Four Weeks Without the Dashboard: What Happens When You Stop Counting

I spent the first week checking a page that no longer existed.

That's the honest version of how this started. On 5 October I stripped every analytics snippet out of this site — the visitor counter, the referrer log, the little dashboard app I'd somehow trained myself to open while the kettle was boiling. Thirty days. No numbers. Not even a rough idea of whether anyone showed up.

I'd like to say it was a noble experiment about creativity. Really it was a reaction to a Tuesday in September when I caught myself rewriting the opening of a piece because I'd convinced myself it would perform better on a chart I don't fully understand, maintained by a company that doesn't tell me how it works.

A confession about dashboards

Analytics are a mood, not a measurement. That's the thing nobody tells you. A number going up feels like approval; a number going flat feels like being ignored, even when the number is ten and the difference between ten and nine is one person who probably clicked the wrong link.

For two years I'd been treating those numbers like weather — checking the forecast before deciding what to wear. Which is a ridiculous way to make anything. You end up writing for the version of your audience that behaves like a spreadsheet: the one who arrives at 3am from a link aggregator, reads for forty seconds and leaves. That person is real, but they're not why I do this.

Week two is very quiet

The second week was the strange one. Not because I missed the data, but because I noticed how much of my day had been structured by it. Open laptop. Check numbers. Feel a thing. Start work.

Without that, I had to find out what I actually thought about a piece, which is harder. My opinions were suddenly unassisted. I'd finish a draft and there was nothing to consult — no bounce rate to blame, no traffic spike to flatter me. Just me, a cup of tea going cold, and the question of whether the thing was any good.

There were three days where I didn't write at all. In the old system I'd have known whether that was allowed.

I also ran the whole thing straight through the end of British Summer Time, which meant the clocks went back on 25 October and my body had no idea what time it was. That may have coloured the results. Everything colours the results when you're the only subject.

What I actually know about UK readers

Here's what the month taught me, and it's mostly about rhythm rather than taste. British reading habits are deeply scheduled. There's the lunchtime window — half twelve to half one, when people are eating a sandwich at a desk and will quietly read something longer than they'd admit to. There's the Sunday evening slot, which is real and gentle and slightly melancholy. There's the post-teatime slump around seven, when the telly is on but nobody's watching it properly.

And there's August, which is a bin fire for anything that requires concentration. Everyone's somewhere else, or wishing they were.

I'd been fighting those rhythms for years, trying to be a thing people check constantly rather than a thing people visit. The month made it obvious: I'm not a feed. I'm a place you go when you've got ten minutes and a brew. That's a better thing to be, and it's much easier to write for.

There's also — and I don't think this is parochial — a particular British ear for being sold to. Readers here can hear a hard sell through three walls. The pieces that landed best, in my own memory of them, were the ones that sounded like a person talking in a kitchen. Not a brand. Not a content strategy. A bloke with an opinion about a microphone.

Three things I got wrong

First, I mistook silence for failure. Halfway through I nearly broke the whole experiment because I'd decided nobody was reading. I had no evidence for that, in either direction. I'd simply filled the vacuum with the worst available story.

Second, I still wrote for an imagined audience — I'd just replaced a spreadsheet with a mental sketch of "a reader". Same problem, softer edges. It's very easy to flatter yourself that you've stopped performing when you've only changed venue.

Third, I over-corrected into navel-gazing. Two of the pieces I published in October were so far up their own backsides that I winced re-reading them in November. Removing the numbers doesn't automatically make you honest. It just removes one excuse.

What I'm keeping

The scripts go back on, but differently. Analytics return as a weekly thing — Monday morning, in a browser I have to log into, with no app on my phone. If a number is going to ruin my afternoon, it can at least have the decency to do it once a week.

I'm keeping the paper notebook, which turned out to be the real instrument of the month. Every time someone replied to an email, said something kind in a comment, or mentioned a piece in passing, I wrote it down. Thirty days produced eleven entries. Those eleven sentences are worth more to me than a month of sessions data, and I can actually remember all of them.

And I'm keeping the rule that one piece a month gets written with no thought whatsoever to how it will land. It'll be the piece I care about most, every time, and I've stopped needing a graph to tell me that.

The kettle's on. Back to normal, mostly.