StoreRouter vs. Datafiniti
Each store's product vs. one merged record per product.
Datafiniti crawls retail sites, merges listings that share an identifier such as a UPC into one record per product, and sells those records through search, bulk downloads and an MCP server. StoreRouter answers search and parse from an index of each covered store's catalog, kept current by change detection: the product as that store sells it, with its variants, per-size stock and price, in one schema.
- /v1/products/search
- /v1/products/parse
- JSON · Markdown
Evidence recorded October 27, 2026 · StoreRouter, formerly Octogen
| Capability | StoreRouter | Datafiniti |
|---|---|---|
| Model | Index of each store's catalog behind one API | Product database merged across sources, sold by the record |
| Unit of record | One store's product: its URL, variants, stock and price | One product merged across sources, with a list of price observations |
| Discovery | search with a plain-language shopping query | Lucene field queries such as brand: and prices.amountMin:<100 |
| Freshness | Change detection keeps the index current; force a live parse | Rolling re-crawls; dateUpdated per record; targeted updates on request |
| Agent access | OAuth MCP, REST, CLI, TypeScript and Python SDKs | Hosted MCP with bearer token (OAuth on claude.ai), REST, web portal |
| Billing | Successful 2xx responses only | One credit per record returned or downloaded; zero results are free |
| Pricing | $2 / 1k parse · $14 / 1k search | $99 / month for 5,000 records to $999 / month for 200,000 |
| Free tier | $5 credit every month | 1,000 records for two weeks |
Fit
Which one fits
Choose StoreRouter when
your agent shops: it takes a request in plain language, finds products across stores, and needs each store's current variants, sizes, stock and price to recommend something a person can buy right now.
Choose Datafiniti when
you need product data in bulk: identifier-keyed records (UPC, EAN, ISBN, ASIN) with price history across merchants, broad US coverage in categories such as automotive, grocery and books, or full dataset downloads for analytics and model training.
Overview
What each tool does
Datafiniti sells product, business, property and people data. Its product data comes from crawling retail sites and licensed providers. Each crawled listing is normalized, then merged with any other listing that shares an identifier: a UPC, EAN or ISBN, or a brand or manufacturer plus manufacturer number. One record can carry prices, descriptions and reviews from several retailers, each with its source URL and the date it was seen. You query records with a Lucene field syntax through the API, the web portal or an MCP server, and download larger result sets as files.
StoreRouter keeps a product index across thousands of covered stores and keeps it current by detecting changes and re-crawling them. search takes a shopping query across stores and returns products; parse turns a product URL into the shared schema: variants, per-size stock, price, images, materials and normalized attributes. You call an API; crawling, unblocking and extraction are ours.
This page compares documented approaches. We have not benchmarked Datafiniti's records, coverage or freshness against StoreRouter's.
Architecture
Where the work happens
- Unit of record. Datafiniti's record is a product merged across sources:
pricesis a list of observations, each with a merchant, a date and source URLs, andcolorsandsizesare lists for the product as a whole. StoreRouter's record is one store's listing, with that store's variants and stock for each size, so an agent can answer "is the medium in stock at this store?" from one response. - Discovery. Datafiniti search is a field filter: the agent or developer writes
categories:(laptops OR desktops) AND brand:Apple AND prices.amountMin:<1000. StoreRoutersearchtakes the shopper's request as written and returns matching products across covered stores. - Freshness. Datafiniti re-crawls sources on rolling schedules and stamps each record with
dateUpdated; a targeted-updates endpoint and a scrape API (up to 10 URLs per request, one request at a time) refresh records from supported sources on demand. StoreRouter answers from an index that tracks changes on covered stores;parsewithresolutionMode: "on_demand_only"andonDemandCachePolicy: "refresh"forces a live read. - Agent workflow. Datafiniti's MCP server exposes query tools: count matches, search up to 50 records, start a download, fetch a record, read the schema (Datafiniti MCP). The agent writes Lucene queries against the schema. StoreRouter's MCP exposes shopping operations:
search_productsandlookup_product. - Reach. Datafiniti covers a wide range of US product categories and identifiers, with history. StoreRouter covers the catalogs in its index;
parsecan resolve other product URLs on demand, with sparser results and no enrichment.
Before you build
Integration notes
- Which stores does each cover?
searchspans StoreRouter's covered catalogs; check catalog coverage for the stores your agent needs.parsecan resolve a product URL from an uncovered store on demand, but that result can be sparser, skips enrichment, and doesn't make the store searchable. Datafiniti publishes record counts by category rather than a store list, and describes its coverage as primarily US.- How does pricing compare?
StoreRouter: $2 per 1,000 parse requests and $14 per 1,000 search requests; only successful 2xx responses are billed (see pricing). Datafiniti sells product records by the month: $99 for 5,000, $259 for 20,000, $499 for 50,000, $999 for 200,000, or unlimited for $60,000 a year (Datafiniti pricing). A search is charged one credit per record returned, so a search that returns ten products costs ten records; a search that finds nothing is free.
At volume, Datafiniti's per-record price falls below StoreRouter's parse price. Compare what the record holds: a merged product with price observations, or one store's live listing with its variants and stock.
- Can I use this inside an AI agent or LLM workflow?
Yes, both. StoreRouter's MCP server signs in with OAuth in every client, works with Claude, Cursor and any MCP client, and returns shopping results directly (MCP docs). Datafiniti's hosted MCP server uses your API token as a bearer header in most clients and OAuth on claude.ai; its tools take Lucene queries, and it publishes guides for using it with Claude and ChatGPT.
- Can I use both?
Yes. Use Datafiniti for identifier-keyed product records, price history and bulk datasets, and StoreRouter when an agent needs to search stores and read each store's current listing.
Give your agent every store.
Mint a key, send search a real shopping query, and parse the products your agent would recommend.
Documentation reviewed · 2026-10-27
Sources and scope
This page compares documented approaches. We have not run a Datafiniti benchmark or established better record completeness, accuracy, coverage, freshness or latency.
- Product data: coverage, access methods, plans, trial.
- How product records are sourced and merged.
- Product data schema and query syntax.
- MCP server: endpoint, authentication, tools.
- How credits work: per-record charging.
Prices are list rates on the review date, before the 10% annual discount.