Glade APIGlade API
Sign inGet an API key

Product research

Product research from query to comparable evidence

Move from a marketplace search to stable product records, variants, offers, reviews, and category context through one normalized Amazon data contract.

Start freeSee the recipe

/api/amazon

One query, a research-ready record

GET/api/amazon/searchGET/api/amazon/productGET/api/amazon/product/variantsGET/api/amazon/categories

Reliable product research preserves where and when every fact was observed. Glade API keeps marketplace, identifier, freshness, and request metadata attached, so teams can compare evidence without silently mixing locales or time windows.

What teams build with it

Start with the question you have. Each card is a practical research chain, not a separate product or integration.

Query to evidence set

Search a marketplace, preserve placement, then resolve the products that deserve deeper analysis.

Endpoint chainsearch -> product -> offers

Variant map

Connect size, color, and configuration siblings without treating them as unrelated listings.

Endpoint chainproduct -> variants

Identifier reconciliation

Move safely between ASIN and GTIN when joining Amazon records to an external catalog.

Endpoint chaingtin-from-asin -> asin-from-gtin

Competitive offer scan

Resolve the record before comparing condition, seller, price, delivery, and availability.

Endpoint chainproduct -> offers

Category discovery

Navigate a normalized taxonomy before drilling into a category and its current products.

Endpoint chaincategories -> category

Demand proxy review

Place review evidence and best-seller position beside the product record without merging their meanings.

Endpoint chainproduct -> reviews -> bestsellers

How the product research workflow works

Four moves turn an open-ended query into a bounded set of comparable Amazon records.

Store request and freshness metadata with the evidence. Reproducible research depends on knowing the selected market and observation time.

  1. GET /api/amazon/search

    Build the initial candidate set

    Run a bounded marketplace search with documented filters while preserving organic and sponsored placement.

    bounded pages · 15 minute cache
  2. GET /api/amazon/product

    Resolve stable product records

    Turn candidate ASINs into normalized titles, identifiers, images, ratings, and availability context.

    one record per ASIN
  3. GET /api/amazon/product/variants

    Connect sibling configurations

    Inspect variant dimensions and related ASINs before calculating assortment or coverage.

    variant cache · 12 hours
  4. GET /api/amazon/product/reviews

    Add bounded customer evidence

    Retrieve explicit review pages and rating filters, keeping source records separate from your interpretation.

    bounded pages · 1 hour cache

APIs used in product research

Use only the operations your research question needs. Every API keeps the same authentication and response envelope.

Compared with the usual setup

Product research becomes fragile when data from different locales, timestamps, and parsers is treated as directly comparable.

CapabilityGlade APITypical setup
Marketplace coverageOne key works across 13 supported Amazon marketplaces.A locale-specific integration or scraper configuration per marketplace.
Response shapeOne versioned response envelope and normalized operation catalog.Different page shapes, selectors, and field names maintained by hand.
Identifier joinsDocumented ASIN and GTIN conversion operations use validated inputs.Identifier mappings are inferred from page content or maintained as custom lookup logic.
FreshnessFetched time, cache state, and operation-specific TTLs travel with the result.Freshness is inferred from a job timestamp or omitted entirely.
Failure handlingTyped failures stay separate from empty or out-of-stock results.Retries, proxy failures, parser failures, and empty pages can look alike.
Pricing modelSuccessful normalized operations consume one published request unit.Proxy, browser, compute, storage, and maintenance costs are metered separately.

Product research questions

Can I start from a search term instead of an ASIN?+

Yes. Use product search to produce a bounded candidate set, then resolve selected ASINs through product details.

Are sponsored and organic positions distinguishable?+

Yes. The search contract keeps placement semantics explicit so downstream ranking work does not silently mix the two.

Can I join Amazon products to an external catalog?+

Use the validated ASIN-to-GTIN and GTIN-to-ASIN operations, then retain the marketplace and request metadata used for the mapping.

How should I compare marketplaces?+

Keep currency, locale, delivery context, and observation time visible. Normalize display and reporting after preserving those source facts.

Does Glade interpret review text for me?+

The source API returns bounded review records. Any semantic analysis should be an explicit downstream step with its own retention and evaluation policy.

Can research results be reproduced?+

Persist the request parameters, request ID, schema version, cache state, and fetched time alongside the records used in a decision.

Ask AI about Glade API

Ask ChatGPTAsk PerplexityAsk GeminiAsk ClaudeAsk Grok

Start a research set with one query

Use a documented search, resolve the products you care about, and keep every observation auditable.

Start freeRead the docs

Request

curl -G "https://gladeapi.com/api/amazon/search" \
  -H "API-KEY: glade_live_YOUR_API_KEY" \
  --data-urlencode "searchTerm=mechanical%20keyboard" \
  --data-urlencode "domain=US" \
  --data-urlencode "page=1"
curl -G "https://gladeapi.com/api/amazon/product" \
  -H "API-KEY: glade_live_YOUR_API_KEY" \
  --data-urlencode "asin=B0D1XD1ZV3" \
  --data-urlencode "domain=US"

Both calls use the same key, response envelope, request metadata, and documented failure model.