A seller showed me his AI follow up
A seller showed me his AI follow up. Nine emails in my inbox used the same line.
Same opening. Same structure. Same careful corporate tone.
He had spent twenty minutes asking ChatGPT to improve it.
The message was longer. More polished.
It still did not sound like him. This is what most people miss about AI writing.
It does not just help you write faster. It quietly removes the only thing that makes a message worth reading.
Your voice. The LLM does not know how you write.
So it falls back to the safest voice it can produce.
Professional. Careful. Generic.
Which is exactly how most inboxes now sound. That bothered me enough to try something simple.
I wrote a small markdown file that Claude reads before writing anything.
Inside it is a profile of how I actually write.
Emails. Posts. Notes.
Claude studies them and builds a rough model of my voice.
Sentence length. Rhythm. How I start a thought. How I finish one.
It also keeps a list of words I never use. The stiff phrases that give AI writing away before the reader reaches the second paragraph.
Now every time I ask Claude for something, it reads that file first. The change is subtle but obvious.
Follow ups sound like something I might have written.
Not cleaner. Not more professional.
Just familiar. I have been using it for a few months.
The editing step feels different now.
I change facts. I rarely change tone.
One seller tried it last week. He said it felt like Claude had finally met him.
Before that he rewrote every output. Now he mostly reads and sends.
Buyers read hundreds of messages every week. Most sound the same.
The ones that sound human are the ones that get answered. If you want the markdown file I use to teach Claude how I write, comment ππΌπΆπ°π² and I will send it.
β»οΈ Repost to help someone in your network π Follow Barry Flanagan for Daily Tech Sales + AI insights