Understanding Governed Publishing

Why StretchSocial is "governed"

StretchSocial runs in a governed, dry-run publishing mode. That means the app is the system of record for your social plan — drafts, approvals, schedules, connections, and queue jobs — while the final step of pushing content to a live network is controlled centrally. Nothing is posted to a social network autonomously, and nothing bypasses human approval. This is deliberate: for agencies and multi-brand teams, an accidental or unreviewed post can be costly, so StretchSocial makes every step explicit and auditable.

The post lifecycle

Every post moves through a clear set of states. Three related fields track its progress:

1. Workflow status — draft → in_review → approved (or blocked if rejected), and scheduled once dated.
2. Approval status — pending → needs_approval → in_review → approved or rejected.
3. Publish status — not_scheduled → scheduled → published.

A typical journey: you create a draft, send it to review, a reviewer approves it, you schedule it (which moves it toward the queue), and — once live delivery is enabled for that network — it is published. A rejected post is blocked and stays out of publishing until it's revised and re-approved.

What the publishing gate allows and blocks

Behind the scenes a publishing gate decides which actions are permitted. In the current governed mode it allows planning and preview actions but blocks live platform writes:

Allowed (dry-run / planning):
- Reading overviews and insights
- Drafting posts
- Previewing schedules
- Inventorying platform accounts
- Checking platform adapter and gate status

Blocked until an administrator enables them per network:
- Live publishing to a social network
- OAuth token refresh
- Direct platform API writes
- Uploading media to a platform
- Deleting a platform post
- Replying to comments
- Sending DMs
- Mutating ad campaigns

Because of this, a post you "schedule" is fully planned and queued, but the actual hand-off to the network happens only once live publishing is switched on for that channel.

The algorithm contract

StretchSocial's intelligence features follow a strict contract. Read-only summaries — campaign summaries, platform-fit summaries, caption-quality review, approval-readiness, and queue-health — are available now. Predictive features such as publish-risk scoring, best-time recommendations, campaign forecasts, audience-fit scoring, and content-reuse suggestions run in dry-run only. Crucially, autonomous mutation is off and human approval is always required — the AI advises, people decide.

How live delivery gets enabled

Activating live publishing to a specific network is an administrator task, not a per-user toggle. It requires several things to be in place for that platform: OAuth and token storage configured, the network's app review or business-account requirements satisfied, and the platform adapter moved from planned to active. Your StretchSuite administrator coordinates this. Until then, StretchSocial gives you a complete, reviewed, scheduled plan that's ready to go the moment a channel is switched on.

Concrete example

Your team drafts a product-launch post targeting LinkedIn and Instagram:

1. The post is created as a draft and sent to review.
2. A manager approves it in the Approvals tab.
3. You schedule it for launch morning; it becomes scheduled and appears in the Queue as two jobs.
4. Live publishing is enabled for LinkedIn but not yet for Instagram, so on launch day the LinkedIn job can be delivered while the Instagram job waits — safely, and visibly, without surprising anyone.

Tips

• Think of scheduling as "lock in the plan," not "post now." Delivery is a separate, governed step.
• Because approval is mandatory, build the review step into your calendar — approve the day before you schedule.
• Ask your administrator which networks currently have live publishing enabled so you know what will actually go out.

FAQ

Does scheduling publish immediately? No — it sets a publish time and queues the post. Live delivery is governed centrally.

Can the AI post on its own? No. Autonomous mutation is disabled and human approval is always required.

Why is a network "planned"? Its live adapter isn't active yet. You can still plan, approve, and schedule for it now.