Name Content Batches So Imports and Reviews Stay Traceable
Create a content batch naming convention with stable article identifiers, explicit revisions and separate publication dates.
TL;DR
- Give each content batch a durable identity by separating batch, article and revision so an article can be traced from brief to import without relying on a mutable title.
- Use a manifest and readable convention as the practical method: map stable IDs to assets, review state and owner, and keep the manifest distinct from the article body.
- Validate the system with explicit revision records and preflight tests so imports update the intended draft and the batch is closed with its resulting CMS identifiers preserved.
Give the batch a durable identity
A content batch naming convention should let someone trace an article from its brief to its imported record without relying on the current title. Titles change during editing. A stable identifier should not.
Separate the batch identifier, article identifier and revision. For a fictional launch, a batch could be called launch-autumn-01, with articles numbered 001 through 020. The article identifier remains the same even if its headline, keyword or publication date changes.
Choose a convention the people doing the work can read. A long encoded filename that nobody can interpret without a script may be unique, but it is a poor handoff tool unless a clear manifest accompanies it.
Keep mutable fields out of identity
A publication date is a plan, not an article's identity. If rescheduling requires renaming every file, the convention creates unnecessary opportunities for broken references.
Likewise, a keyword belongs in metadata. It may change when research reveals overlap with an existing page. The stable ID should still connect the revised assignment to its earlier review and decisions.
| Field | Intended role |
|---|---|
| Batch ID | Groups a defined delivery of work |
| Article ID | Identifies one article across revisions |
| Revision | Distinguishes a changed deliverable |
| Title and keyword | Describe the current assignment |
| Publication date | Records the current scheduling decision |
A filename such as launch-autumn-01-007-v03.json can be sufficient. The exact punctuation matters less than consistent parsing and a documented meaning.
Maintain a manifest for the batch
Create a manifest that maps each stable ID to its title, source file, image, review state and destination record. It should also identify the person responsible for resolving discrepancies.
Do not make the manifest a second unofficial source of article content. Its purpose is traceability. Keep the body in the designated article file or CMS version and link to it.
Our content brief template can hold the assignment details. The manifest connects that brief to the assets and publication record created later.
Treat revisions as explicit events
When an approved article changes, record what changed and why. A new revision should not silently replace the evidence that an earlier version was reviewed.
For imports, use the stable ID to determine whether a record is new or an update. Do not assume a different headline means a different article. Conversely, two articles with similar titles may have different reader tasks and deserve separate IDs.
A practical revision record includes the previous revision, the changed fields and the review required. A keyword-only correction may need a different check from a rewritten comparison or replaced product screenshot.
Test collisions before delivery
Run a small preflight before importing the batch. Check for duplicate article IDs, missing files, repeated destination slugs and assets attached to the wrong article.
Include a deliberate test case in a nonproduction environment: rename a title while preserving its ID and verify that the importer updates the intended draft. Then attempt to reuse an ID for a different article and verify that the process stops for a decision.
The content approval workflow should identify which revision was accepted. An approval that refers only to a mutable title leaves room for ambiguity.
Close the batch without erasing its history
After import, record the resulting CMS identifiers and any rejected or deferred articles. A batch can be complete even when a deliberately excluded assignment has no published page, provided the disposition is clear.
Keep the naming guide close to the manifest and include a short example. New contributors should be able to add a file without inventing a competing pattern.
A useful convention makes ordinary work easier: locating an image, retrying a failed import, checking an approved version and explaining why a planned article was not released. If it cannot support those tasks, simplify the convention before scaling the batch.