Data Health
Monitor your product data quality, find missing attributes, and fix issues that affect search and filter accuracy.
Data Health
BetaThe Data Health page gives you a clear picture of your product data quality. Incomplete or inconsistent data leads to poor search results and filters with gaps. The Data Health tools help you find and fix these issues.
Access Data Health from the app dashboard at AI > Data Health (or navigate directly to /app/data-health).
Data Quality Score
At the top of the page, you see your overall Data Quality Score -- a percentage from 0-100 that summarizes how complete and consistent your product data is.
| Score | Rating | Meaning |
|---|---|---|
| 90-100 | Excellent | Your data is well-structured and complete |
| 70-89 | Good | Minor gaps that could affect some searches |
| 50-69 | Fair | Noticeable data issues affecting search quality |
| Below 50 | Poor | Significant gaps -- many products have missing attributes |
The score is calculated based on the completeness of key fields across your catalog: titles, descriptions, images, tags, variant options, and configured metafields.
Metafield Doctor
The Metafield Doctor is the primary tool on the Data Health page. It scans your entire product catalog for missing or inconsistent attributes.
Running a Scan
- Click Run Metafield Doctor
- The scan processes your products and checks each one for data issues
- Progress is shown in real time: products scanned, issues found, estimated time remaining
What It Checks
| Check | What It Looks For |
|---|---|
| Missing descriptions | Products with empty or very short descriptions |
| Missing images | Products with no images or only placeholder images |
| Empty tags | Products with no tags assigned |
| Inconsistent option names | Variant option names that should be standardized (e.g., "Colour" vs. "Color") |
| Missing metafield values | Products that lack values for metafields you have configured as filters |
| Price anomalies | Products with $0 prices or compare-at prices lower than the actual price |
| Duplicate titles | Products with identical or nearly identical titles |
Scan Results
After the scan completes, issues are organized by category. Each category shows:
- Issue count -- How many products are affected
- Severity -- High (directly impacts search), Medium (affects filter accuracy), or Low (cosmetic)
- Affected products -- A list of products with the issue, each linking to the product in your Shopify admin
Applying Suggestions
For certain issue types, the Metafield Doctor can suggest fixes automatically:
- Auto-fill missing metafields -- If similar products have a value for a metafield, the doctor suggests applying the same value to products that are missing it. For example, if 90% of your "T-Shirts" have
material: Cotton, it suggests setting the remaining 10% toCottonas well. - Standardize option names -- If some products use "Colour" and others use "Color", the doctor suggests normalizing to a single spelling.
- Tag suggestions -- Based on product titles and descriptions, the doctor suggests tags for untagged products.
To apply suggestions:
- Review the suggested changes in the Suggestions tab
- Check or uncheck individual suggestions
- Click Apply Selected
Suggestions write to product metafields only. They never modify your product titles, descriptions, prices, or images.
Manual Fix Mode
If you prefer to fix issues yourself, switch to Manual Fix Mode in the results view. This gives you:
- A spreadsheet-like interface showing all products with a given issue
- Inline editing for metafield values
- Bulk select and bulk edit capabilities
- The ability to open any product in your Shopify admin with one click
Exporting Issues as CSV
To work with data issues outside of FilterIQ:
- Run a scan or view previous scan results
- Click Export CSV in the top-right corner of the results page
- Choose which issue categories to include
- The CSV downloads with columns for: Product ID, Product Title, Handle, Issue Type, Issue Details, Current Value, Suggested Value
You can use the exported CSV to make bulk corrections in Shopify using a CSV import, a spreadsheet app, or a third-party bulk editor.
Why Data Health Matters
Your search and filter quality is directly tied to your data quality:
- Missing descriptions mean fewer keyword matches when shoppers search
- Missing tags mean products are excluded from tag-based filters
- Inconsistent option names split what should be a single filter into two (e.g., "Color" and "Colour" appearing as separate filters)
- Empty metafields create gaps in custom filters where some products disappear
Running the Metafield Doctor regularly -- especially after bulk product imports or catalog restructuring -- keeps your search experience accurate and complete.
The Data Health tools are available on the Pro and Plus plans. The Free and Starter plans can view the Data Quality Score but cannot run the Metafield Doctor or apply suggestions.
AI providers and data flow
Data Health scans (Data Quality Score + Metafield Doctor) send catalog metadata to one of our approved AI sub-processors — currently Anthropic (Claude API) or OpenAI (GPT API). Both are disclosed in our Sub-processor List on the Privacy Policy page, which is the authoritative live list generated from docs/governance/SUBPROCESSORS.yaml. Catalog data is minimized and redacted before submission; neither provider is authorized to use submitted merchant data to train public models.