✓ TestedWorks with CSV, Excel, Google Sheets → JSON-LD Schema
By dataclean.to team · 2026-02-13
Product schema markup is the most widely implemented structured data type in e-commerce, powering the price tags, star ratings, and availability badges that appear in Google search results. If your product catalog lives in CSV spreadsheets, converting it to valid Product schema requires cleaning every field to match schema.org's strict property formats. dataclean.to automates this conversion — from a messy product CSV to validated JSON-LD markup with properly nested Offer, AggregateRating, and Brand objects.
The Problem
E-commerce CSVs contain product data that looks correct to humans but fails schema validation in systematic ways. Prices include currency symbols, thousand separators, or "from $X" ranges that aren't valid for the Offer.price property. Product identifiers (GTINs, MPNs, ISBNs) have transcription errors that fail check digit validation. Availability uses natural language ("ships in 2-3 days") instead of schema.org's ItemAvailability enumeration values. Ratings combine value and scale ("4.5/5") instead of separating them. SKU counts and review numbers are formatted with commas that break numeric parsing. Each product in the catalog may have a different combination of these issues. Product schema type specification
How to Fix It
1
Upload your product catalog CSV
Import your CSV containing product data. The platform handles catalogs from any e-commerce system — Shopify exports, WooCommerce dumps, Amazon seller files, or custom inventory spreadsheets. No pre-formatting of column names is needed.
2
Detect product data field types
dataclean.to identifies product-specific data: prices with currency detection, GTIN/UPC/EAN identifiers with format recognition, rating values, image URLs, and categorical fields like brand, color, size, and material.
3
Clean and validate product data
Apply product-specific cleaning: prices extract to numeric values with ISO 4217 currency codes, GTINs validate against check digit algorithms, availability maps to schema.org ItemAvailability values (InStock, OutOfStock, PreOrder, etc.), and ratings split into ratingValue and ratingCount components.
4
Map columns to Product schema properties
Map your columns to Product properties: name, description, image, sku, gtin, mpn, brand (as Brand or Organization), offers (as Offer with price, priceCurrency, availability), and aggregateRating (as AggregateRating with ratingValue, reviewCount, bestRating).
5
Export Product JSON-LD
Generate JSON-LD for each product with properly nested Offer, Brand, and AggregateRating objects. Test with Google's Rich Results Test to confirm product rich result eligibility. Deploy via the auto-embed script or download for manual page integration.
Frequently Asked Questions
Which Product schema properties trigger rich results in Google?
Google's product rich results primarily use: name, image, offers (price + availability), aggregateRating (stars), and review data. The more of these you provide with valid values, the more enhanced your search listing becomes. Brand, GTIN, and SKU also contribute to product identification in Google's systems.
How does dataclean.to handle products with multiple price variants?
If your CSV has separate rows per variant (size, color, etc.), each row generates its own Product JSON-LD with the specific variant's price. If a single row contains a price range, the platform can structure it as an AggregateOffer with lowPrice and highPrice properties.
Can I include product reviews in the schema markup?
If your CSV includes aggregate review data (average rating, review count), these map to the aggregateRating property. Individual reviews can be included as Review objects if your data has individual review text and ratings. Both contribute to star rating display in search results.
Example: Input → Output
name
sku
price
brand
rating
Wireless Headphones Pro
WHP-001
$89.99
SoundTech
4.5
wireless headphones pro
WHP001
89.99
soundtech
4.5/5
Red rows show common data quality issues. dataclean.to normalizes and generates JSON-LD automatically.