Introducing WinPure Clean&Match Enterprise v12: A New Era of Intelligent Data Matching and AI-Powered Data Quality

WinPure Clean & Match Enterprise v12 introduces a re-engineered matching engine, AdaptiveMatch™ intelligent matching recommendations, MatchGraph™ relationship visualisation and AIConnect™ connectivity. Built for organisations that need accurate, explainable data without giving up control of their processing environment, v12 brings these innovations into a unified, no-code data quality platform.

Introducing WinPure Clean&Match Enterprise v12

On a large deduplication job, much of the work goes into configuring and running match rules, then reviewing the output with little visibility into why the software treated two records as duplicates. We see this first hand, because our team works closely with enterprise data teams whose sources, volumes and matching requirements keep changing. They are in need of a matching solution that can adapt as their data changes, while giving them full visibility into why a record was marked a duplicate, what relationship does that record have with other similar records and how teams can resolve these variations into a consolidated entity or identity that stands as the single source of truth across current and future datasets.

This gap led us to build – or rather – rebuild our match engine entirely. Based on our customer feedback and our team’s relentless innovation in the data quality space, WinPure Clean&Match v12 is redefining the future of data matching – and data quality.

Our team has made sure the v12 responds directly to those needs, giving teams matching that works with the complex nature of their data (not against it!), and results they can inspect, along with consolidated records that stay consistent as new data arrives. Teams also keep control over where processing happens and how WinPure connects to their wider systems:

• Matching that adapts to your data: AdaptiveMatch™ assigns every column a data type at import and pre-fills its match condition, including the algorithm, fuzzy level and weight. Users review and adjust every suggestion before the match runs.
• Visibility into every match: Match Graph™ opens any match group as a relationship graph. It shows the score and rule behind each pair, and the bridge records that hold the group together.
• Consolidated records that persist: MatchAI™ resolves entities through our embedded entity resolution engine, and SmartMaster AI™ selects master records. GoldenID™ keeps golden record identifiers stable when data is matched again.
• AI on your own terms: AIConnect™ connects supported AI services using your organisation’s own provider credentials.
• Data quality inside your wider systems: A REST API and MCP Server connect WinPure to enterprise applications and compatible AI assistants.

This is just the tip of the iceberg for what is to come. As companies continue to invest in data quality for AI projects, WinPure aims to remain at the forefront with innovations that balances automation and AI-assistance along with security and control. We believe data is an organisation’s most valuable asset, which is why we work hard to make sure it stays within your environment, controlled by you. WinPure’s artificial intelligence technologies are all built into the software and does not require you to transmit data outside of your environment into any third-party app or platform (of course with the MCP, the choice is entirely yours!).

Clean&Match v12’s Breakthrough Performance: One Million Records Matched in 14.6 Seconds

It’s been quite a journey building the Clean&Match v12 to meet expectations not just on accuracy but also on speed and volume.

In our first internal benchmark test, the WinPure data match engine processed one million records, matched in 14.6 seconds! The match definition used six columns, combining exact conditions with WinPure Fuzzy comparisons at a 95% similarity threshold. It ran on a single Dell Precision workstation with no server cluster behind it. 

In data matching, speed does matter, especially when you are tuning rules against millions of records or preparing data for a migration with a fixed cutover date. Every rerun that takes an hour is an adjustment a team may decide not to make.

While we are excited with this breakthrough performance, we would also like to add an important caveat:

This benchmark test comes from our own internal setup, using our own sample dataset, with configurations designed on our data type and structure. We attempted to match first name and last name on state and email address, using WinPure’s proprietary fuzzy algorithm for names along with exact and numeric matches for numbers and post codes. 

The number that matters most is the one you get on your own data, and we encourage every team evaluating v12 to run that test. Take a file you already know, apply the match rules you use today and time the run on your own hardware. Then adjust a threshold and run it again. At the speed we measured, a million-record match finishes quickly enough to make that kind of iterative tuning practical within a single working session. More of the team’s time goes into reviewing matches and refining rules, and less into waiting for results.

