The dispatch desk — publishing & returns

LinkedIn ships itself. The rest ship honest.

Approved posts publish natively to LinkedIn — an in-house dispatcher, no browser automation, no Zapier — and video uploads natively to YouTube. For Instagram, Facebook and X we’d rather show you the manual steps than fake a connector. They’re printed below.

Start free — 2,000 credits Read the manual path ↓

Publishing on every plan · one connection per client workspace, fenced at the database

Fig. 1 — the channel matrix, as it stands today Updated when reality changes — not before
ChannelDispatchWhat that means
LinkedInPersonal profiles Native Publish now or schedule — the dispatcher posts it for you. OAuth per client workspace, tokens encrypted at rest.
YouTubeVideo Native Rendered videos upload straight to the channel — resumable upload, title, description, tags, thumbnail. Lands private so you flip it live when ready.
LinkedInCompany pages Coming soon Built and waiting on LinkedIn’s API approval — coming soon means exactly that, not “secretly broken”.
Instagram · FacebookMeta Manual — formatted Natively-formatted drafts, one click to copy, you post in the app. Native connectors are on the roadmap behind Meta’s app review.
XTwitter Manual — formatted Drafts truncated to the real 280 characters including hashtags — what you copy is what fits.
Native = the dispatcher posts it · manual = formatted copy, you post it No browser automation · no Zapier · ever
The dispatcher — engineering datasheet

Boring, reliable machinery. On purpose.

Scheduled posts are dispatched by a Postgres cron worker that wakes every minute, claims due posts atomically — so a post can never double-fire — and publishes through LinkedIn’s official API.

If a dispatch fails, the card says failed, and why. Transient errors retry on the next run; permanent ones — an expired connection, a policy rejection — stop and tell you. No silent success theatre.

Schedule it and close the laptop. That’s the feature.

Dispatch spec — LinkedInAs implemented
WorkerPostgres cron, every minute — no third-party queue in the loop
ClaimingAtomic — status-guarded update; a due post is dispatched exactly once
Failure handlingTransient errors retry next run; permanent errors mark the card failed with the reason recorded
TokensAES-256-GCM encrypted at rest, decrypted only in memory; auto-refresh with a 5-minute buffer
IsolationOne connection per client workspace — fenced at the database, not the UI
Guard rail10 publishes per hour per user — a fat-finger brake, not a quota
The manual path — said out loud

Here is exactly what manual publish looks like.

Most tools bury this. We’d rather print it: until Instagram’s connector clears Meta’s review, publishing there is a copy-paste with the formatting already done. About a minute per post — and you should know that before you buy, not after.

A fake connector costs you a client. A printed manual step costs you a minute.

Dispatch ticket — Instagram, manual~1 minute
1

Open the card’s Instagram tab — the preview shows the post as that platform expects it.

2

Click “Copy for Instagram” — the caption lands on your clipboard with hashtags already tucked below the fold . . . Instagram-style.

3

Save the artwork from the preview — it was generated at the post’s real shape, so there’s nothing to crop.

4

Paste and post in Instagram’s own composer. Done — mark the card published so the calendar stays true.

Same ticket for Facebook · X drafts arrive pre-cut to 280 characters · native connectors replace this ticket when they clear review
The returns desk — what comes back

Five numbers, on a schedule.

Published LinkedIn posts report back likes, comments, shares, impressions and clicks — synced by a quota-aware engine that respects LinkedIn’s API budget instead of burning it: fresh posts sync daily; older posts weekly; after thirty days they rest.

Not real-time. Not pretending to be.

Personal-profile posts carry likes, comments and shares today — impressions and clicks arrive with company pages, because that’s what LinkedIn’s API actually offers each. We report what the platform reports.

Returns scheduleAs implemented
Posts under 7 days oldSynced daily
Posts 7–30 days oldSynced weekly
Posts over 30 daysAt rest
API budgetQuota-aware — capped per run
Latest figures, not trend lines — each sync updates the card’s current numbers
Said plainly — how we keep this honest
On company pages

The code is written; LinkedIn’s API approval isn’t ours to schedule. Until it clears, company pages stay “coming soon” on this page — we don’t sell what a platform hasn’t switched on.

On analytics

You get the five metrics LinkedIn’s API provides, on the schedule printed above — the latest figures, not invented reach scores or “engagement grades.” When the platform gives us more, you’ll get more.

The dispatch window is open

Schedule it. Close the laptop.

Start free with 2,000 credits — connect a client’s LinkedIn, approve a post, and let the dispatcher do the boring part. Roughly a week of real production, on us.

Start with 2,000 free credits See the rate card

No card · No trial countdown · No per-client fees