Ensuring accurate and consistent master data across retail systems
Case study with Rimi

THE CUSTOMER

Rimi Baltic is one of the largest retailers in the Baltic States, operating since 2004. The company manages four store formats: Rimi Hyper, Rimi Super, Rimi Mini and Rimi Express, across more than 300 locations, offering extensive opportunities for growth and development. Rimi Baltic employs over 11,000 people throughout the region and is part of the Salling Group, Denmark’s largest retail organization.

CHALLENGES

Prior to implementing SelectZero, Rimi Baltic faced fragmented and manual data quality management practices. The company lacked a structured and centralized solution for defining and monitoring data quality across multiple data sources and business units. This made it difficult to ensure consistency and accuracy on a regular basis.

Inaccurate master data, particularly product-related, had a direct impact on the supply chain, store operations, e-commerce, finance, and overall customer experience. Incorrect product labeling and tax data also introduced compliance risks, while manual validation processes proved inefficient, time-consuming, and prone to human error.

This in turn affected Rimi’s business operations in numerous ways. The lack of automation meant an inefficient use of resources, with manual validations consuming significant time and requiring dedicated full-time employees. Limited automation capabilities also made it difficult to perform efficient data health checks. Over time, these issues contributed to an erosion of trust in data, resulting in reactive decision-making and negatively affecting regular reporting and analytics.

WHY SELECTZERO?

Since 2023 Rimi Baltic has evolved from piloting SelectZero’s solution to becoming a full customer. Recognizing SelectZero as a promising and adaptable platform, Rimi’s feedback played a key role in refining the solution to better meet the needs of the retail company.

By replacing time-consuming manual checks with automated processes, SelectZero has enabled Rimi Baltic to operate more efficiently, make decisions with greater confidence, and minimize business and compliance risks. Its flexible rule definition, scheduling, and automation capabilities enable the company to maintain consistent and accurate product information, prevent compliance penalties, and reallocate resources to higher-value activities, all while driving continuous improvement in data governance practices.

“SelectZero’s data quality improvement tool is user-friendly and intuitive, helping to streamline our approach to data management. The vendor’s clear communication helped us explore data challenges from new angles, making collaboration smooth and insightful.”

– Andris Vedigs, Innovation Lead, Rimi

Gartner has published its latest Market Guide for Data Observability Tools, and SelectZero is proud to be recognized once again as a Representative Vendor.

We are proud to announce that SelectZero has been recognized as a Representative Vendor in the Gartner Market Guide for Data Observability Tools. This recognition reflects the growing importance of data observability as a core capability for modern organizations, and validates the work our team has put into building a platform that makes data truly reliable.

Why data observability matters

The stakes around data quality have never been higher. Poor data quality costs businesses an estimated $3.1 trillion annually in the US alone, and the average organization spends around 15 hours resolving a single data quality incident. At the same time, AI initiatives, regulatory requirements like BCBS239, DORA, and GDPR, and increasingly data-driven decision-making mean that unreliable data is no longer just a technical headache, it’s a business risk.

Data Observability gives organizations the visibility they need to detect, diagnose, and prevent data issues before they cause damage.

What SelectZero does

SelectZero is a data observability platform built to turn data from liability into reliability. Our platform covers the full spectrum of observability needs: automated data validation with real-time alerts, column-level data profiling with ML-based anomaly detection, end-to-end data lineage, a quality-driven data catalog, and a business glossary that bridges the gap between technical teams and business stakeholders.

We also believe that good tooling should be accessible to the whole data team, not just engineers. SelectZero is built for data and ETL engineers, BI analysts, and business stakeholders alike.

What this recognition means to us

Being included in the Gartner Market Guide sits alongside recognition from customers such as SEB, Citadele Bank, and Telia, who have trusted SelectZero to help them manage and improve data quality at scale.

We see this as motivation to keep building. If your organization is evaluating data observability tools, we’d be glad to demonstrate what SelectZero can do. Book a demo or start a free 14-day trial here, or reach out to our team directly.

SelectZero 2026.3 is here, bringing a range of new features and improvements across the platform. This release focuses on deeper data quality insights, broader connectivity, and a more streamlined user experience. From row-level execution diffs and percentage-based test cases to AWS S3 integration and enhanced admin tooling, there is something for every team working to raise the bar on data quality.

