Nobody Left

The longest, strangest letter I've sent all year, and what happened after I pressed send

By Rafi Mercer

There's a rule in this business. Nobody writes it down, but everybody follows it.

Keep it short. Get to the point. Three lines, a picture, a button. People are busy, attention is expensive, and every extra paragraph is a reason for somebody to leave.

I've been in digital work since 2000 and I've heard that rule in every room I've ever sat in. It's usually delivered kindly, by people who have the numbers to back it up. Sometimes they do have the numbers. Shorter emails often do perform better, in the way that shorter everything often performs better, which tells you less about people than it does about the things being sent to them.

On Friday I did roughly the opposite.

One record. No headings. No bullet points. No offer in the first screen. About two thousand words that began with waking up with Sade already playing and ended somewhere near Compass Point, via Paris, a tango, and an accordion I hadn't expected to find. And then a question at the bottom, asked properly rather than as a growth tactic: have you ever woken with a sound in your head rather than a song?

Then I pressed send and went and made coffee, which is what you do.

I don't watch opens. Opens tell you almost nothing now, and they've told you even less since the mail clients started opening things on people's behalf. What I watch is departures. The unsubscribe number is the only fully honest thing in the whole apparatus, because it costs the reader something to do it — a small deliberate act, two clicks, a decision. Nobody unsubscribes by accident.

This week, nobody left.

Not one member cancelled. On the wider list, a handful of people went, the sort of number that's indistinguishable from ordinary housekeeping — inbox clear-outs, addresses that have stopped being used, people who signed up for something they'd already decided they didn't want.

And a few notes came back. Not many. Enough.

That's the part I've been turning over since.

Because the received wisdom says that email should have cost me people. It was long. It was slow. It wandered. It didn't tell anyone what to do. It was, by every measure I was taught, an inefficient use of somebody's Friday morning.

And yet the week I asked most of my readers is the week fewest of them reached for the door.

I don't think this is a mystery. I think we've simply got the diagnosis wrong, and we've had it wrong for about fifteen years.

Attention isn't scarce. Attention is unasked for.

People aren't sitting there guarding their concentration like a dwindling fuel supply. They're perfectly capable of giving forty minutes to something. They do it constantly — films, matches, arguments, novels, long walks, entire afternoons lost to a task nobody asked them to do. What they've stopped doing is giving forty minutes to things that don't deserve it, and almost everything that arrives in an inbox doesn't deserve it. So they've become quick. Ruthlessly, sensibly quick.

Then something arrives that was actually made, and the quickness switches off.

I keep meeting this idea from different directions, which is usually the sign that it's true rather than convenient.

It's the whole premise of a listening bar, for instance. Somebody decides to open a room where the main activity is sitting down and hearing a record properly, at volume, without conversation as the primary event. On paper that's the least commercial idea of the last thirty years. Nobody has time. Nobody sits still. Everybody has the entire recorded history of music in their pocket and no patience left at all.

Except the rooms keep opening. Tokyo, Paris, Philadelphia, Marseille, cities where you wouldn't have predicted it at all. People pay to sit in the dark and listen to an album in the correct order. Some of them travel to do it.

The prediction was wrong because it was measuring the wrong thing. It measured how quickly people click, and concluded that's how people are.

It's also how the record works, which is what made the whole week rhyme. Nightclubbing is not a fast album. It's built on space — gaps held open on purpose, a voice standing several feet back from the microphone, a bassline given more authority than the singer. Nothing about it hurries to meet you. Play it as background and it flattens into mood and you'll wonder what the fuss was. Give it a room, a volume, and forty minutes of not doing anything else, and it turns into architecture.

The record demands the same thing the letter demanded. And the people who subscribe to a thing called Tracks & Tales are, it turns out, exactly the people who were always going to say yes to that.

Which is perhaps the least surprising discovery I've ever made, and it still took me by surprise.

There's a version of me — the version that spent two decades optimising things — who would have run that email through a checklist before sending it. Cut the opening. Move the record to the top. Add a subheading so it can be scanned. Trim it by forty per cent. Put a button after the third paragraph.

Every one of those changes would have improved a metric. All of them together would have destroyed the thing.

Because the length wasn't the risk. Length is never really the risk. What people actually leave over is emptiness — the sense, arriving about ten seconds in, that nothing is being given and something is being asked for. You can deliver that in eighty words. Most brands manage it daily.

The notes that came back were the part I hadn't accounted for. You don't get replies from an efficient email. Efficient emails are designed to be resolved: read, clicked, deleted, forgotten. Nobody writes back to a resolution. They write back when something has been left open, and a question at the end of a long slow letter is about as open as it gets.

So next Friday I'll write another one. It'll probably be long. If the record needs a thousand words it'll get a thousand, and if it needs two thousand it'll get those, and the button will sit where it always sits, at the bottom, waiting for whoever wants it.

The rule is still there. Everyone still follows it. Shorter, faster, clearer, fewer words, more buttons.

I'd just point out what it's actually asking you to believe — that the people reading you can't sit still for ten minutes. Say that out loud and it sounds like contempt, because it is.

Mine sat still for a record they mostly remember as a photograph.

Nobody left.

Why measure unsubscribes rather than opens?

Because opens have been unreliable since mail clients began loading images on the reader's behalf, and because leaving costs something. Unsubscribing is deliberate — a decision, two clicks, a small act of judgement. It's one of the few numbers in email that still means what it appears to mean.

Does long-form writing actually work for a music publication?

It works when the subject earns it. Length is not a strategy on its own; it's a consequence of having something to say. A record like Nightclubbing has enough underneath it to sustain two thousand words. A press release doesn't sustain eighty.

What is the Listening Club?

One album, played in full, listened to properly, with interludes and a little context and nothing else. It runs monthly and it's online, so it happens wherever you are. The only requirement is a willingness to sit still for the length of a record.

For listening recommendations, city guides, and the quieter corners of music culture, find me at Tracks & Tales throughout the week — or subscribe and I'll write to you on Friday.

Every month, The Listening Club gathers around the world. Join here.