# Changelog of Campaign Offer Guard: Promotion & Ad Review (`critd/campaign-offer-guard`) Actor

- **URL**: https://apify.com/critd/campaign-offer-guard/changelog.md
- **Full Actor documentation**: https://apify.com/critd/campaign-offer-guard.md

## Campaign Offer Guard release notes

Customer-facing changes and compatibility notes. The README owns the current preparation, pricing and recovery contract. Preserve original files when changing a saved workflow.

### Offer collision review — optional supplied declarations

Compare fixed-price offers explicitly declared to share a placement and audience. The report preserves exact source rows, price scales, effective/display-window endpoints, ambiguous revisions and unanswered questions. Distinct affected offers and campaigns are counted without multiplying supplied spend. Prelaunch reviews can use explicit empty campaign/observation tables.

The existing review, fictional public task, price and billing/recovery behavior remain available. Collision declarations are optional; complete findings share existing input, source-reference and result limits. Over-limit reviews refuse before publication and the completed-report charge. This does not establish approval precedence, verify checkout or authorize launch.

### October 8, 2026 — maintenance update in the 0.2 family

The refreshed build retains fixed-price review while consolidating the underlying file-reading implementation. A published fictional example now gives a direct starting point. Normal pricing applies to that example: $0.25 per completed report plus $0.00005 per start.

Results, input limits, explicit mapping and existing prices are retained. Component-level improvements are not a promise of faster hosted runs or measured customer time savings.

### 0.2.0 — clearer preparation corrections

**Know what to fix before trying again.** Failed-run guidance can identify up to five supported corrections, such as required mappings or invalid formats, with remaining checked-stage corrections identified when available. Private source headers and values stay out of diagnostics. Unknown failures retain general guidance.

Invalid preparation does not request a completed-report charge; startup charges can still apply. A valid review with an unresolved approval clock remains a completed review. Ordinary findings, prices and recovery behavior are unchanged.

### 0.1.0 — prepared-file review on Apify

Compare prepared approved-offer, campaign and observed-offer records, with explicit CSV/JSON column mapping. Open the readable report and working exception list, and follow the source-row references when a detail needs checking.

The fictional example uses normal pricing. Completed results are published before the report charge is requested; uncertain charges are not automatically replayed. A valid unresolved review is billable. No live advertising-account connection, checkout test, blanket vendor-export compatibility or savings calculation is included.