What Can Data Managers and Engineers Expect from WinPure v12?

Our work with enterprise customers gives us a close view of what makes matching projects demanding. We have built v12 around those practical challenges, solving the time spent in manually configuring match rules to verifying duplicates and creating master records. With the v12, WinPure has not only built a match engine, but an entire data quality ecosystem that starts from raw data to master records that become the source of truth, easy to maintain and update in concurrent match runs.

Some of WinPure’s key capabilities that users can expect:

✅ Automation with full control: With 20 years in the business, our team knows how important it is for users to maintain control over their data. Automation should make the work easier while leaving people responsible for the decisions. In v12, recommendations for match conditions, cleaning options and validation patterns give you a starting point for each column, but, should you prefer setting your own rules and configuration, you can do so with complete control. You can review and change every suggestion, keeping the final say over your rules, thresholds and merges.

✅ A data library that acts as your guiding board: Once you have worked out how your organisation’s data should be handled, that knowledge is saved in a data library that you can keep building and adding to. However, you wouldn’t need to start from scratch. With WinPure’s  100+ built in data types, v12 lets you define your own, with their cleaning options, validation patterns and match conditions. You can save those standards and share the library between machines, giving colleagues a consistent approach and new analysts the benefit of work you have already done.

✅ The whole workflow in one application: We want users to spend their time working through data problems, with fewer interruptions to export files and prepare them for another tool. Profiling, cleansing, address verification, matching, entity resolution and golden record creation sit together in WinPure, supported by automation and audit logging. You can follow the work through to master record creation, then establish a repeatable process for the next dataset.

✅ Speed that gives you room to investigate: Waiting for another matching run can make it difficult to test every adjustment you want to make. That is why the rebuilt engine matters to us. In our internal benchmark, it matched one million records in 14.6 seconds. Performance varies by workload, but faster processing gives you more opportunity to test thresholds, examine groups and refine your approach before approving the output.

✅ Results you can explain with confidence: We know that approving a match carries responsibility, particularly when someone later asks why those records belong together. MatchGraph™ lets you examine the connections within a group, including the score and rule behind each link. Match Explanation™ provides the corresponding context for entity resolution, showing why records belong to an entity and why others remain possible relations. Alongside audit logs recording who changed what and when, these capabilities give you evidence to review and discuss with data owners.

These priorities guided the changes we made in v12. Take a look at some of the new features in v12 and how they work in practice.

What’s New in WinPure v12? Key Features and Improvements

AdaptiveMatch™: The Next Generation of Intelligent Data Matching

Traditionally, choosing the right matching approach would take time; you’d have to identify the type of data you’re working with and whether a match algorithm meets requirements (for example, can a simple fuzzy match algorithm handle international names?). This is particularly stressful when each field presents different patterns of variation. A method that works well for personal names may be less useful for company names, while identifiers often need an exact comparison.

We built AdaptiveMatch™ to make that configuration work faster. You no longer need to decide which algorithm can help with the match, as our engine automatically maps your data to the right kind of algorithm (WinPure supports over a dozen algorithms, each designed to meet a specific data type challenge). It examines the characteristics of individual fields and recommends suitable comparison methods, giving you a starting point that reflects your data. From there, you can review and refine the recommendations through an intuitive interface that allows for:

Intelligent Recommendations with User Control

• Choose the right method for each field: Apply different algorithms to names, organisations, addresses and other columns, so each comparison reflects the type of evidence that field provides.
• Set thresholds that suit your project: Adjust how much similarity a match requires, taking account of the variations you expect and the consequences of linking the wrong records.
• Give stronger evidence more weight: Control how much each field contributes to a match, allowing reliable identifiers or contact details to carry more weight than less distinctive values.
• Find connections across columns: Compare related information even when different sources place it in different columns, helping you investigate relationships that a comparison of corresponding fields could miss.
• Build rules with less manual setup: Use the recommendations to establish your initial configuration, then adjust it as you test the rules and review the results.

