How to Auto-Push a WordPress Post the Moment It's Published
You finish a post on staging, schedule it the normal WordPress way, and expect it to show up on production the moment it goes live. It doesn't — because nothing in WordPress core connects "a scheduled post went live" to "send it somewhere else." That connection has to exist somewhere, and whether it does determines whether auto-push actually fires or quietly never runs.
The hook that actually matters
There's a tempting shortcut: hook publish_post or save_post and push whenever either fires. Both are wrong for this, for the same reason — they fire on every save of an already-published post too, not just the moment it first goes live. Auto-push built on either would re-push on every minor edit, which isn't "auto-push on publish," it's "auto-push on every save."
The hook that actually captures the right moment is transition_post_status, specifically only when the new status is publish and the previous status wasn't. That's one hook that correctly covers both an editor clicking Publish directly and a scheduled post going live through WP-Cron — WordPress fires the exact same transition either way, so one hook covers both triggers without separate wiring for "immediate" and "scheduled." It also means a routine edit-and-update of a post that was already published doesn't re-fire it at all — only an explicit Push click, or an unpublish-then-republish, sends it again.
Why this needs to be opt-in, per post
Auto-push isn't a global switch — it's a choice made on each individual post, saved as that post's own configuration, alongside the connection it should push to. The alternative (a single sitewide "auto-push everything" toggle) would auto-push things nobody meant to send automatically — a draft post type nobody connected to a workflow, a page that's deliberately staging-only. Scoping the decision to the post itself, made by whoever's actually editing it, avoids that.
What happens after it fires
Auto-push doesn't talk to the destination site directly — it just decides that a push should happen, then hands off to the same queue a manual Bulk-toolbar push or a scheduled deployment already uses. That matters more than it sounds: the actual validation, the reference resolution, the conflict detection — none of it is auto-push-specific code that could drift out of sync with what a human clicking "Push" gets. It's the identical pipeline, just triggered a different way.
From there it behaves like any other queued item: most failures get retried with increasing delay before giving up, but a handful of error types — an invalid package, a failed authentication, an unsupported post type — never get retried at all, since retrying something that will never succeed just burns through the backoff schedule for nothing. There's no popup or notice in the editor either way; the result shows up on the Queue page, the same place every other deployment's status lives.
The part that's easy to get wrong
Auto-push still depends on something actually triggering WordPress's own scheduler afterward. WP-Cron's built-in scheduler only runs when someone visits the site — on a low-traffic staging site, a post scheduled for 3am might not actually fire until someone happens to load a page well after. The fix isn't in PostDeploy; it's a real server-level cron hitting wp-cron.php on a fixed schedule instead of relying on site traffic to trigger it. If a scheduled post "didn't auto-push," checking the Queue page first — is the item sitting there queued but not yet processed? — usually answers whether it's a WP-Cron timing issue or something else entirely.
Try it
Auto-push on publish is a Pro feature. PostDeploy Free moves Posts, Pages, and custom post types on demand. PostDeploy Pro adds auto-push, ACF/Elementor/WPML reference resolution, and Rollback.