Eliminating Data Inconsistencies
TL;DR: Tilores’ Consistency feature stops entity resolution from creating conflicting identifiers within a single resolved entity. Instead of only comparing records pair-wise, consistency rules validate a new record against all existing records in the target entity, so one customer or company cannot end up with two VAT numbers, SSNs, or license IDs.
| Approach | How it works | Risk | Best for |
|---|---|---|---|
| Pair-wise matching rules only | Compares two records at a time; transitive chains can pull in conflicting records | Multiple VAT numbers or SSNs within one entity | Initial deduplication where identifiers overlap |
| Consistency rules (Tilores) | Before merging, validates new record against all existing records in the target entity | Conflicts detected and merge blocked before they enter the entity | Identifiers that must be unique per entity (VAT, SSN, license ID) |
On this page
- The Hidden Cost of Inconsistent Entity Data
- Introducing Consistency: Business Intelligence You Can Trust
- Real-World Business Impact
- Why Consistency Rules, Not Just Matching Rules?
- How It Works: Smart Validation Without Performance Sacrifice
- The Bottom Line: Better Decisions Through Better Data
- Getting Started
- Frequently Asked Questions
When entity resolution systems identify multiple records as the same business or customer entity, they can inadvertently create new data quality problems. One of the most persistent challenges? Inconsistent entity data that creates confusion, compliance risks, and poor decision-making. Today, we announce a new solution: Tilores’ new Consistency feature.
The Hidden Cost of Inconsistent Entity Data
Picture this scenario: Your entity resolution system identifies three company records as the same business entity based on matching company names. However, due to pair-wise matching logic, this single entity now contains three different VAT numbers. Suddenly, your “clean” data becomes unreliable for:
- Compliance reporting - Which VAT number is correct for regulatory filings?
- Financial analysis - How do you reconcile transactions across multiple tax identifiers?
- Customer management - Sales teams receive conflicting information about the same client
- Risk assessment - Credit decisions become challenging when entities have multiple inconsistent identifiers
This creates a fundamental trust problem with your data that ripples through every business decision.
Introducing Consistency: Business Intelligence You Can Trust
Tilores’ new Consistency feature solves this challenge by ensuring that critical attributes remain consistent across all records within a single entity. Unlike traditional pair-wise matching that can create these inconsistencies, our consistency rules validate that essential identifiers—like VAT numbers, social security numbers, or license IDs—remain unique and consistent within each resolved entity.
The key insight is that consistency requirements vary by identifier type. While it’s perfectly acceptable for a company entity to have multiple customer IDs from different systems, having multiple VAT numbers indicates a data quality problem. The Consistency feature is designed to be applied selectively—you configure it only for those identifiers where duplicates signal incorrect matching, not for those where multiple values are expected and legitimate.
Data quality issues that originate in entity resolution can cascade into downstream master data management pipelines, compliance workflows, and customer-facing systems. Tilores sits alongside MDM, CDPs, and governance tooling rather than replacing them, so consistency at the resolution layer means cleaner inputs everywhere.
Real-World Business Impact
For Financial Services
Banks using Tilores for customer due diligence can now ensure that each customer entity has only one social security number, eliminating the risk of accidentally merging different individuals’ financial profiles. This reduces compliance violations and improves regulatory reporting accuracy.
For B2B Companies
Enterprise sales teams can trust that each company entity has a single, consistent VAT number, enabling accurate invoicing, tax reporting, and financial forecasting. No more manual verification of conflicting tax identifiers.
For Healthcare Organizations
Patient safety improves when medical records ensure consistent patient identifiers. The consistency feature prevents dangerous scenarios where multiple patients might be merged due to similar names while maintaining different medical record numbers.
For regulated industries like banking and insurance, consistent entity data also feeds directly into EBA connected-clients identification and KYC-AML workflows where a single incorrect merge can trigger reporting failures.
Why Consistency Rules, Not Just Matching Rules?
