Marketing

Content Repurposing Automation: Preserve the Source Before Changing the Format

For a content or marketing-operations owner turning one approved article, webinar or recording into channel-specific drafts without losing claim provenance.

Uli PrantzBuilds and operates all-agents
Published
Key takeaways

Key takeaways

  • Freeze an approved source version, claim ledger, rights record and channel brief before generating derivatives.
  • Repurposing may compress, reorder and reframe supported material; it must not manufacture a new fact, customer result, quote or product claim.
  • Generate drafts into review states. A named editor approves each channel asset and the publishing action separately.
  • Retain source spans, draft versions, reviewer edits, publication IDs and a withdrawal path so one corrected claim can be found everywhere it traveled.
Automate content repurposing from an approved, versioned source—not from a topic prompt. Extract a claim ledger with source spans, select supported material for a channel brief, generate a draft within declared transformation rules, validate citations and prohibited additions, then require editorial approval before publishing. Preserve lineage and publication IDs so a corrected or withdrawn source claim can be found across every derivative.

The source packet is the contract for every derivative

A usable packet separates source truth from channel instruction.
FieldExampleControl
Source versionApproved article or transcript hash and dateReject stale or unapproved input
Claim ledgerClaim, exact source span, citation and qualificationNo unsupported additions
Rights recordOwner, consent, permitted media and expiryBlock unapproved reuse
Channel briefAudience, format, length, voice and desired actionAdapt shape without changing facts
Prohibited movesNew statistics, invented quotes, guarantees and unapproved product claimsFail validation
ReviewerNamed editor and specialist when neededNo self-approval by the drafting step

Run from approved source to approved channel asset

  1. 01Freeze the sourceTrigger
    Create a project and source-version key only after editorial approval.
  2. 02Build the claim ledgerAI judgment
    Extract candidate claims with exact spans; return unsupported or ambiguous items for review.
  3. 03Approve reusable materialHuman approval
    Confirm claims, quotes, rights, exclusions and allowed transformations.
  4. 04Select for one channelCode
    Combine approved claims with the current channel brief and asset key.
  5. 05Draft within boundsAI judgment
    Compress, reorder and reframe without adding facts, outcomes or quotations.
  6. 06Validate mechanicallyCode
    Check source references, required labels, length, duplicates and prohibited terms.
  7. 07Approve the assetHuman approval
    An editor accepts the actual copy and media for the named channel.
  8. 08Publish idempotentlyCode
    Use one asset key, store the publication ID and preserve the withdrawal route.

Distinguish a permitted transformation from a new claim

Repurposing changes presentation; it does not create new evidence.
Draft moveRouteReason
Shorten a supported explanationAllowed with source linkMeaning and qualification remain
Turn a list into a channel-native sequenceAllowed after editor reviewOrder changes, evidence does not
Convert “may” to “will”RejectIt changes the claim strength
Create a customer quote from a summaryRejectA quote requires exact attributable words and rights
Add a platform-specific claimReturn to source approvalNew claim needs current evidence
Publish the same copy everywhereRewrite to each briefChannel fit is part of usefulness

Make correction and withdrawal a first-class path

Google's guidance emphasizes accuracy, quality, relevance and added value for AI-assisted content. Lineage makes those requirements operational: a correction to one source claim can query every derivative by claim ID, pause scheduled assets, notify their owners and record the corrected publication.

  • Source superseded: block new derivatives and compare approved claims between versions.
  • Rights expire: locate every affected asset and unpublish or replace the media.
  • Unsupported draft sentence: reject the asset rather than asking the model to invent a citation.
  • Partial publish failure: preserve successful publication IDs and retry only missing destinations.
  • Editor correction: update the claim ledger or channel rule so the same error becomes a regression case.

A webinar statistic is omitted because its source is not in the packet

Simulated repurposing runSimulated example data

An approved webinar transcript contains a speaker's unsourced market statistic. The editor approves the operating example but marks the statistic unavailable for reuse. A social-post draft reintroduces the number from model memory.

