To improve the matching accuracy and handle various input variations (e.g., using only the company name + address or phone number), the matching algorithm uses a cascading series of matching steps, referred to as a Match waterfall.
Each enriched company profile output will feature a Match score and a Match method attribute. The values of these attributes determine the level of match confidence and provide information about the criteria used to match the input data with the company information in the Veridion database.
Match score
The Match score (match_score) is a numerical value that indicates the level of confidence for a match. Match scores range from 0 to 1, with higher values indicating greater confidence in the match.
Note
The Match score is a quantitative indicator of the strength of the match.
Read more about the
match_scoreattribute.
By using the match_score, you can establish a match threshold, enabling you to filter out company results that do not meet your desired confidence level.
Match method
The Match Method (match_method) indicates the step in the Match waterfall (13 steps in total) that triggered the final match. It provides information about the criteria used to match the input data with the company information in Veridion's database.
Note
The
Match methodis a qualitative indicator of the matching technique used.Read more about the
match_methodattribute.
By analyzing the match method, you can determine how the match was made and make informed decisions regarding the relevance of the result to your desired output.
Match waterfall
This is a cascading series of matching steps used by the matching algorithm. These steps become gradually less strict and combine the matching fields in different ways, considering the reliability of each field during the matching process.
For example, a country-based match on a country provides less confidence than a city-based match. Therefore, the algorithm first tries to find a city-level match before considering a country-level match as an alternative.
Here is the complete list of available match methods (or the matching waterfall structure), arranged in descending order based on the level of confidence:
| Match Method | Description |
|---|---|
website | Exact match with the website. |
website_name | Partial match with the website and an exact match with the company name. |
exact_name_street_or_postcode | Exact match with the company name and street or postcode. |
exact_name_phone_city_or_region | Exact match with company name and phone number and city or region. |
fuzzy_name_street_or_postcode | Fuzzy match with company name and street or postcode. |
fuzzy_name_phone_city_or_region | Fuzzy match with company name and phone and city or region. |
exact_name_city | Exact match with the company name and city. |
fuzzy_name_city | Fuzzy match with company name and city. |
exact_name_region | Exact match with the company name and region. |
fuzzy_name_phone_country | Fuzzy match with company name, phone, and country. |
fuzzy_name_region_country | Fuzzy match with company name, region, and country. |
fuzzy_name_phone | Fuzzy match with company name and phone. |
exact_name_country | Exact match with the company name and country. |
Analyze a Company match
Please refer to the How to analyze a Company match section for more information on Match scores and Match methods.
General notes
Company name match
The matching logic uses multiple Veridion datapoints, with different relevance weights:
- Company legal and commercial names extracted from the company's website
- Company names extracted from social media like Facebook, Twitter, Instagram etc. (where available, if the company has a good social presence)
An exact match on a company name requires any of the input names to match exactly any of the available names in the Veridion database.
A fuzzy match on a company name requires any of the input names to match partially any of the available names in the Veridion database, based on a set of rules (e.g., fuzzy matching is not allowed on legal entity terminations such as LTD, LLC etc.).
Location match
The API normalizes the input location before running the match. This allows the API to work with any level of address specificity and format.
Location matching is performed against either the main or secondary locations of the company.
Main vs. secondary locations
- Main location
A company's main location refers to its headquarters or primary place of business. It is typically the central hub where the company's core operations are managed, and key decisions are made.
- Secondary locations
Secondary locations include all other places where the company maintains a business presence, such as regional offices, branch locations, factories, or stores. These locations reflect the geographic extent of a company's operations beyond its main location.
The API matches locations using both string and geo-point datapoints:
- String-level match: the input country, region, and/or city must match exactly the corresponding information in our records.
- Geo-point match: we use the latitude and longitude information extracted from the input address to identify the closest location in our records within a predefined radius.
Geo-point limits
Geo-point matches are not permitted for country-only matches, and the maximum allowed search radius is 20km.
Phone number match
The API normalizes the input phone number before running the match. This allows the phone number to be passed in various formats, with or without country/region prefixes.
An exact match on a phone number requires the normalized input to match exactly any of the primary or secondary phone numbers in the Veridion database.
Enable Advanced Matching
By default, the API operates with limited information, such as the company name and country.
However, when extra information is supplied (e.g., a more detailed company address, website, or phone number), the matching process is designed to handle some inaccuracy in the input.
As a result, even in cases of typographical errors or minor differences in the input company name, the API can still make a match by using additional details, such as the phone number, to identify the appropriate company.