What the data for this category looks like
Core data for this category comes from landing page access logs, heatmaps, form submission data, and ad source parameters. Regularly produced documents include landing page plans, event tracking requirement documents, A/B test reports, and form field lists. The data update frequency is weekly or within a single campaign cycle. Most document structures include structured fields such as form field names and ad channel parameters, as well as unstructured analysis content such as page path analysis conclusions. Fields have no unified fixed units, and must be matched based on specific campaign scenarios.
Constraints for Knowledge Base Retrieval and Recall
The data sources for landing page conversion optimization are mostly real-time or near-real-time data at the campaign level. Retrieval and recall functions must support filtering by campaign time range to avoid interference from irrelevant data across campaigns. Documents contain a large number of structured fields and unstructured analysis content, so both semantic matching and precise keyword matching must be supported to ensure that detailed content such as buried point fields and form rules is recalled. The weekly or campaign-level update frequency means the knowledge base must support incremental updates. Retrieval results must fit the exclusive configuration of the current campaign, and cannot be generalized to historical campaigns. In addition, troubleshooting scenarios have high timeliness requirements, and overly long retrieval waiting times must be avoided.
How to Set Configurations
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
similarity | 0.75–0.85 | Semantic similarity range for matching landing page-related documents, to avoid recalling irrelevant historical campaign documents |
limit | Top 10–15 results | The total number of troubleshooting documents for landing page conversion optimization is limited. Excessive results will exceed the context length and affect subsequent processing |
searchMode | Vector search + keyword hybrid | Balances semantic matching and precise keyword matching, and can recall structured content such as buried point fields and ad parameters at the same time |
usingReRank | Enabled | Landing page documents contain a large number of detailed fields, and re-ranking can improve the ranking priority of highly relevant results |
datasetSearchUsingExtensionQuery | Enabled | Can supplement context such as current campaign ad source parameters and campaign time to improve retrieval accuracy |
collectionIds | Bind the exclusive dataset ID of the current campaign | Avoid recalling irrelevant documents from other campaigns, and ensure that retrieval results fit the current business scenario |
Three Common Configuration Mistakes
- Phenomenon: An error
fail to create post presigned urloccurs when uploading .txt or .csv files related to landing pages. Cause: The storage bucket permissions associated with the knowledge base are not configured correctly, or the current campaign's dataset ID is not bound. - Phenomenon: Non-current campaign landing page documents are recalled during retrieval. Cause: The
collectionIdsparameter is not specified, or an incorrect dataset ID is bound, causing the retrieval scope to cover irrelevant knowledge bases. - Phenomenon: When importing a .csv format ad source parameter file, the prompt
datasetId is required for S3 filesis displayed. Cause: The associated dataset ID is not specified in the configuration, so files stored in S3 cannot be matched to the corresponding knowledge base.
How to Confirm the Configuration Is Complete
- Upload a test landing page plan document, check that there are no preset URL-related errors during the upload process, and confirm that the file is successfully imported into the knowledge base.
- Enter a specific troubleshooting question, such as "Form field missing troubleshooting", and check whether the recall results include the current campaign's form field list document.
- Go to the knowledge base configuration panel of FastGPT v4.17.0, confirm that parameters such as
usingReRankanddatasetSearchUsingExtensionQueryhave been enabled as set. - Import test .csv format ad source parameter data, check that it can be imported normally without datasetId-related errors.
Note: The configuration values provided here are common starting points. They should be evaluated against your own specific samples and scenarios.
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.