Enriching transaction data with structured place information is critical for accurate analytics, fraud detection, and AI workflows. SafeGraph Places provides verified store-level context to help fintechs, payment processors, and data platforms clean and standardize messy merchant strings at scale.
In transaction data, the same physical store can appear under dozens of messy, inconsistent names. These discrepancies aren’t just cosmetic – they introduce real friction in analytics, UX, and risk analysis. In this blog, we’ll walk through a few real-world examples of how messy merchant strings create friction – and how structured enrichment solves it.
1. Messy Strings for the Same Store A single Whole Foods location might appear as:
2. Brand Confusion Across Sources "Ashley Home Store" vs. "Ashley Furniture" vs. "Ashley Outlet" — all the same brand, but difficult to reconcile without identifiers.
3. Hidden Geographic Impossibilities A card used in Atlanta (WF MKT 1615) is used again 20 mins later in Chicago (WholeFoods #2921). Without place context, fraud like this could be missed.
These examples highlight a common challenge: raw transaction strings are messy and inconsistent, making it hard to extract accurate merchant-level insights. That’s where enrichment comes in.
Here’s how teams are already using SafeGraph’s enriched POI data to solve this:
Payment Processors
FinTech Platforms
Merchant Intelligence Teams
AI & ML Teams
Spend Analytics / Consumer Research
📌 Case Study: Plaid, a leading fintech infrastructure provider, uses SafeGraph Places data to improve the precision of their transaction enrichment. By integrating store_id and other SafeGraph attributes, Plaid connects roughly 50% of card-present transactions to verified merchant locations — significantly improving match accuracy while reducing manual cleanup cycles. (Read the case study)
SafeGraph’s store_id, name_aliases, mcc, and naics_code columns anchor ambiguous transaction strings to a single, verified POI.
The result:
✅ Fewer false positives
✅ Cleaner attribution
✅ Stronger fraud signals
✅ Better analytics
✅ More usable user experiences
Definition: The unique ID associated with the store as provided and maintained by the store/brand itself.
Most store_ids are alphanumeric and can be found directly on official store locators. Some are displayed in plain sight next to the store name while others are embedded within the store locator URL or other non-obvious places. For example, the store_id for a Dunkin’ store is “352872” (appended at the end of the URL).
store_id is especially useful as a join key when working with transaction data. For example, “TJ256Y8” may be the only location-specific information within a transaction dataset. A Places dataset that also contains "TJ256Y8" as a store_id enables a join to contextualize transaction data (or other internal, store-level data) with SafeGraph places information.
Definition: An array of alternative names for the place. These can include common colloquial names, registered business names, store-specific names within a brand, and/or parent company names. Values are ordered by string length from smallest to largest.
Do you call your local coffee shop by its old name while transplants call it the updated name under new ownership? Is it "Ashley Furniture" or "Ashley Home Store," and are both correct? These sorts of name differences permeate transaction data, making joining to places data especially challenging.
We’re excited to bridge this gap through name_aliases, to accelerate the training of entity resolution models or building embeddings that support semantic search and categorization for fintech use-cases.
Definition: An array of Merchant Category Codes describing the business.
Merchant Category Codes (MCCs) classify businesses based on the type of goods or services they offer. These codes are widely used across payment ecosystems to categorize, track, and analyze transactions.
The mcc column in SafeGraph contains an array of all known MCCs associated with the business, ordered from most to least commonly used at the point of sale. Example: [5942, 5814, 5945].
Definition: SafeGraph uses the 2017 North American Industry Classification System (NAICS) developed by the US Census Bureau.
NAICS codes are hierarchical:
While developed for US industry classification, NAICS has proven effective internationally and will continue to be our core classification system until further notice.
These fields are designed for use directly in enrichment workflows. Here’s how to join and filter your transaction logs:
With store_id and name_aliases, SafeGraph provides scalable tools to turn noisy merchant descriptors into precise POI matches. As AI continues to reshape fintech and commerce, data cleanliness becomes a competitive advantage. Enriched POI data from SafeGraph helps ensure your models, agents, and analytics operate on trusted, contextualized merchant signals.
