Post Audit
For X creators with 300 to 5,000 followers

Find out which of your posts actually work.

Post Audit reads your latest posts through the official X API, classifies every post by type, hook, addressee and artefact, and compares each group with your own median. Every finding comes with a number from your own account.

Have an invite? Log in with X

Free Quick Check · read-only access · delete everything with one click

2.5×
replies vs. your other posts

Contrarian posts get 2.5× the replies of your other posts.

Measured against your posts from the same weeks, they get 1.6× the usual replies (28 posts); your other posts get 0.6× (36). Per post: 5.1 against 2.1 replies.

Strong signal · 28 vs. 36 posts

From the example report of @lucas_builds

How it works

Three steps, about two minutes

  1. 1

    Connect X, read-only

    Your latest 100 original posts with their real impressions. We ask for read access only, never for the right to post.

  2. 2

    Every post gets classified

    16 fields per post: type, opening line, addressee, artefact, tone and more. The model reads your text, never your numbers.

  3. 3

    Each group meets your median

    Which groups beat your own usual reach, by how much, on how many posts. Findings, not a dashboard.

Example

What a report says

Findings from the audit of @lucas_builds, measured on 94 of his posts.

See the full example report
2.5×
replies vs. your other posts

Contrarian posts get 2.5× the replies of your other posts.

Measured against your posts from the same weeks, they get 1.6× the usual replies (28 posts); your other posts get 0.6× (36). Per post: 5.1 against 2.1 replies.

Strong signal · 28 vs. 36 posts
−65%
replies vs. your other posts

Posts about your own tool get 65% fewer replies than your other posts.

Measured against your posts from the same weeks, they get 0.4× the usual replies (10 posts); your other posts get 1.2× (54). Per post: 1.2 against 3.8 replies.

Early signal, small sample · 10 vs. 54 posts
−31%
reach vs. your other posts

Posts without an artefact get 31% less reach than your other posts.

Measured against your posts from the same weeks, their median post reaches 0.9× (40 posts); your other posts reach 1.3× (24).

Moderate signal · 40 vs. 24 posts

Post type

one of six tables in the free Quick Check
GroupPostsMedian reach vs. your medianBookmarksReplies / post
question81.2×06.5
own finding141.1×43.4
opinion191.0×13.7
Pricing

Start free. Pay once if you want the full picture.

Quick Check

Free
  • Your latest 100 original posts
  • Three findings with numbers
  • Shareable result card

Weekly Report

9 € /month
  • Your new posts every week
  • What changed against last week
  • Full Audit included

What Post Audit is

An auditor for your own posts. It checks claims against your numbers and shows its work: every finding names the group, the effect and the sample size.

What it is not

Not a scheduler, not a writing assistant, no follower tricks, no scraping. It uses the official X API with your consent and reads only.

Questions
Do you post on my behalf?

No. Post Audit asks X only for read access (tweet.read, users.read, offline.access). It cannot post, like, follow or send messages.

What exactly do you read?

Your profile (handle, follower count) and your own posts with their metrics: impressions, replies, likes, bookmarks, reposts and quotes. The Full Audit also reads your replies and, for each reply, the follower count of the account you replied to.

How are posts classified?

An AI model (Claude by Anthropic) reads the text of each post and fills in 16 fields from a fixed list: post type, opening line, addressee, artefact, tone, stance and more. It never sees your numbers. The comparison with your median is plain statistics.

Why not just use X analytics?

X shows you numbers per post. It does not tell you that your question posts get twice the replies but no bookmarks, or that posts opening with a mention lose most of their reach. That is the gap Post Audit fills.

Can I delete my data?

Yes, with one click in your account. We delete your account, tokens, posts, labels and reports, and revoke our access at X.

Who sees my results?

Only you. A result becomes public only if you create a share link yourself, and it shows only the headline findings. You can revoke the link any time.