For example, two supplier records may have slightly different company names but identical postcodes and telephone numbers. AdaptiveMatch™ lets you choose how to compare those fields and how much weight to give each one, so the combined evidence informs the matching decision.

AdaptiveMatch™ works alongside WinPure’s wider entity resolution capabilities and SmartMaster AI™ golden record workflows. Its recommendations are designed to support informed configuration, not to replace human validation of consequential matches.

MatchGraph™: Bringing Transparency and Explainability to Data Matching

Once you have configured the rules, the next challenge is understanding the groups they produce. An unexpected record can be difficult to assess when you cannot easily see how it connects to the others. MatchGraph™ brings those connections into view through interactive networks of records and relationships, giving you more context when reviewing the output.

Understanding Direct and Indirect Matches

Consider three records:

A matches B at 98%, and B matches C at 96%, while A and C fall below the direct matching threshold. All three may still belong to the same group through their connection to B.

MatchGraph™ makes that structure visible, allowing you to explore direct and indirect connections, inspect relationship scores and identify the designated master record. When a group contains a questionable record, you can focus on the connection that brought it into the group.

That visibility gives you a clearer basis for deciding how to handle the records and explaining the result to a colleague or data owner before consolidation.

AIConnect™: Connecting Data Quality with Artificial Intelligence

As we extend WinPure’s AI connectivity, we want organisations to have flexibility over the services they use. AIConnect™ introduces a way to connect supported AI providers within the platform, alongside its specialist cleansing and matching engines.

Bring Your Own Large Language Model

With a Bring Your Own Large Language Model approach, you can connect a supported service using your own provider credentials. This allows you to choose a provider that fits your organisation’s requirements and use the AI interactions available through that integration.

The selected service and released functionality determine which interactions are supported. Because an external connection may transmit information to the provider, you can assess its data handling against your policies before enabling it, while WinPure’s core processing remains within your chosen environment.

Live MCP Server for AI Assistants and Agents

WinPure v12 includes a live Model Context Protocol (MCP) Server, enabling compatible AI assistants and agents to connect to supported WinPure data quality functionality through a structured integration interface. This complements AIConnect™ and the REST API, extending how authorised users and applications can interact with the platform.

Available MCP tools, supported clients, permissions and deployment requirements depend on the released server configuration and should be confirmed against WinPure documentation.

Local WinPure processing and external AI services have different data-handling implications. Before enabling a provider, organisations should review what information may be transmitted, the provider’s terms, credential handling and their own security policies.

REST API: Connecting Data Quality Across the Enterprise

For many customers, a matching project sits within a larger migration, application or recurring data process. We have expanded WinPure’s connectivity so supported data quality operations can become part of those existing workflows.

Connect WinPure to Your Applications

The REST API in v12 gives external applications access to supported WinPure functionality. Engineers can use the available operations to incorporate data quality work into migration pipelines, internal applications and repeatable processes.

This creates a route for connecting work completed in WinPure with the systems that need its capabilities, reducing reliance on separately initiated desktop tasks where an API operation can support the process.

Connect Data Quality to Enterprise Workflows

Potential applications include integrating data preparation with migration projects, connecting internal business applications to supported data quality services and building repeatable workflows around available API operations. The precise endpoints, authentication options and supported actions should be checked against the released REST API documentation.

Both the REST API and MCP Server are available in v12. The REST API supports application integration, while MCP provides a structured connection for compatible AI assistants and agents to access supported capabilities.

Data Security, Privacy and Governance by Design

Data quality tools frequently process sensitive customer, citizen and business information. For many organisations, the ability to choose the processing environment is as important as matching accuracy.

WinPure’s Windows-based architecture supports operation on on-premises infrastructure, compatible virtual machines and Windows environments hosted on Microsoft Azure or AWS. This gives teams flexibility to align core data processing with their infrastructure and governance requirements.

