Data Reconciliation
Automated Data Reconciliation Glossary
A
Audit Trail
A chronological record of reconciliation activity showing what data was compared, when it was compared, what discrepancies were detected, and how they were resolved.
Automated Data Reconciliation
The practice of using software to continuously compare data across two or more systems, identify discrepancies against defined matching rules, and surface them for review or automatic resolution without manual record-by-record comparison.
B
Break
A detected mismatch between systems being reconciled. A break may occur when a record is missing from one system, values conflict, or a difference falls outside an approved tolerance.
Break Resolution
The process of investigating, correcting, and closing a detected reconciliation break. Resolution may be manual or automated depending on the configured rules.
C
Control Total
An aggregate value used to validate that transferred or represented data is complete and consistent, such as record counts, transaction totals, or monetary totals.
Cross-System Comparison
The comparison of corresponding data between two or more systems to determine whether expected records and values agree.
D
Data Consistency Check
A validation that determines whether corresponding data remains consistent across systems according to defined rules, transformations, and tolerances.
Data Discrepancy
A difference between corresponding data sets or records that does not meet defined reconciliation criteria.
Data Integrity Validation
The process of checking whether data remains accurate, complete, and consistent as it moves between or is maintained across systems.
Data Mismatch Detection
The identification of records or values that fail defined reconciliation matching criteria.
E
Exception Management
The process of identifying, routing, tracking, investigating, and resolving reconciliation exceptions or breaks.
Exception Workflow
The sequence of actions followed after a reconciliation break is detected, including assignment, investigation, resolution, approval, and closure where applicable.
F
False Positive
A reconciliation alert indicating a difference that is not a genuine data problem, often caused by an expected transformation, timing difference, rounding difference, or poorly configured tolerance.
M
Matching Engine
The component of a reconciliation solution that applies matching rules to align records and evaluate whether corresponding data agrees.
Matching Key
A field or combination of fields used to align corresponding records between systems, such as transaction ID, order number, customer ID, or account number.
Matching Rule
The logic used to determine whether records or values from different systems should be considered equivalent. Rules may use exact, tolerance-based, or transformation-aware matching.
N
Normalization
The process of standardizing data formats before comparison so equivalent values can be evaluated consistently across systems.
R
Reconciliation Engine
The processing layer that receives or extracts data, applies normalization and matching logic, identifies matches and breaks, and produces reconciliation results.
Reconciliation Frequency
How often a reconciliation process runs, such as hourly, daily, monthly, or near real time.
Reconciliation Process
The structured process of comparing corresponding data across systems, identifying discrepancies, and managing them through resolution workflows.
Reconciliation Report
A summary of reconciliation results showing matched records, unmatched records, breaks, totals, exception status, and other control information.
S
Source System
The system from which data is being reconciled and is generally treated as the originating or authoritative reference for the comparison.
Source-to-Target Validation
The process of checking whether data delivered from a source system is accurately represented in a target system after expected transformations are applied.
System-to-System Data Matching
The comparison and alignment of corresponding records across two or more systems to determine whether they represent the same expected data.
T
Target System
The downstream system against which source data is compared. The target may contain an exact copy of source data or a transformed representation governed by reconciliation rules.
Tolerance
An acceptable range of difference within which two values are considered reconciled. Tolerances are commonly used for rounding, currency conversion, timing, and other expected variations.
Transformation-Aware Matching
A reconciliation approach that recognizes an expected difference between source and target values when that difference follows a defined transformation rule.
U
Unmatched Record
A record that exists in one system but cannot be matched to a corresponding record in the other system according to configured matching rules.
V
Validation Rule
A predefined condition used to determine whether reconciled data meets expected completeness, consistency, equality, or tolerance requirements.
Key Reconciliation Concepts
Exact Match
A comparison in which corresponding values must be identical to be considered reconciled.
Tolerance-Based Match
A comparison in which values are considered reconciled when their difference falls within an approved tolerance.
Record-Level Reconciliation
Comparison performed at the individual record level using one or more matching keys.
Aggregate-Level Reconciliation
Comparison using totals or aggregated values, such as transaction counts or monetary control totals.
Continuous Reconciliation
Ongoing reconciliation performed at defined intervals or in near real time rather than as a one-time validation.
Key Acronyms
| Acronym | Meaning |
| API | Application Programming Interface |
| ERP | Enterprise Resource Planning |
| ETL | Extract, Transform, Load |
| POS | Point of Sale |
| QA | Quality Assurance |
| SQL | Structured Query Language |
Related Reading
See Data Reconciliation in 4DAlert
Explore how 4DAlert implements the concepts in this guide as a working platform.