Users can now see the differences between previous and new executions directly within the execution history. Each run is compared against the previous one, clearly showing which invalid rows are newly introduced, which have been resolved, and which remain active. This makes it easier than ever to track data quality progress over time and stay on top of recurring issues.

Note: Test case result CSV row limit has been reduced to a maximum of 10000 rows.

Users can now choose a new test case type that allows adding a separate SQL query for the total row count alongside the usual invalid row count query. When executed, SelectZero automatically calculates the percentage result, making it straightforward to define and evaluate proportion-based data quality rules without complex single-query workarounds. The invalid row count, total row count, and calculated percentage are all displayed in the execution results for full transparency.

It is now possible to see all upstream and downstream dependencies for any object, enabling simpler column-level impact analysis and root cause investigation. Derived from column-level lineage, this gives users a clear picture of what is affected when a data quality issue is discovered, as well as what upstream sources may be responsible for a failing test case.

SelectZero now supports connecting to AWS S3 via Amazon Athena, allowing users to query files stored in S3 buckets directly within the platform. This opens up data quality management for teams working with cloud-based file storage as part of their data infrastructure.

Links stored in custom fields are now rendered as clickable, taking users directly to the relevant external resource. This makes custom fields more useful for teams who reference external documentation, tickets, or tools as part of their workflow.

Admins can now view different log types directly within the platform without needing to download them first. The logs are accessible through the Logs modal, streamlining troubleshooting and reducing the steps needed to inspect platform activity.

The new version of SelectZero is packed with new features and improvements designed to make data quality management faster, smarter, and more flexible. This release introduces Standard validations functionality for quick no-code test creation, a drag-and-drop mapping interface for data reconciliation, and enhanced alerting at the test case level. Teams can now track full change history on test cases, export reports as Excel files, and receive richer webhook notifications with test case results. On the integrations side, Power BI on-premises (PowerBI Report Server) and SAS are now supported, and the Glossary has been updated to handle duplicate terms across different contexts.

Setting up data quality validations just got a lot faster. The new Standard validations functionality lets you pick from a set of predefined validation types, including column completeness, allowed values, numeric ranges, regex checks, date/time checks, duplicate checks, data availability, and data freshness. Simply fill in the required details and you are good to go. No SQL writing required. Standard validations can also be created directly from the Catalog page when opening the Test cases section for any Object.

Creating data reconciliation tests is now more intuitive with a new drag-and-drop mapping interface. Select your source and target objects, then simply drag columns onto each other to define the mapping. WHERE filters can be added as needed, and the SQL is automatically generated from your mapping. Existing SQL is also parsed back into the visual mapping view when reopening a test case.

You can now track all changes made to a test case over time with the new change history feature. It provides a full audit trail of edits, making it easy to see what was changed, when, and by whom. This helps teams maintain accountability and transparency across their data quality test cases.

Alerts can now be configured directly on individual test cases, giving you more granular control over notifications. You can choose to receive alerts based on test case execution results, triggered by a schedule or API, as well as alerts when changes are made to the test case itself.

In addition to the existing HTML reports, test suite reports can now also be generated and sent by email as Excel files. The file includes a summary sheet with all test cases and their statuses, plus individual sheets for each test case with detailed results where available.

Webhook alerts (e.g. Slack, Teams, Jira) can now include a list of test cases and their results. This gives receiving systems and team channels richer context about what passed or failed, without needing to open the full report.

The Glossary now supports duplicate term names across different Glossaries, giving teams the flexibility to manage terms that share a name but carry different meanings in different contexts.

The Power BI connector now supports on-premises Power BI Report Server, in addition to the existing Power BI cloud service. Organizations running Power BI on-premises can now connect and monitor their reports directly, including field-level lineage.

SelectZero now connects to SAS via the IOM connector, allowing to run data quality tests directly against SAS 9.4 and Workspace Server environments. The SAS connector is available alongside the existing suite of database and platform integrations.

The Glossary term detail view has been redesigned with dedicated tabs for Graph, Relations, and Assets, making it easier to explore and manage term context. The legacy view has been unified with the Catalog layout, enabling a side-by-side experience with configurable panel widths that are saved per user for a more personalized workflow.

The platform now keeps a full change history for individual Glossary terms, catalog objects, and attributes. All updates are retained, making it easy to review what changed, when, and by whom, and to revert unwanted changes when needed.

Users can configure automatic notifications for changes to specific Glossary terms or Catalog objects, such as schema drift. Alerts are delivered by email and include direct links to the affected entities, making it easy to review changes as they happen.

