What the data for this category looks like
Advertising and marketing research draws data from real-time performance data in ad delivery backends, publicly available media rate cards and scheduling documents, competitor ad material libraries, industry research reports, and similar sources. Data update frequencies span real-time, hourly, monthly, and quarterly intervals. Most structured data exists in table format, with fields including delivery date, channel, material identifier, performance metrics, and additional relevant fields. Unstructured data includes creative copy, video scripts, poster source files, and similar content. Some quantitative metrics include standard business units. Overall data types are diverse, and associated relationships are complex. A single round of research dialogue may reference content from multiple sources and formats simultaneously.
What constraints do these characteristics impose on conversation logging and auditing
Multi-source, multi-format data requires conversation logs to fully record data identifiers and reference paths linked to each dialogue round. This prevents ambiguous data traceability during audits. Data with varying update frequencies requires logs to synchronously record data acquisition timestamps. This ensures audits can match the data source state at the time the dialogue was initiated. References to unstructured materials require logs to include material version information. This prevents audit results from conflicting with actual delivered content due to material iterations. Advertising and marketing research has frequent compliance verification requirements. Logs must fully record compliance verification logic and results for each dialogue round, to meet subsequent compliance audit needs. Large data volume requires log storage to support path partitioning by business scenario. This avoids wasted storage resources and reduced query efficiency.
How to configure the settings
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
apiLogEnabled | true | Compatible with v4.8.3 and later versions. Enables full logging of conversation API calls to meet auditing requirements for advertising and marketing research |
maxContextHistory | 15-25 previous entries | Advertising and marketing research dialogues often involve multiple rounds of delivery strategy iteration. This range balances log storage volume and context completeness |
logRetentionDays | 90 days | Advertising marketing compliance audits typically require retaining at least 90 days of dialogue data. This value aligns with general compliance requirements |
auditLogIncludeSourceData | true | Advertising and marketing research dialogues often link structured delivery data and unstructured materials. Enabling this setting fully records the data traceability chain |
LOG_EXPORT_PATH | /var/log/fastgpt/marketing_research/ | Partitions storage paths by business scenario to facilitate subsequent archiving and audit queries |
apiResponseLogMatch | true | Compatible with v4.8.10 and later versions. Enables matching verification between API call logs and returned content to prevent discrepancies between logs and actual responses |
The parameter values provided on this page are common starting points for configuration. Actual values are affected by material formats, data volume, and business rules. Specific issues require individual analysis. It is recommended to test on your own samples before finalizing settings.
Three common configuration errors
- Symptom: After calling the conversation API for v4.8.10 or later, log details show content that does not match the actual returned response, or remains static. Cause: The
apiResponseLogMatchconfiguration item is not enabled. This causes logs to only record initial request parameters, without capturing the final generated dialogue content. - Symptom: After configuring
LOG_EXPORT_PATH, the system does not generate corresponding log files. Cause: The FastGPT process lacks read/write permissions for the target storage path, or the path format includes invalid characters. - Symptom: After multiple rounds of research dialogues, the audit log does not include structured delivery data content referenced in the previous round. Cause: The
auditLogIncludeSourceDataconfiguration item is not enabled, so associated business data is not written to the audit log.
How to verify correct configuration
- Initiate a dialogue referencing structured delivery data and describing unstructured materials. Check whether backend logs record complete request parameters, intermediate generation steps, and final response content.
- Confirm that time-partitioned log files are generated in the target storage path, and that file permissions match the configured
LOG_EXPORT_PATH. - Call the conversation API for v4.8.3 or later versions. Verify that log details include association information between the current dialogue and historical context.
- Trigger a preset advertising compliance verification logic. Confirm that the audit log records verification results and trigger conditions.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-14.