The Unfinished Shelf: Giving the Work That Stopped Somewhere to Live

There's a drawer in my desk that has never opened smoothly. It holds a shooting schedule for a pilot that died at the read-through, a printed mood board for a brand film that was pulled a fortnight before the shoot, and a call sheet for a documentary that lost its funding in the last round. For years I treated that drawer as evidence of failure. Lately I've started treating it as the most honest page on my website.

The highlight reel is only half the story

Spend ten minutes on any UK creative's site and you'll see the same architecture: the finished, the approved, the launched. The showreel cut to the best four seconds. The case study with the tidy metrics. It's a good shopfront, and I'm not arguing against it. But it's also a kind of fiction — one that implies a career made entirely of things that worked.

Nobody's career looks like that. Not the directors whose work I admire, not the editors who keep getting booked, not the writers with a shelf of scripts. What separates the ones who last isn't a clean record. It's what they did with the mess.

A commissioner scanning your site isn't only asking whether you can do the thing. They're quietly asking what happens when the thing goes sideways — because on their side of the table, it often does. A page of finished work can't answer that. A small, well-framed record of the work that stopped can.

Three kinds of dead, and why the distinction matters

Not every abandoned project deserves the same treatment, and lumping them together is how you end up with a misery wall.

Killed by someone else. The commission that got pulled, the series that didn't get a second run, the funding round that closed. These are the easiest to write about honestly, because the story is simple: good work, wrong moment. The lesson is usually about process — what you'd prototype earlier, what you'd ask about budget before you said yes.

Killed by you. The short film you shelved because the edit wasn't earning its length. The pitch you withdrew because the brief was drifting somewhere you didn't want to go. These entries are worth more than the first kind, frankly. They're public evidence of judgement, and judgement is the thing bookers can't get from a showreel.

Mothballed. Not dead, just asleep. The feature treatment waiting on a producer, the podcast format parked until the right co-host appears. Include these sparingly. A dormant project with no date attached can read as wishful thinking, so be clear it's dormant and say what would wake it up.

Redaction is the whole job

The instinct when you first build a page like this is to show everything. Resist it. Most of the material in that drawer is not yours to publish, and treating it as though it is will cost you the next job faster than any gap in your CV.

The rules I work to:

• No client names without written permission. "A UK broadcaster" is almost always enough. So is "a heritage brand" or "a regional theatre."
• No talent, no crew, no contributors unless they've agreed. A face is not a design element.
• No music, no footage of uncleared material, even in a still. A single frame can carry a sync issue.
• No rate cards, no internal emails, no budget figures. Obviously. But it happens.
• Describe the shape of the project, not the client. "A six-part returnable format for a streamer" does the work without naming anyone.

Where an NDA covers the whole thing, don't publish the project — publish the lesson. "A project I can't talk about taught me to storyboard the first ten minutes before pitching" is a legitimate entry. It's honest, it's useful, and it doesn't break anything.

The format that keeps it useful

Three fields, no more:

1. What it was meant to be. One sentence, present tense, no apologising.
2. Where it stopped. One sentence, factual. "The commissioner changed strategy" beats a paragraph of grievance.
3. What I kept. The technique, the relationship, the piece of craft that survived into paid work.

Cap the page at six entries. A long list of failures starts to look like a pattern; a short, edited selection looks like a portfolio. Order them by what they teach, not by when they died.

And write in the past tense, in the third person if it helps. Distance is what stops the page reading as a complaint.

What bookers actually take from it

I've had two conversations in the past year that started with someone mentioning a cancelled project on my site. Neither was a sympathy call. One was a producer checking whether I could handle a brief that had already changed twice. The other was a collaborator who wanted to know how I talk about work that doesn't land — because they'd been in that room before and wanted to know I wouldn't make it weird.

That's the trade. You give up the illusion of an unbroken run, and in return you get to show range, judgement and a working relationship with failure. For a creative site with your own name on it, that's a better deal than another case study with a tidy metric.

The drawer still sticks. The page doesn't.