#05 - Your best post is always the one you almost didn't publish.
Why the filter that protects you also makes you forgettable.
I almost didn’t publish “I Don’t Have a Voice. I Just Talk.”
It sat in my drafts for nine days. Twice I rewrote the opening to sound less rude. Once I deleted it entirely and started over. Once I considered shelving it because it felt too direct, too obvious, too “everyone already knows this.”
I published it on a Tuesday morning at 6:47am, before the rational part of my brain came online.
By that night it had done more inbound for The Ghost than three months of polished writing.
This is a pattern. Yours, too.
Every founder I work with has a folder of drafts they almost published. The thing the customer said. The opinion they were going to put on LinkedIn until they remembered they had clients. The post that started “Most people in my industry are wrong about X.” The post that started with a number that felt too specific to be safe.
Almost every one of those drafts is better than what they actually shipped that week.
The reason is simple, and it’s the reason this newsletter exists: the filter that stops you publishing the post is the same filter that strips the voice out of the post you do publish.
If you almost didn’t hit send, the post was probably good. If you hit send without flinching, the post was probably forgettable.
There’s a counter-intuitive rule in here. It’s not “publish whatever pisses you off.” It’s narrower than that.
The post you almost don’t publish is the right one when it has all three of these:
A specific moment, name, or number that you can defend. A real opinion that someone reasonable could disagree with. A cause-and-effect that explains why the moment matters.
That’s the same three things I named two weeks ago. Specific evidence. Real opinion. Named cause and effect.
If a draft has all three, the part of your brain that wants to delete it is doing exactly the wrong job. That brain is the freelancer who smooths your edges. The AI trained on the median. The internal editor rewarded for fifteen years of not being weird in public.
The brain saying “don’t post this” is the same brain that produced the average LinkedIn post you’re currently embarrassed by.
You should listen to it less.
I’m not going to tell you to “be brave.” That’s the kind of advice that makes me want to throw a laptop. Brave is not the move.
The move is operational.
Three things that genuinely fix this:
One. Write it, then take a 90-minute break before you read it again. The version of you that wrote it is the unfiltered one. The version of you reviewing it 90 minutes later is the filtered one. Whichever one wins the argument decides what gets published. Most of the time, the unfiltered version should win. You’ll know because the filtered version is always the one suggesting “softer” edits.
Two. Send the post to one person before you publish. Not a committee. Not your team. One person whose taste you trust, who isn’t paid by you, who’ll tell you the truth in one sentence. If they say “ship it,” ship. If they say “this isn’t you,” keep digging.
Three. Keep a graveyard. A folder of drafts you didn’t publish. Read it monthly. Notice the pattern. Most founders have a graveyard full of better posts than what made it to the feed. The graveyard is your evidence cabinet for what your unfiltered voice actually sounds like.
That’s it. No frameworks. No prompts. Three operational moves that override the filter.
The post I almost didn’t publish is the one bringing this newsletter the most subscribers.
Yours is in your drafts right now.
It’s the one with a specific number, a real opinion, and a cause and effect you’d rather not own in public. The one that names a person, a moment, or an industry. The one you’ve already half-deleted twice.
Hit send before the filter catches up.
That’s the work.
Sarra The Ghost 👻
If you want a second pair of eyes on the post in your drafts, that’s what a Voice Audit is