Glossary
Plain-English definitions for every term Postbrander uses. Useful if you're trying to work out what we mean by "branded series" or what a "needs_refresh" verdict actually does.
- Anti-AI score
- A 0-100 score given to every post that measures how AI-ish the writing reads, against 90+ banned words and 30+ banned sentence patterns. Anti-AI score, explained →
- Auto-fix
- One-click rewrite of just the sentences in a post that tripped an anti-AI rule. The rest of the post stays as you wrote it. Costs one AI generation.
- BYOK (Bring Your Own Key)
- Connect your own Anthropic API key on the Basic plan and the monthly AI generation cap goes away. You pay Anthropic directly.
- Branded series
- A reusable post format with a recurring title like 'BURF'S QUICK WINS #27'. Postbrander remembers the last number used and increments automatically.
- Fact-check
- Web-search-backed check on a draft's claims and timing. Returns one of three verdicts: fresh (auto-queue), needs_refresh (rewrite in your voice), contradicted (replace). Fact-check, explained →
- First-comment seeder
- Pre-write a follow-up comment on a post. Postbrander posts it from your account the second the main post goes live. LinkedIn rewards early author engagement.
- Format for LinkedIn
- AI button in the post editor that splits walls of text into short paragraphs, blank lines between them, bolds the hook, converts mid-sentence lists into bullets. Doesn't change wording.
- Generation modes
- Standard: fast, no research. Research: web search before drafting. Premium: research + self-review pass. Each costs different amounts of your monthly quota.
- Hotstreak
- Anonymous leaderboard of Postbrander users ranked by current posting streak. Visible at /leaderboard. Handles are deterministic + anonymised (e.g. BoldFox42).
- Negative parallelism
- The 'This isn't X. This is Y.' sentence pattern. The single biggest AI tell, banned by our anti-AI engine across all generations.
- Onboarding checklist
- Five-step setup widget in the sidebar that hides when you finish: connect LinkedIn, set voice, add topics, generate a post, publish your first.
- Post coach
- Live panel under the post editor that scores length (per format), hook quality, hashtag count, ending style. Format-aware: polls don't need 2,000 chars, the coach knows. The Post coach →
- Recycle
- Take a high-performing past post and rewrite it with a fresh angle (different hook, different framing, same core idea). Available at /recycle.
- Series numbering
- When a post uses a branded series prefix, the AI looks at recent posts for the highest number used and increments automatically. No more #25 jumping to #32.
- Temporal framing check
- Part of fact-check: flags posts that use 'today' / 'just announced' / 'this week' framing when the actual event is 14+ days old.
- Topic bank
- Your list of recurring themes you want to be known for on LinkedIn. Drives what Postbrander writes about when you click Generate without a specific prompt.
- Variants
- Three structurally different versions of the same post (story-led, data-led, contrarian). Not three rolls of the temperature dial — actual structural difference. Pick the one that fits.
- Verify-after-publish
- After publishing to LinkedIn, we fetch the post back to confirm it actually exists. Catches silent removals by LinkedIn's content moderation (e.g. short Giphy MP4s).
- Voice profile
- Configuration that trains the AI on your specific writing style: tone, words to avoid, example posts, hashtag preferences, length range. Setting up a voice profile →
- Voice-to-post
- Tap the mic, talk for 60 seconds, get a polished 2,000-character LinkedIn post in your voice. Faster than typing, captures phrasing better than a topic prompt.
Missing something?
If you've hit a term in the app you don't understand, email james@seoburf.com and we'll add the definition here.