What the data for this category looks like
Core data for campaign setup comes from advertising platforms, user profile libraries, budget scheduling tables, and attribution tools. Daily produced documents include four types: campaign plans, targeting settings, budget schedules, and conversion tracking configurations. These are organized per individual campaign activity. They include fields such as targeted audience package identifiers, budget amounts, conversion event names, and delivery time periods. Updates are synchronized daily or whenever a campaign activity is updated.
What Constraints Do These Characteristics Impose on Document Parsing and Chunking?
The structure of campaign setup documents organized per individual activity means chunks cannot be split across activities. Splitting across activities will lose the complete configuration logic for a single campaign. Mixed fields from multiple data sources require the parsing process to retain associations between different field types. This prevents targeting and budget configurations from becoming disconnected after splitting. The high-frequency update characteristic requires the parsing process to include reasonable timeout fault tolerance. This avoids parsing interruptions caused by frequent updates. Structured tables and nested configurations require the parsing process to support non-standard formats such as merged cells and nested fields. Without this support, field identification errors or missing data will occur.
How to set configurations
The following configurations are based on the parameter logic of FastGPT v4.17.0.
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
chunkSize | 800–1200 characters | Adapts to the single-campaign configuration length of campaign setup documents, preventing loss of association between targeting and budget configurations after splitting |
chunkSplitter | By paragraph + keyword anchoring | Campaign setup documents contain structured paragraphs and configuration items. Splitting by paragraph preserves configuration context, while keyword anchoring targets core fields such as targeting and budget |
PARSE_FILE_TIMEOUT_SECONDS | 300–600 seconds | Campaign setup documents often include multiple campaign configurations, leading to longer parsing times. This setting avoids parsing interruptions due to timeout |
UPLOAD_FILE_MAX_SIZE | 1000 MB | Supports bulk import of multi-campaign campaign setup documents, adapting to bulk operational needs for agency operations |
customPdfParse | Enabled | Most campaign setup documents are exported in PDF format. Enabling custom parsing preserves table and field structures, preventing configuration items from being lost during OCR recognition |
chunkSettingMode | Manual configuration | Adapts to differences in document structures across different campaign activities. Manually adjust chunking parameters to match the field association logic of different documents |
Three common misconfigurations
- Issue: An error reporting empty fields appears after parsing a campaign setup Excel document that contains merged cells. Cause: Structured file parsing adaptation is not enabled. The default parsing cannot recognize nested configuration fields, leading to lost associated data.
- Issue: The number of chunks after document parsing is excessively high, leading to prolonged vectorization processing time. Cause:
chunkSizeis set too small. This splits a single targeting configuration into multiple independent chunks, increasing the total workload of vectorization calculations. - Issue: Non-document text queries trigger the document parsing process. Cause:
chunkSettingModeis not configured correctly. General queries are incorrectly bound to the document parsing link, leading to abnormal resource usage.
How to confirm configurations are properly set
- Upload a standard campaign setup Excel document, view the parsed chunk preview, and confirm that the targeting, budget, and conversion configurations for each campaign activity are not split across different chunks.
- Upload a single campaign setup PDF document under 100 MB, wait for parsing to complete, and confirm that no timeout error is triggered.
- Test non-document text queries, confirm that the document parsing process is not activated, and that parsing only occurs when uploading documents.
- Check the uploaded file size limit, confirm that bulk documents matching business requirements can be uploaded without triggering a block prompt.
The values provided here 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.