As for how long should be a playlist... I don't like playlists with hundreds of songs... If it's a thematic playlist maybe it's ok to have a great number of songs, but if we're talking about mood playlists, I think it's better an small selection. My favourite playlists don't have more than 50 songs! I think 50 it's the perfect number to transmit the essence, but if you are someone like me, that I'm constantly looking for new music and updating my playlists, it's difficult to control the tracks you have in the playlists hahah I usually start with 50, but I'm adding new songs frequently... “
Carlos - Indie Mono (Spotify Channel)

One told me theirs were variable depending on what each playlist needed to do and achieve but that it was mostly about good fit in terms of song choice and sensible lengths.

A curator from one channel informed me that their users liked 'slightly longer playlists' , which when i asked how long he said "over 20 , around 30-40 ideally"

4. Coffee Pit Stop..

A look at some Spotify’s own playlists listing the word ‘Coffee’ show a consistent play towards a 4 hour playlist, covering half the day, or enough time to rotate the playlist without customers noticing, which backs up the notion that mood playlists have a requirement to be longer when compared to new music or commute playlists.

Coffee Table Jazz: 4hrs, Soul Coffee (4hrs), Coffee Pop 4hrs, Your Favourite Coffeehouse (4hrs 20minutes)

Conclusions

Length is variable but mood playlists are longer, in general compared to new music playlists. So you could decide the length of time your playlist might get played for (eg. Wedding party/ dinner for 2 hours) , the context (i.e bar, coffee shop, on the move?), and whether your playlist fits into one of these three categories

• New Music (short - medium length, 30-50) *up to 2hrs
• Catalogue Music, Best Ofs (medium, 30-75)  *2-4hrs
• Mood (longer, 75+) *4hrs+

I’ve listed catalogue as medium because you could argue for and against longer playlists. I tend to keep mine at about an hour to 70 minutes length for these for the main apart from playlists like 90’s Pop Records which by it’s title alludes to a more encompassing list (this is 232 songs long and going!) so could argue for more flexibility there.

Index. A selected list of Playlist track counts..

Spotify playlist track numbers (UK) *accurate as of publish date

New Music
Discover: 30
New Music Friday UK/US/CA/SE: 50-70
One Week One Playlist: 16
All New All Now: 50   
The Indie List: 56
New Music Commute: 59

Mood
Sleep Into the Ocean: 20 (2hrs in length)
Coffee Table Jazz: 47
Your Coffee Break: 81
Folk & Indie Coffee Break: 55
Jazz & Coffee: 50
Coffee Pop: 61
Wake Up & Smell The Coffee: 35
Country Coffeehouse: 50
Caffeine Rush: 93
Singer Songwriter Coffee Break: 56
Folk & Indie Coffee Break: 46
Black Coffee & Gritty Blues: 24

Filtr (Sony) *selected
New Music Playlists: New Music (58) | Hot Right Now (68) | A List Hits (62)
Catalogue: 90s Indie (62) | Ultimate 00’s playlist (47) | 
Mood: Running Tracks (90) | New Chillout (86) | Poolside Beats (85) |Relaxing Classical (102) | Coffee Table Jazz (26) | Soul (114)

Digster (Universal) *selected
New Music Playlists: Hits! (77) | Music You Should Know (66) |New Christian Music (59)
Catalogue: Welcome To The 90s (50) | Reggae (102)
Mood: Indie Chill (113) | Road Trip: Singalong Songs (94) | Coffee Shop Sounds (101)

Topsify (Warner) *selected
New Music Playlists: Tomorrows Hits (50) | Future Dance Anthems (56) | Buzz 2016 (23) | New Music Fridays UK (64) | UK Top Tracks (100)
Catalogue: Soulful Christmas (105) |
Mood: Ultimate Workout (61) | Ultimate Love Songs (300)| Songs To Fall Asleep To (51)

Spotify- New About Section

I'm quite liking the refreshed 'About' section on Spotifys artist pages. Whereas before you had pictures and a biog text sourced from AMG, we can now see the top 5 playlists your artist was discovered in, rolling monthly plays and top 5 biggest cities.

It's been mentioned that Spotify are now into the numbers game.  If youtube and soundcloud are the rivals then it's plays not subscriptions to a artist that count first. It's no longer enough to say band X have 5000 followers, it looks a lot better to say 50,000 monthly plays and it gives a more accurate idea on how many people are really listening.

