What this category of data looks like
Core data for ad creative review comes from ad placement platforms, landing page resources, product qualification documents, and advertising law compliance clauses. Daily produced documents include ad creatives awaiting review, landing page screenshots, manual review opinions, and rectification records. Data update frequency is daily or real-time, with updates triggered immediately as placement plans adjust or violation decisions are made. Data fields include creative unique identifiers, placement channel types, copy content, landing page URLs, violation type tags, rectification requirements, and more. There are no unified fixed units; fields are dynamically adjusted based on placement scenarios and compliance requirements.
What constraints do these characteristics impose on the "workflow orchestration" link
The data sources for ad creative review are scattered and update flexibly. Workflows must support parallel pull nodes from multiple data sources to ensure real-time or daily updates of creatives and compliance clauses. The business logic of review by individual item requires workflows to be configured with loop processing nodes to adapt to the full-process verification of a single creative. The variety of document types and dynamically adjusted fields requires support for dynamic variable binding and field mapping to avoid adaptation failures caused by hardcoding. The consequences of errors are directly linked to ad placement permissions. Workflows must be configured with pre-checks and exception interception links to reduce the probability of manual intervention errors.
How to set the configurations
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
| Loop termination condition | Triggered when the violation tag field in variables is detected | Matches the violation judgment logic of ad creative review, automatically terminating invalid loops |
| Number of creatives processed per batch | First 20 | Adapts to server resource load for single-batch reviews, avoiding node timeouts |
| Timeout threshold | 600 seconds | Covers the average time required for multi-data source pulling, copy checking, and landing page verification |
| Variable reference format | {{}} format compatibility mode | Adapts to FastGPT v4.17.0 variable reference rules, fixing compatibility issues with older versions |
nodeId binding | Bind to the review result storage node | Ensures all review results after loop processing are accurately collected to the specified storage location |
| Maximum character limit per single check | 800–1200 characters | Matches the typical length of single ad creative copy, avoiding node processing overload |
Three common configuration errors
- The loop does not terminate automatically after execution, and logs show the violation tag field in
variablesis not correctly identified. The cause is that the loop condition is not bound to the correctvariablesfield, and does not match the violation judgment logic of ad creatives. - When calling an ad platform interface via an HTTP node, variable resolution fails and returns empty. The cause is that the {{}} format variable compatibility setting is not enabled, and using the old / pattern for variable retrieval leads to resolution failure.
- No review results are produced after the workflow is triggered, and no data is written to the storage node bound by
nodeId. The cause is that the loop is not configured with correct output variable mapping, leading to processed results not being passed to the storage node.
How to confirm the configuration is correct
- Submit a single test creative with a clear violation tag, verify all workflow node statuses are successful, and confirm the loop terminates automatically.
- Trigger a batch review task, check the storage node bound by
nodeId, and confirm all review results are correctly written. - Simulate a high-load scenario, submit more than 20 test creatives, and confirm no node timeout errors are triggered.
- Adjust the variable reference format to the old / pattern, verify variable resolution failure occurs, and confirm the compatibility setting is effective.
The values provided are common starting points and should be measured against the reader's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-10-19.