Fictional worked example

Catalog variant preparation

Prepare 4 of 8 fictional variant rows. Hold 3 unresolved rows and exclude 1 exact duplicate.

Static demonstration · No client outcomes or live-system changes

Fictional demonstration. All data is invented.

Prepare 4 of 8 fictional variant rows. Hold 3 unresolved rows and exclude 1 exact duplicate.

Eight invented storage and packaging variant rows with inconsistent SKU/option formatting, one missing size and one conflicting SKU pair.

Before and after

RecordBeforeAfter
V001 bin-10-clr ; capacity 10 l; color clearBIN-10-CLR; Capacity 10 L; Color Clear
V006Blank size; SKU contains A4Held; size remains blank
V007 + V008Same normalized SKU; different capacity/priceBoth held for a decision

1. What changed

  • Trim outer SKU whitespace and uppercase the supplied SKU. Preserve separators and digits.
  • Apply the explicit demo option labels Capacity, Size and Color.
  • Use only the supplied demo value map: 10/20/30/45 l to the same value with L; A5 stays A5; clear/blue/natural/gray become Clear/Blue/Natural/Gray.
  • Preserve product IDs and prices. No dimensions, inventory, images or missing options are invented.
  • Exclude only exact duplicates across all raw business fields, ignoring source row ID and dataset label.
  • Hold a missing required option value. Never infer a size from a SKU.
  • Hold both rows when a normalized SKU points to conflicting variant details.

2. Calculations and checks

  • 8 input rows = 4 ready + 3 held + 1 exact duplicate.
  • V005 exactly duplicates V004.
  • V006 is held for a missing size. V007 and V008 are held for conflicting details under CRATE-30-GRY.
  • Every source row is represented exactly once as ready, held or duplicate.
  • Prepared SKUs are unique and every prepared row has both option values.
  • Every prepared price and product ID matches its source.
  • Blank size stays blank in exceptions; conflicting SKU rows are retained for review.

3. Decisions still needed

  • V006: supply the size.
  • V007 and V008: confirm the correct SKU or the correct capacity and price for each variant.
  • Agree on the destination platform template before testing an import.

Limits of this example

  • This is a generic review template. No production import or platform integration has been tested or validated.
  • Prepared rows follow only the clearly defined demo option map. Real catalogs may require additional fields or rules.
  • dataset_label and source_row_id are audit columns. Map or omit them only after the target template is agreed.

Bring the problem. We’ll define the project.

A few lines about what you have and what you need are enough to begin.

Tell me what you need