Skip to main content

Overview

Pipeline Settings let you customize how the Bronze → Silver → Gold pipeline behaves for your workspace. You can define custom field mappings (Silver), adjust scoring weights (Gold), enable auto-triggers, and use preview mode to test changes safely.

Accessing pipeline settings

Via UI:
  1. Navigate to SettingsPipeline
  2. The settings page shows tabs for Silver Mappings, Gold Weights, and Behavior
Via API:
  • GET /api/workspace/{workspaceId}/settings/pipeline — read current settings
  • PUT /api/workspace/{workspaceId}/settings/pipeline — update settings
Settings are applied at the workspace level and affect all catalogs in the workspace. Team-level overrides are available on Enterprise plans.

Silver field mappings

Silver mappings define how non-standard source fields are mapped to the Alana product schema.

Why you need mappings

When supplier files use column names like "item_code" instead of "sku", or "prod_name" instead of "title", Silver won’t know how to map them without explicit configuration.

Mapping rules

Each mapping rule has:
PropertyTypeDescription
sourceFieldstringThe column name as it appears in the source file
targetFieldstringThe Alana schema field to map to
transformstring (optional)Transformation to apply: "uppercase", "lowercase", "trim", "prefix:X", "suffix:X"

Example mappings

{
  "silverMappings": [
    {
      "sourceField": "product_name",
      "targetField": "title",
      "transform": "trim"
    },
    {
      "sourceField": "item_code",
      "targetField": "sku"
    },
    {
      "sourceField": "cat",
      "targetField": "categoryPath",
      "transform": "prefix:Apparel > "
    },
    {
      "sourceField": "vendor",
      "targetField": "brand"
    }
  ]
}

Available target fields

Target fieldTypeDescription
titlestringProduct name
skustringStock-keeping unit
gtinstringGlobal Trade Item Number
pricenumberSelling price
originalPricenumberPre-discount price
currencystringISO 4217 code
brandstringBrand name
categoryPathstringCategory hierarchy (> separated)
descriptionstringFull product description
shortDescriptionstringBrief description
primaryImageUrlstringMain image URL
availabilitystringStock status

Gold scoring weights

Gold weights define the importance of each of the 7 scoring stages. All weights must sum to 100.

Default weights

StageDefault weight
Identity (SKU, GTIN, brand)20%
Taxonomy (category depth)15%
Content (title, description, bullets)25%
Media (images, video)20%
Pricing (price, currency, original)10%
Attributes (category-specific specs)5%
SEO (slug, meta, keywords)5%

Custom weight example

For a media-heavy catalog (e.g., fashion photography):
{
  "goldWeights": {
    "identity": 15,
    "taxonomy": 10,
    "content": 20,
    "media": 35,
    "pricing": 10,
    "attributes": 5,
    "seo": 5
  }
}
All weight values must sum to exactly 100. The API returns a 422 error if the sum is incorrect.

Behavior settings

SettingTypeDefaultDescription
autoTriggerSilverbooleanfalseRun Silver automatically after every Bronze ingest
autoTriggerGoldbooleanfalseRun Gold automatically after Silver completes
previewModebooleanfalseSimulate pipeline changes without writing to products
autoTriggerGold: true is not recommended for large catalogs. Gold scoring has a cost and can be slow for 10,000+ products. Use manual or scheduled triggering instead.

Preview mode

When previewMode: true, pipeline operations simulate transformations and return what would happen — without modifying any product records. Use preview mode to:
  • Test a new Silver mapping before applying it to real data
  • See how new Gold weights would affect scores
  • Validate field mapping logic without risk

Example preview response (Silver)

{
  "preview": true,
  "results": [
    {
      "productId": "prod_abc",
      "status": "would_succeed",
      "fieldsWouldNormalize": ["title", "categoryPath", "brand"],
      "currentValues": {
        "title": "blue running shoe",
        "categoryPath": "shoes"
      },
      "newValues": {
        "title": "Blue Running Shoe",
        "categoryPath": "Apparel > Footwear > Running"
      }
    }
  ]
}

Read current settings

curl "https://app.alana.shopping/api/workspace/WORKSPACE_ID/settings/pipeline" \
  -H "Authorization: Bearer YOUR_API_KEY"

Update settings

curl -X PUT "https://app.alana.shopping/api/workspace/WORKSPACE_ID/settings/pipeline" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "silverMappings": [
      { "sourceField": "product_name", "targetField": "title" },
      { "sourceField": "item_code", "targetField": "sku" }
    ],
    "goldWeights": {
      "identity": 20,
      "taxonomy": 15,
      "content": 25,
      "media": 20,
      "pricing": 10,
      "attributes": 5,
      "seo": 5
    },
    "autoTriggerSilver": true,
    "autoTriggerGold": false,
    "previewMode": false
  }'

Score coherence

Gold scores are recalculated only when Gold is triggered. If you change goldWeights, existing scores on products are stale until Gold is re-run. A banner in the UI warns when settings have changed since the last scoring run. To refresh scores after changing weights:
# Re-run Gold on all products after updating weights
curl -X POST "https://app.alana.shopping/api/workspace/WORKSPACE_ID/catalogs/CATALOG_ID/batch/gold" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"scope": "all"}'

Best practices

If your supplier files use non-standard column names, configure Silver mappings before importing. This ensures your first import lands correctly without requiring a re-import.
Before applying new mappings to a live catalog, enable preview mode and run Silver on a small selection. Review the newValues output to confirm mappings behave as expected.
After updating Gold weights, trigger a full catalog re-analysis so all product scores reflect the new configuration.
If you have a recurring import job (daily supplier feed, nightly CSV), enable autoTriggerSilver: true so imported products are normalized automatically.
Last modified on March 18, 2026