The validator finds no approved claim ID and rejects the asset. The next draft uses the supported example and links to the webinar source. The failed sentence becomes a regression case; it is never published and no citation is fabricated.

Pair the editorial procedure with the lineage blueprint

The SOP defines approval and correction responsibilities. The blueprint adds versions, claim IDs, idempotency, retries, publication records and acceptance tests. Adapt it to the actual channels and rights policy; no connector is implied.

SOP

Content repurposing SOP

Editable procedure for source approval, claim extraction, channel adaptation, review, publishing and correction.

Preview the file
# Content repurposing SOP

Version: 1.0

Owner: Content or marketing operations

Effective date: [YYYY-MM-DD]

Review cadence: [cadence and owner]

## Purpose and scope

Create channel-specific derivative drafts from one approved source while preserving factual meaning, claim provenance, rights, editorial approval and correction control. This procedure does not establish copyright, endorsement or licensing rights and does not permit autonomous publishing.

## Trigger, inputs and prerequisites

- Trigger: source version receives editorial approval for repurposing.
- Inputs: source file/version/hash, claim ledger, citations, quote/rights record, prohibited moves, channel brief, editor and publication destinations.
- Prerequisites: approved voice/claims policy, rights policy, channel requirements, correction/withdrawal route, test cases and publishing permissions.

## Roles

- Source owner: approves the canonical source and supported claims.
- Rights owner: confirms permitted text, image, audio, video and quote reuse.
- Channel editor: approves the actual derivative and publication.
- Operations owner: owns versions, workflow failures and correction propagation.

## Procedure

1. Freeze the approved source version and create a project ID.
2. Extract candidate claims with exact source spans, citations and qualifications.
3. Have the source/rights owners approve reusable claims and media.
4. Create one asset ID from project, source version, channel and brief version.
5. Select only approved claims relevant to that brief.
6. Generate a draft that may compress, reorder or reframe but may not add facts, outcomes or quotations.
7. Validate source references, required labels, prohibited terms, length and duplicate asset keys.
8. Route unsupported sentences, rights gaps and ambiguous qualifications to review.
9. Have the named editor approve the exact copy/media and then approve publication.
10. Store publication ID, URL, source/claim links, reviewer edits and withdrawal path.

## Decisions, exceptions and approvals

- New fact, statistic, product claim or customer result: return to source research and approval.
- Quote: require exact words, attribution and permitted use.
- Source superseded: block new assets and review affected scheduled/published derivatives.
- Rights expired: pause or withdraw every asset using the affected material.
- Partial publication: retry only missing destinations after reconciling existing IDs.
- Publication, correction and withdrawal: named human approval.

## Output and completion

Output one approved channel asset with asset/source/brief versions, claim IDs, citations, rights state, reviewer, publication ID and correction link. A generated draft is not a completed output.

## Measures

Track unsupported-claim rejection rate, editor correction themes, duplicate prevention, publication failures, time in review, claim-withdrawal propagation and assets lacking a source span. Do not use raw asset count as a quality measure.

## Worked example (simulated)

An approved webinar includes an unsourced statistic. The claim ledger marks it unavailable for reuse. A draft introduces the number from model memory, so validation rejects the asset. The corrected draft uses only the supported operating example, and the failure becomes a regression case.

## Revision log

| Version | Date | Change | Approver |
| --- | --- | --- | --- |
| 1.0 | [YYYY-MM-DD] | Adapted from public template | [name/role] |
Workflow blueprint

Content repurposing workflow blueprint

Version, lineage, generation, approval, idempotency and withdrawal design for derivative assets.

Preview the file
# Content repurposing workflow blueprint

## Trigger and termination

- Trigger: `{source_id, source_version, approval_id, rights_version}` becomes approved for repurposing.
- Success: approved derivative has a reconciled publication record or an explicitly approved draft-only terminal state.
- Stop: revoked source/rights approval, unsupported claim, invalid channel brief, expired editor authority or owner-issued stop.

