Documentation
Product VideosAPI ReferenceChangelogFAQDashboard

Data verification methodology: 1x1 Matching

Understand how 1×1 Matching works in Database Validation.

This page outlines the 1x1 Matching methodology used in the Database Validation solution. The underlying rules are pre-set to optimize user approval rates without compromising the level of verification assurance.

📘

Note

The data matching rules are standard for all Sumsub clients and not available for customization.

Database Validation

The 1x1 Matching methodology requires matching user’s personal data against one source to receive a Match on their identity.

Verification result

Name

ID number

Date of birth
Match
Match
Match
Any Value
Partial Match
Partial Match
Match
Any Value
No Match

All other combinations
📘

Note

  • All available data sources are screened until a Match on the identity is made. If a Match is not found, the system then repeatedly checks those sources for a Partial Match.
  • By default, applicants who receive a Partial Match verification result are approved. If you prefer to reject them, contact your Customer Success Manager or the Support team.

Response definitions

The tables below describe the possible response values. Attribute responses come directly from data sources, while category results are derived by Sumsub based on the Data matching scenarios.

Attribute responses

Response

Description
Match

The provided attribute matches the record in the database.
No Match

The provided attribute does not match the record in the database.
No Data

No database record is available for this attribute.

Category results

Response

Description
Match

The attribute results meet at least one Match combination for the category.
Partial Match

The attribute results meet a Partial Match combination, but no Match combination.
No Match

The attribute results meet no Match or Partial Match combination.

Data matching scenarios

This section details the matching scenarios in each data category that are relevant within Database Validation: Name, Date of birth, and ID number.

Name matching

Some countries, particularly those that follow Spanish naming conventions, record two surnames for each person:

  • Paternal name: the first surname, traditionally inherited from the father.
  • Maternal name: the second surname, traditionally inherited from the mother.

When a data source stores these surnames separately, the Paternal name is treated as the applicant's Last name, and the Maternal name is treated as an additional last name component. If the Last name submitted for the applicant contains two words separated by a space, the value is split: the first word is compared as the Paternal name, and the second word as the Maternal name. Because the Paternal name and the Maternal name can each match on their own, an applicant can still receive a Name Match when a source record contains only one of their surnames.

Name Match requires meeting any of the following combinations:

  • First name Match + Last name Match
  • First name Match + Maternal name Match + Paternal name Match
  • First name Match + Maternal name Match
  • First name Match + Paternal name Match
  • First initial Match + Last name Match
  • First initial Match + Maternal name Match
  • First initial Match + Paternal name Match
  • First initial Match + Maternal name Match + Paternal name Match

Name Partial Match requires meeting any of the following combinations:

  • First name
    Match
  • Last name
    Match
  • Paternal name
    Match
  • Maternal name
    Match
  • Maternal name
    Match
    + Paternal name
    Match

Date of birth matching

Date of birth Match requires meeting the following combination:

  • Year of birth Match + Month of birth Match + Day of birth Match
📘

Note

Certain countries may have exceptions to the default Date of birth matching rules.

ID number matching

ID number Match requires meeting the following combination:

  • Identification number
    Match

Data matching sensitivity

These are the criteria for fuzzy matching on different data attributes against country sources.

  • Name

    Name matching requires the data input to have 70% of the Levenshtein character similarity in comparison to the source record.

  • Date of birth and ID number

    Date of birth and ID number matching requires an exact match between the data input and the source record.

Examples

The following examples demonstrate how the matching sensitivity level works in practice:

Data field

Input data

Source record

Result

first_name

Jeanette

Jeanette
Match

first_name

Jeanotte

Jeanette
Match

first_name

Gene

Jeanette
No Match

last_name

Richardson

Richardson
Match

last_name

Richardsen

Richardson
Match

last_name

Richarliset

Richardson
No Match

Updated about 4 hours ago

Did this page help you?