That said it still doesn't quite give an accurate picture of how many of those listens are just 'mood' or one track in a big playlist or just casual. Spotify are working on a system to see what proportion of those numbers are actual potential fans who haven't yet followed, rather than one off listens, but it's good to see.

The playlists 'discovered on' section is neat. Showing the Top 5 only -presumably as its public facing and they're not data geeks like us-, it gives an opportunity for smaller playlist channels to get profile as say a champion of brand new artist and release. Stay ahead of the game and on top of the newest releases and you might find your playlist in here.

You can see the number of unique 'discovered' plays too, so if you're in the top 5 you can see a little slice of how many listeners heard an artist for the very first time. Showing what playlist really adds for an artist fanbase and crucially, whilst we wait for Spotify to open up with more data on playlists, how many of your followers actually listen to what you add..

A very cool, useful, update from Spotify HQ.
 

Vanessa White of The Saturdays, 30,000 plays, compared to 143 followers after 2 singles

Top Cities now show who is listening in the 'Where People Listen' section. In this case New York is ahead because it's likely New Yorkers listen to..

..the Fresh Finds playlist which has added 5.6k listens in 10 days and shows in the 'Discovered On' section within About. My playlist is 3rd showing the benefit of adding a song to your playlist early in it's cycle.

Editorial Bias and Filtr, Digster & Topsify

This is a big subject and in some ways this article will pose more questions than answers but it's one I'm asked regularly

"do you pitch songs to Filtr, Digster, Topsify?"

The answer is of course yes. But there's a caveat. Filtr is Sony, Digster (now called HITS) is Universal and Topsify is what was playlists.net and owned by Warner Music Group.

A quick peruse of the playlists on these channels show that the number of tracks that aren't the label groups own repertoire are minimal.. take for instance Filtr's new music playlist which is 90% Sony  with about 10% from UMG and other artists where Sony are the label in the US, e.g Adele's Hello being included which is released and marketed by Columbia in the USA but XL in the UK.

So although a user might see 'Future Dance Anthems' playlist, what they're actually being fed is a small percentage of the music that could actually be the top dance hits because for instance Topsify will likely have a policy to not add UMG/Indie/Sony songs or others to their lists, focussing 90+% on their owned repertoire.

A decent and increasing wedge of marketing money is being set aside and spent on playlist marketing and users of Spotify , Apple and Deezer on these playlists will be getting a list picked from a limited range.

This is why, unlike Apple, it's good that both Deezer and Spotify are focussed on making User playlists available to find on search. Yes user playlists will be biased but without the pressure of a corporate decision to only include tracks from a small pool of choice.

Do users care or even know that often channels that don't explicitly say who is curating them, and deliberately exclude songs to the detriment of that list?

Many of the main channels have built up their followings with big marketing pushes funded by hefty budgets, but the cost is sometimes authenticity.

Why make a blog about playlists?

Hi! Thanks for visiting this page to journey in looking at trends within playlists and playlisting as a form.

I've started this blog as a way to connect further and ask questions about an area of music I am passionate about, interested in and involved with daily. Playlists!

Key questions I hope to pose and answer will be as much to do with the artistic form of how tracks are chosen and the editorial side as it will effective marketing techniques , good practice and asking questions on new developments.

To give you my background I grew up listening to the radio and the music in my parents car and loved it. My dad played classical music interspersed  with the Police, Camel, The Orb and Supertramp. My mum is a true child of the 60's so the Beatles era through to Enya and Kate Bush. One of my favourite radio shows was Paul Conroy on Galaxy 101 on Sunday afternoons. His blend of funk and soul massively influenced me and got me into the genre.

From this interest I learnt to DJ and mix , did stints at hospital radio and my local bbc station as well as co founding a station at UWE in Bristol called The Hub. This led to work at hmv and then into the music industry proper where I help to sell records to key accounts like Amazon.

The first place I was really able to make a playlist was burning a cd of tracks from a computer to an audio disc from iTunes. It was so easy compared to tape and mini disk. I used to make compilations of funk soul and disco, often for the purposes of having 'go to' CDs for the car or DJing.

Of course having to pay for a download and then burn the cd was a long process but I felt I owned it a bit more. I still pay for downloads alongside LPs  CDs and streaming now so in terms of my favoured platform I like elements of all of them.

Anyway enough of me.. onwards to the juicy part..