## State model

`received -> claim_extraction -> claim_review -> ready -> drafting -> validating -> editorial_review -> publication_approval -> publishing -> published`

Alternatives: `returned`, `rejected`, `withdrawn`, `stopped`, `superseded`. Approval never occurs in the drafting step.

## Data contract

- Source: ID, version, hash, approval and canonical location.
- Claim: ID, text, exact source span, citation, qualification, allowed transformations and approval.
- Rights: material ID, owner/consent, allowed channels, expiry and restrictions.
- Brief: version, channel, audience, format, length, voice, desired action and prohibited moves.
- Asset: ID, source/claim/brief versions, draft, validation results, reviewer, publication IDs and correction state.

## Ordered steps and actors

1. **Trigger:** receive approved source version.
2. **AI (bounded):** extract claims/spans or mark ambiguous.
3. **Human:** approve reusable claims and rights.
4. **Code:** select approved claims for one brief and compute asset key.
5. **AI (bounded):** draft without new facts, quotes or outcomes.
6. **Code:** validate lineage, labels, prohibited terms and duplicate keys.
7. **Human:** edit/approve exact asset and publication.
8. **Code:** publish idempotently and retain effect IDs.
9. **Human + code:** correct or withdraw through linked assets when needed.

## Idempotency, retry and rollback

- Key: `asset:{source_id}:{source_version}:{channel}:{brief_version}:{variant_key}`.
- Key each publish effect by asset/version/destination. Reconcile returned IDs before retry.
- A changed draft creates a new version, not an overwritten approved artifact.
- Rollback pauses scheduled effects and withdraws/replaces published assets through recorded destination IDs.

## Permissions and sensitive data

Drafting reads approved source/claim fields only. Publishing credentials are unavailable to the drafting step. Exclude private customer/personal data from prompts and test with fictitious fixtures. Restrict withdrawal to named editors/owners.

## Exception routes

- Unsupported/ambiguous claim -> source editor.
- Quote or media rights gap -> rights owner.
- Channel-policy failure -> channel editor.
- Partial publish -> operations owner.
- Source correction/withdrawal -> all linked asset owners.
- No owner -> safe stop.

## Acceptance tests

1. Unapproved source cannot trigger drafting.
2. Claim without source span cannot enter a draft.
3. Qualification such as “may” cannot become “will.”
4. Quote without exact text/rights is rejected.
5. Drafting step cannot publish or self-approve.
6. Duplicate event creates no duplicate asset/publication.
7. Partial publish retries only missing destinations.
8. Source withdrawal finds and pauses every linked scheduled asset.

## Audit record

Retain source/claim/rights/brief versions, prompts or generation configuration where appropriate, validations, drafts, edits, approvals, publication IDs, retries, corrections and withdrawal events.

Limitations and when not to use this

  • This playbook does not claim an all-agents connector to a CMS, social network, video, transcription, design or scheduling platform.
  • A source packet does not establish copyright, publicity, endorsement or licensing rights; use the organization's qualified review where required.
  • Automatic checks cannot replace editorial judgment about accuracy, context, usefulness, brand, audience or harm.
  • The example is simulated and represents no reach, ranking, engagement, production-time or revenue result.

Sources

  1. Guidance on using generative AI contentGoogle Search Central Accessed 10 August 2026
  2. Spam policies for Google web searchGoogle Search Central Accessed 10 August 2026
  3. Protecting Personal Information: A Guide for BusinessFederal Trade Commission Accessed 10 August 2026

Download the repurposing SOP

Define one approved source, claim ledger, channel brief and editor before generating the first derivative.

Download the repurposing SOP
About the author

Uli Prantz

Builds and operates all-agents

Uli Prantz builds all-agents, the process-automation platform this site documents. He writes about the operational side of automating recurring business work: where deterministic code beats model judgment, where it does not, and where a human still has to approve.

Bring one process. We will scope it in 30 minutes.

You leave the call knowing whether it is a fit, what can become code and what still needs a person.

Book a discovery call