Control, Accuracy and Accountability

WinPure supports better information management through cleansing, profiling, duplicate identification, matching and master record selection. MatchGraph™ adds visibility into how records become connected, helping reviewers investigate possible false matches. Together, these capabilities can contribute to stronger data governance practices.

Deployment control does not automatically establish UK GDPR or other regulatory compliance. Organisations remain responsible for access management, retention, lawful processing and other appropriate safeguards. Similarly, an external AIConnect™ provider may receive information according to the integration used; provider-specific arrangements must be assessed separately.

A Modern, More Intuitive User Experience

WinPure v12 also refreshes the desktop experience with a cleaner interface, streamlined navigation, quick actions and easier access to recent projects. These changes are designed to help users return to ongoing work and reach common data quality functions with less friction.

The most significant usability advances come from the new workflows themselves. AdaptiveMatch™ helps users configure sophisticated matching without coding, while MatchGraph™ makes the results easier to interpret visually. The result is a more accessible experience for both experienced data specialists and business users.

One Unified Platform for Data Quality and Entity Resolution

WinPure v12 brings complementary data quality operations into a single environment, reducing the need to move datasets between separate specialist tools.

• CLEAN: Cleanse and standardise inconsistent values and formats.
• PROFILE – Data Quality Insights™: Examine completeness, patterns and quality issues.
• MATCH: Use exact, fuzzy and phonetic matching, enhanced by AdaptiveMatch™.
• MATCH AI: Apply Senzing-powered entity resolution to complex relationships.
• GOLDEN RECORDS – SmartMaster AI™: Select master records and support consolidation.
• AUTOMATION: Configure repeatable data quality workflows.
• AUDIT: Review supported operations and matching outcomes.

A typical workflow starts by profiling incoming data, correcting inconsistencies and validating relevant fields. Teams can then match records, inspect groups and select master records before applying the process again to future datasets. For more background on the broader category, see WinPure’s guide to entity resolution software options.

Who Is WinPure Clean & Match Enterprise v12 Designed For?

WinPure v12 is designed for teams that manage inconsistent or duplicated information and need a practical way to improve it without developing their own matching software.

• Government and public services: Investigate duplicate citizen and service records while retaining infrastructure control.
• Financial services and insurance: Improve customer and organisation data consistency and review entity relationships.
• Healthcare and life sciences: Prepare and reconcile datasets within appropriate privacy and clinical governance processes.
• Retail and CRM teams: Reduce duplicate customer profiles across channels and applications.
• IT and migration teams: Clean and match data before consolidating legacy systems.
• MDM and AI teams: Create more consistent records for analytics, automation and AI initiatives.

The platform is particularly relevant where organisations need advanced matching, no-code operation, visual explainability and flexible Windows deployment in one solution.

Test Your Toughest Dataset on WinPure Clean&Match v12!

Put v12 to work on the dataset that has been hardest to match and consolidate. Book a personalised free trial to see how it handles your data, investigate the connections it finds and judge the results against your own project requirements.

Frequently Asked Questions About WinPure v12

Written by

Team WinPure

The WinPure Team brings together over 100 years of combined expertise in data quality, data management, data matching, entity resolution and software innovation. Our growing team of experienced specialists combines deep industry knowledge with a passion for advancing data quality technology, helping organisations worldwide improve data accuracy, identify and resolve duplicate records, streamline operations and unlock the full potential of their data.

Reviewed by

Farah Kim

Farah Kim is a human centric product marketer who specialises in making complex data management topics accessible to business and technical audiences. With a background in Computer Science, Linguistics, and Media Communications, she bridges the gap between technology and business by translating data quality, entity resolution, data matching, and governance challenges into practical, actionable insights. At WinPure, she works closely with product and customer teams to educate organisations on building trusted, high quality data for analytics, AI, compliance, and operational success.