Industries
Description:
The industries that describe the business model of the business. Enigma uses the North American Industry Classification System (NAICS) hierarchy.
Child attributes (and data file structure):
Column Name | Data Type | Description | Example |
---|---|---|---|
classification_code | string | The NAICS code of the business industry. | 2361 |
classification_description | string | The description of the NAICS code. | Residential Building Construction |
classification_type | string | The type of code shown. | NAICS 2017 |
JSON Sample:
{
"industries":
[
{
"classification_type": "NAICS_2017",
"classification_code": "2361",
"classification_description": "Residential Building Construction",
}
}
Coverage:
- Businesses: 69.7%
- Business Locations: 60.7%
Time structure:
- Current point in time. No historical information.
Data sources:
- Public data (e.g. corporate filings, PPP applications, etc.) where a business specifies their industry
- Online business directories where a business specifies their industry
- Information and keywords on a business’ website
- MCC codes from credit card transactions
- Third party verification services
Methodology:
- All of the data sources listed above go into a model to determine a businesses’ industry
Enigma only tags a business with an industry code if we have greater than 85%+ confidence that the industry specification is correct. - Enigma’s system attempts to tag the most granular NAICS code possible while maintaining high accuracy. The system starts by attempting to tag a six digit code, but if the confidence level is less than 85%, then Enigma will try to provide five digit granularity. If Enigma’s system does not have high enough confidence in a five digit code, then it will move on to a four digit code. This process continues all the way to a two digit code.
- As of March 2022, the breakdown of NAICs codes by granularity, given the business has an industry, is:
2 digit granularity or deeper | 100% |
3 digit granularity or deeper | 81% |
4 digit granularity or deeper | 51% |
5 digit granularity or deeper | 35% |
6 digit granularity | 27% |
- A business location profile may have multiple industry codes. This is reflective that locations may have primary and secondary industries (e.g. a business can be a restaurant and can also sell packaged food).
- Business profiles, however, can only have one primary industry. This is determined by looking at the most prevalent industry across all locations associated with that business.
Updated 6 months ago