For the best experience, please enable JavaScript in your browser.

Skip to main content

Write LinkedIn Posts in Your Own Voice, Not a Bot's

By Lisa

Write LinkedIn Posts in Your Own Voice, Not a Bot's

To write LinkedIn posts in your own voice with AI, you have to show the AI your voice first, because a blank prompt gets you the average of every post it has ever seen. Average sounds like "I'm thrilled to share." That isn't you, and your connections can tell.

Why do AI LinkedIn posts sound generic?

AI posts sound generic because the model has nothing specific to copy, so it reaches for the safest phrase every time. The safest phrase is the one a thousand people already wrote.

You can spot it in three places: an opener that announces feelings ("Excited to share"), a middle that lists lessons in a tidy row, and a closer that asks "What do you think?" Nothing in it came from your week. A reader scrolling at 8 a.m. recognizes that shape before they finish the first line.

How do you write LinkedIn posts in your own voice?

Give the AI examples of how you actually write, then give it a brief with facts only you know. There are three sources worth using, in this order of usefulness.

  1. Your own posts. Paste five or six that sound like you on a good day.
  2. Your LinkedIn data export, if you have a long history and don't want to pick by hand.
  3. Writers whose style you want to be closer to. Pick up to three, and treat them as a direction, not a costume.

SocialFaktory's LinkedIn writer takes all three. You can teach it a voice from up to three roster writers, from your own LinkedIn posts, or from your LinkedIn data export. The roster holds forty real writers whose voice was learned from their X posts, tagged contrarian, storyteller, data-driven, build in public and educator. Teaching a voice costs nothing, and you can change it any time.

What are filler phrases, and how do you catch them?

Filler phrases are stock lines that fill space without saying anything, and you can catch them by asking "who else could have written this sentence?" If the answer is anyone, cut it.

Weak: "I'm thrilled to announce that after months of hard work, we've launched our new onboarding flow."

Rewrite: "We shipped a new onboarding flow on Tuesday. 31 of the first 50 users finished it."

The first has a feeling, an effort and an adjective, and no fact. The second has a day and a number you could check. The numbers here are examples; use yours.

SocialFaktory's LinkedIn writer runs a gate for this: every draft is checked against a list of filler phrases and the 3,000-character limit, and written again when it fails. That saves you the first pass. It doesn't replace the second one, which is yours.

What should your editing pass change?

Your pass should change three things: the first line, one detail only you know, and the last line. Those are the places where a post either sounds like a person or doesn't.

Before: "Leadership is about listening. Here are three things I learned from my team this year."

After: "My team told me our Monday meeting was a status update nobody needed. They were right. We replaced it with a 15-minute call on one problem."

The after has a specific meeting, a specific change and a person's pushback in it. Add one detail like that to every draft, and read the post aloud. If you wouldn't say a sentence to a colleague over coffee, change it. LinkedIn hook examples can help with the first line.

What is the checklist before you publish?

Run six checks before you publish, and you'll catch most of what makes a post sound machine-made.

  1. The first line contains a fact, a number or a scene.
  2. Every sentence is one you'd say out loud.
  3. One detail could only have come from you.
  4. No "thrilled," "excited to share" or "game plan" openers.
  5. The post ends on a point, not a question to the room.
  6. The preview looks right.

Preview shows the post exactly as it will look on LinkedIn. Nothing reaches your LinkedIn feed until you press Publish or schedule it. If you plan to queue posts for the week, how to schedule LinkedIn posts covers both LinkedIn's own scheduler and SocialFaktory's.

Get started and teach SocialFaktory your voice with five of your own posts.

In one sentence: Write LinkedIn posts in your own voice by teaching the AI your real posts, catching filler phrases, adding one detail only you know, and rereading the draft aloud before you publish.

Frequently asked questions

Why do AI-written LinkedIn posts sound the same?
Without examples of your writing, the AI reaches for the most common phrasing. Voice samples and specific facts in the brief give it something else.
How many of my own posts should I use to teach a voice?
Five or six that sound like you on a good day is a workable start. More is fine, and the LinkedIn data export covers a long history.
What voice sources does SocialFaktory support for LinkedIn?
Up to three roster writers, your own LinkedIn posts, or your LinkedIn data export. You can change the voice any time.
What is the roster?
Forty real writers whose voice was learned from their X posts. They are tagged contrarian, storyteller, data-driven, build in public and educator.
What is a filler phrase?
A stock line that takes space and says nothing, like "I am thrilled to announce." Every draft is checked against a list of them and rewritten if it fails.
Will the post go live before I see it?
No. Nothing reaches your LinkedIn feed until you press Publish or schedule it, and the preview shows exactly how it will look.

We use cookies to keep you signed in, and our own cookieless measurement to see which pages actually help. Accept and we also place advertising cookies from Google, Meta, TikTok, Reddit, Pinterest, LinkedIn, Microsoft and OpenAI, which measure the campaigns that bring people here. None of them load before you accept. Read our privacy notice