Backed by
Combinator
CombinatorTransaction Enrichment {API}
Identify merchants from bank statement strings and enrich them with brand data.
Turn descriptor strings like "SQ *BLUE BOTTLE OAK" into clean merchant data — name, logo, domain, colors, category. Built for fintech, personal finance, expense, and fraud-detection workloads.
Get API AccessNo card required
View Documentation
What You Get
Each API call converts transaction text into structured brand data.
Company name
Clean, official business name
Domain & logo
Official website and high-res logo
Brand metadata
Colors, description, and social profiles
Category
Business classification and industry
How It Works
We use advanced matching algorithms to identify merchants from transaction descriptions.
— step 01 —
Send transaction text
Like GREGORYS COFFEE $9.99
— step 02 —
String parsed
Descriptor parsing and optional location, MCC, or phone context
— step 03 —
Merchant matched
Known merchant database and pattern analysis
— step 04 —
Brand returned
Complete business identity with domain, logo, and category
API Response
Example brand match for Nespresso (optional fields may be absent)
POST /v1/brand/retrieve{
"status": "ok",
"brand": {
"domain": "nespresso.com",
"title": "Nespresso",
"description": "World leader in coffee...",
"colors": [{ "hex": "#8a8c92", "name": "Ancient Pewter" }],
"logos": [{ "url": "https://media.context.dev/...", "mode": "light", "type": "logo" }]
},
"code": 200
}Frequently asked questions
Common questions about the Context.dev Transaction Enrichment API.
Am I billed for failed requests?
No. You are not billed for failed requests or requests where we are blocked (rarely happens). Credits are only consumed on successful responses.
How do I clean up a bank statement transaction?
POST one raw descriptor string to /v1/brand/retrieve with type set to by_transaction. The API returns the matched brand record when it can identify the merchant; a miss returns a not-found error.
What merchant fields are returned?
When a match is found, the Brand response can include the merchant title, domain, description, discovered logos and colors, social URLs, and industry data. Fields are optional. Pair it with the separate NAICS or SIC APIs if you need those classifications.
Can I use this for expense management or fraud detection?
Yes. Common use cases: auto-categorizing transactions in budgeting apps, surfacing rich merchant cards in card statements, deduplicating recurring subscriptions, and feeding fraud models with stable merchant identifiers (domain) instead of raw descriptor strings that change between processors.
What is a transaction enrichment API?
A transaction enrichment API converts the messy strings on bank statements ("SQ *BLUE BOTTLE OAK", "AMZN MKTPL*A12B3") into structured merchant data: clean name, official domain, logo, brand colors, category, and description. Context.dev returns the same data shape used in modern fintech UIs and personal finance apps.
Is this a Plaid Enrich or Ntropy alternative?
Context.dev can handle the merchant-brand enrichment step from a transaction descriptor. It is not a replacement for transaction feeds, account connectivity, categorization, or every feature of broader financial-data products.
Does it handle international merchants and small businesses?
Coverage varies by merchant and descriptor. Supply country_gl, city, MCC, or phone context when available. Use high_confidence_only when you prefer an abstention over a less certain match, and handle not-found responses explicitly. We do not publish an unmeasured long-tail or UK coverage percentage.
Am I billed for failed requests?
No. You are not billed for failed requests or requests where we are blocked (rarely happens). Credits are only consumed on successful responses.
What is a transaction enrichment API?
A transaction enrichment API converts the messy strings on bank statements ("SQ *BLUE BOTTLE OAK", "AMZN MKTPL*A12B3") into structured merchant data: clean name, official domain, logo, brand colors, category, and description. Context.dev returns the same data shape used in modern fintech UIs and personal finance apps.
How do I clean up a bank statement transaction?
POST one raw descriptor string to /v1/brand/retrieve with type set to by_transaction. The API returns the matched brand record when it can identify the merchant; a miss returns a not-found error.
Is this a Plaid Enrich or Ntropy alternative?
Context.dev can handle the merchant-brand enrichment step from a transaction descriptor. It is not a replacement for transaction feeds, account connectivity, categorization, or every feature of broader financial-data products.
What merchant fields are returned?
When a match is found, the Brand response can include the merchant title, domain, description, discovered logos and colors, social URLs, and industry data. Fields are optional. Pair it with the separate NAICS or SIC APIs if you need those classifications.
Does it handle international merchants and small businesses?
Coverage varies by merchant and descriptor. Supply country_gl, city, MCC, or phone context when available. Use high_confidence_only when you prefer an abstention over a less certain match, and handle not-found responses explicitly. We do not publish an unmeasured long-tail or UK coverage percentage.
Can I use this for expense management or fraud detection?
Yes. Common use cases: auto-categorizing transactions in budgeting apps, surfacing rich merchant cards in card statements, deduplicating recurring subscriptions, and feeding fraud models with stable merchant identifiers (domain) instead of raw descriptor strings that change between processors.
Ship an agent that actually knows things.
Free tier, 10-minute integration, and the same API powering agents at Mintlify, daily.dev, and Propane. No credit card to start.