Every feed in FeedPrep has a health report that gives you a quick overview of data quality and configuration completeness. The health report helps you identify missing mappings, gaps in your data, and areas where your normalization setup can be improved.
Completeness Score
The health report includes a completeness score from 0 to 100 that summarizes the overall health of your feed. The score is calculated from two weighted components:
- Column mapping completeness (40%) — what percentage of your canonical attributes have been mapped to a source column in the feed.
- Critical attribute coverage (60%) — what percentage of rows have actual data present for attributes you have marked as critical.
A score of 100 means every critical attribute is mapped and every row has data for those attributes. In practice, most well-configured feeds score between 70 and 95, since supplier feeds rarely have 100% coverage for every attribute.
Critical Issues
Critical issues are problems that are likely to cause data quality failures in your downstream systems (PIM, webshop, marketplace). The health report flags two types of critical issues:
- Missing column mapping for critical attributes — a critical attribute in your schema has no source column mapped to it. This means the attribute will be empty for every row in the feed.
- Missing data for critical attributes — a critical attribute is mapped to a source column, but that column is empty for some rows. The report shows the exact count and percentage of rows affected.
Critical issues should be resolved before exporting. They typically indicate a gap in the supplier's data or a missing column mapping that needs attention.
Warnings
Warnings highlight issues that are not blocking but may affect data quality:
- Missing data for recommended attributes — attributes that are not critical but are marked as recommended in your schema have empty values for some rows. The report shows affected counts and percentages.
Recommended attributes are typically things like secondary images, detailed descriptions, or supplementary specifications. Missing data here won't break your feed, but filling these gaps can improve the quality of your product listings.
Info
The info section provides additional context about your feed configuration:
- Default value rules configured — how many default value rules are active for this feed's adapter. Default values fill in missing data automatically.
- Unmapped columns count — how many columns exist in the source file but have not been mapped to any attribute. These may contain useful data you have not yet incorporated.
- Active transform rules count — how many transform rules (math operations, text manipulations, conditional logic) are configured for this feed's adapter.
Statistics
The health report also includes basic statistics about the feed:
| Statistic | Description |
|---|---|
| Total rows | The number of product rows in the feed. |
| Columns detected | The total number of columns found in the source file. |
| Columns mapped | The number of source columns that have been mapped to canonical attributes. |
Refreshing the Health Report
The health report is generated automatically when a feed is processed. If you make changes to your adapter configuration — such as adding new column mappings, updating value mappings, or creating transform rules — you can refresh the report on demand to see how those changes affect the feed's health score and issues.
This makes the health report a useful feedback loop while configuring your normalization: make a change, refresh the report, and see whether the issue count goes down and the score goes up.