You might wonder why we created a separate consistency feature instead of simply using matching rules. The answer lies in how pair-wise matching works versus entity-wide validation.
Consider these three records for “John Smith”:
- Record A: Has SSN “1234567890”
- Record B: Has no SSN
- Record C: Has SSN “0987654321”
With traditional matching rules, Record A matches with Record B (same name, no conflicting SSN), and Record B matches with Record C (same name, no conflicting SSN). Through this chain, all three records end up in the same entity—even though Records A and C have different SSNs.
Consistency rules solve this by evaluating the new record against all existing records in the potential target entity. When Record C is being processed, the consistency rule checks it against both Records A and B. Since Record C conflicts with Record A’s SSN, the entire entity is rejected as a match, preventing the inconsistent merge.
Understanding how entity resolution works in practice makes it clear why pair-wise logic alone cannot catch transitive inconsistencies, and entity-wide validation is a separate concern.
How It Works: Smart Validation Without Performance Sacrifice
The Consistency feature operates intelligently:
- Validates during entity resolution - Before adding a new record to an entity, the system checks all existing records within that entity
- Handles missing data gracefully - Records without certain identifiers can still be matched, maintaining flexibility
- Optimized for performance - Internal optimizations ensure consistency checking doesn’t slow down your resolution process
Check our technical docs for detailed information about how to use this new feature.
The Bottom Line: Better Decisions Through Better Data
With Tilores’ Consistency feature, organizations can:
- Reduce manual data validation efforts significantly
- Improve compliance accuracy with consistent regulatory identifiers
- Accelerate decision-making with trustworthy, consistent entity data
- Minimize operational risk from data inconsistencies
Getting Started
The Consistency feature is available now for all Tilores customers. Our team is ready to help you configure consistency rules tailored to your specific use case and data schema.
Ready to eliminate data inconsistencies and unlock reliable business intelligence? Contact our team to learn how the Consistency feature can transform your entity resolution process.
Tilores is the modern entity resolution platform that helps organizations create a single source of truth from fragmented data. Learn more at tilores.io
The W3C Data Quality Vocabulary (DQV) includes consistency as a named data quality dimension, characterising it as data that is free from contradiction and coherent with other data in a specific context of use, a principle that Tilores’ Consistency feature operationalises directly in the entity resolution layer.
See what resolved entity data does for your business — and your AI.
FAQ
What is the Tilores Consistency feature?
The Tilores Consistency feature is a validation layer that ensures critical attributes (like VAT numbers, social security numbers, or license IDs) remain unique and consistent across all records within a single resolved entity. It checks a new record against all existing records in the potential target entity before completing a merge, preventing conflicting identifiers from ending up in the same entity.
Why can pair-wise matching create data inconsistencies?
Pair-wise matching compares records two at a time. If Record A matches Record B and Record B matches Record C, all three end up in the same entity even when Records A and C have conflicting identifiers. For example, Record A might have SSN 1234567890 and Record C might have SSN 0987654321 and pair-wise matching misses that conflict because it never directly compares A and C. Consistency rules solve this by evaluating the new record against all existing records in the potential target entity.
Which identifiers should use consistency rules?
Consistency rules should be applied selectively: only for identifiers where duplicates signal incorrect matching. VAT numbers, social security numbers, and license IDs are good candidates because a single entity should have exactly one of each. Customer IDs from different source systems are not good candidates because a company entity legitimately has multiple customer IDs across systems.
Does consistency checking affect entity resolution performance?
No. The Consistency feature includes internal optimizations that ensure consistency checking does not slow down the resolution process. It also handles missing data gracefully: records without certain identifiers can still be matched, maintaining flexibility while enforcing consistency where it matters.
Who can use the Tilores Consistency feature?
The Consistency feature is available now for all Tilores customers. You can review the technical documentation at docs.tilotech.io to configure consistency rules tailored to your specific use case and data schema.
See it on your own data: book a demo for a walkthrough on your records, or get the evaluation build to try resolved entity data locally.
See what resolved entity data does for your business — and your AI.