Personal Systems

A writing backlog should not be a guilt machine

6 min read

I have ruined enough writing systems by making them too virtuous.

A backlog starts as a relief. Finally, a place to put the thought before it evaporates. A link, a sentence, a weird phrase from a meeting, the outline for a post I might write later. It feels responsible. It feels like I am protecting future work from present chaos.

Then the list grows teeth.

Three weeks later there are forty-seven notes in the queue and none of them look like ideas anymore. They look like tiny obligations. The original spark is gone, but the title is still sitting there, implying that I owe it something. This is how a writing backlog turns into a guilt machine. It stops storing possibility and starts manufacturing accusation.

That is a bad trade. I would rather lose a few ideas than build a system that makes writing feel like debt collection.

the backlog is not the work

The backlog is a capture tool. It is not the writing. It is not the plan. It is not the proof that the work matters. It is a shelf.

That sounds obvious, but I forget it constantly. I will make a nice list, tag the notes, split them into themes, add a status field, and tell myself I am building momentum. Sometimes I am. More often I am arranging the pantry because cooking feels harder.

The check I use now is simple: did touching the backlog make it easier to write today? If the answer is no, the backlog is probably asking for attention it has not earned.

A good backlog should lower the cost of starting. It should remind me what I was thinking, where the idea came from, and what kind of shape it might take. It should not require a weekly ceremony. It should not punish me for changing interests. It should not preserve every half-thought like a legal record.

Some ideas age badly. That is fine. A stale note is not a failed promise. It is just a note that did not keep enough charge.

capture less context than you think

I used to think every saved idea needed a bunch of context. Source link. Quote. Category. Possible angle. Related posts. Maybe a rough outline. That can be useful for research, but it is usually too much for everyday writing.

Most ideas only need three things:

  • the sentence that made me stop
  • the reason I thought it might matter
  • the smallest next move

That last one is the important part. “Write about backlogs” is vague. “Explain why old writing notes start to feel like chores” is better. It gives me an entrance.

I do not want a backlog item to be complete. I want it to be restartable.

There is a difference. A complete note tries to preserve the whole idea for later. A restartable note preserves the doorway. Future me does not need a perfect memory of the moment. He needs enough friction removed to write the first paragraph.

delete aggressively

The best improvement I have made to my writing backlog is deleting more of it.

Not archiving. Not moving to “someday.” Deleting.

If a note has been sitting around for months and I feel nothing when I read it, it can go. If the title sounds like something a productivity system invented, it can go. If I only keep it because it once seemed smart, it can go. I do not need a museum of former interests.

This is harder than it sounds because deleting an idea feels like admitting waste. But the waste already happened if the note is dead. Keeping it around just charges rent.

The archive has a different job. Finished work belongs there. Drafts that still have blood in them can stay. But a capture backlog should be more like a workbench than a storage unit. If there is no room to move, the tool is broken.

let the backlog be uneven

The other mistake is trying to make every idea enter the same pipeline.

Some posts start as a title. Some start as a complaint. Some start as a code example. Some are just one paragraph that should never become more than one paragraph. A good backlog lets those shapes stay weird for a while.

I do not want every note promoted through the same stages: captured, outlined, drafted, edited, scheduled. That works for some teams and some kinds of publishing. For a personal technical site, it can turn small thoughts into project management theater.

The better workflow is looser:

  • quick notes can become posts directly
  • technical ideas need enough scratch space for examples
  • personal observations need time to prove they have more than mood behind them
  • old notes need an easy way out

That last part matters. Every backlog needs an exit ramp. Otherwise the only direction is accumulation.

the useful metric is pull

I do not measure a writing backlog by how many ideas it contains. That is like measuring a kitchen by how many ingredients are expiring in the fridge.

The metric I care about is pull. When I sit down to write, do any of the notes pull me back into the thought? Do they make me want to open the editor? Do they remind me of a real problem, a real annoyance, a real system I want to explain?

If yes, the backlog is working.

If no, the backlog is probably too full, too abstract, or too polite. Polite notes are a warning sign. “Thoughts on developer experience” is useless. “Good CLI output is a kind of API” has an opinion in it. I can work with that.

This is also why guilt is such a bad signal. Guilt can make me open the backlog, but it cannot make the writing better. The best posts usually come from irritation, curiosity, or the feeling that I finally know how to explain something that has been bothering me. Guilt produces dutiful paragraphs. Nobody needs more of those.

a smaller default

My current default is intentionally plain. Capture the idea. Add one sentence of context. Add the next move if I know it. Revisit it when I am actually looking for something to write. Delete it when it goes cold.

No weekly reckoning. No fake priority scores. No giant “content engine” pretending that a personal site is a newsroom.

The backlog should keep writing available, not morally urgent. It should hold the idea lightly enough that I can pick it back up, change it, merge it with something else, or throw it away without a ceremony.

That is the boundary I want: a backlog is allowed to remember. It is not allowed to accuse.

Jeremy London

About Jeremy London

Engineering leader and builder in Denver. I write about AI platforms, agents, security, reliability, homelab infrastructure, and the parts of engineering work that have to survive production.