Test cases can now evaluate results against a percentage threshold instead of a strict pass or fail. The measured percentage is propagated across the platform and reflected in data quality metrics and dashboards. For example, a validation that checks whether at least 90% of rows meet a rule will display the actual percentage result in dashboard visualizations.

Individual Test cases can now be added directly to Dashboard tiles, making it possible to build custom, validation-based data quality metrics and monitor specific Test cases.

Test case SQL can now be viewed in full screen using a dedicated button in the SQL editor, making complex validations easier to read and edit. In addition, users can define custom row-level result queries by using subqueries in the SQL projection, enabling more flexible and advanced Test case logic (see documentation for details).

New Custom Field types are now available, allowing values to be selected from existing Glossary terms, entered as dates, or captured using larger text areas. This enables richer metadata and more structured documentation across the platform.

Custom fields can now be assigned to Business rules, allowing teams to capture additional metadata and standardize documentation alongside existing rule definitions.

Following the 2007–2008 global financial crisis, it became painfully clear that many of the largest global financial services firms did not have the tools, methodology or data governance principles needed to accurately measure their risk exposures, with often disastrous consequences. Data was siloed, incomplete, and inconsistent, making timely and effective risk management nearly impossible.

In 2013, the Basel Committee on Banking Supervision (BCBS) issued BCBS 239, a set of principles for effective risk data aggregation and reporting. Supervisors have since used these principles as a benchmark and embedded them into local regulatory frameworks, requiring banks to upgrade their governance, data, and reporting capabilities. These principles have been implemented by regulators worldwide and now form a core part of how large banks are assessed on risk data aggregation and reporting.

In May 2024, the ECB published its Guide on effective risk data aggregation and risk reporting, which reinforces and specifies BCBS 239-aligned expectations for banks under its supervision. According to the ECB: “The ability of institutions to effectively manage and aggregate risk-related data represents an essential precondition for sound decision-making and strong risk governance.”

The BCBS has published 14 principles as part of BCBS 239, which are classified into four categories:

• Governance and infrastructure: Requires banks to implement strong governance frameworks and maintain robust IT infrastructure to support risk data.
• Risk data aggregation capabilities: Focuses on the accuracy, completeness, timeliness, and adaptability of risk data.
• Risk reporting practices: Calls for clear, comprehensive, and frequent reporting that supports decision-making at all levels.
• Supervisory review and cooperation: Emphasizes regular regulatory reviews and collaborative oversight mechanisms.

BCBS 239 and Data Observability

If BCBS 239 is the “what” of risk data management then data observability is the “how.” Data observability is the ability to fully understand, monitor, and troubleshoot data as it moves through a system. It gives you insight into the health and behavior of your data pipelines. The capabilities associated with data observability (continuous monitoring of data quality, lineage and reporting) are directly linked to the supervisory expectations articulated in BCBS 239:

• Accuracy and integrity: Automated monitoring, anomaly detection, and validation ensure that risk data remains accurate and reliable throughout ingestion and aggregation, reducing manual errors and supporting reporting needs in both normal and stressed conditions.
• Completeness: Visibility into all data sources, pipelines, and transformations makes it possible to confirm that every material data element, across business lines, entities, asset types, regions, and risk types, is captured, reconciled, and included in aggregated risk views.
• Timeliness: Continuous tracking of data freshness, pipeline delays, and update cycles ensures that aggregated risk data is always up to date and available at the speed required by the bank’s reporting frequency and the volatility of the underlying risk.
• Adaptability: Clear lineage, flexible monitoring, and rapid validation capabilities allow institutions to incorporate new data sources, respond to regulatory or internal reporting changes, and support on-demand supervisory requests.

Ultimately, data observability strengthens the core capabilities that BCBS 239 depends on, spanning everything from data capture to reporting.

SelectZero and BCBS 239

SelectZero is a data observability platform built to maintain data quality, data integrity, and regulatory readiness. The platform is designed to support compliance with BCBS 239 by helping financial institutions meet key requirements for risk data aggregation and reporting.

Data validations

Ensure that data is clean, complete, and trustworthy before it reaches downstream systems. Validations help detect anomalies, missing fields, and formatting issues early in the pipeline, which reduces manual work and strengthens confidence in data.

Automated data catalog and lineage tracking

Gives full visibility into data assets, including definitions, ownership, dependencies, and quality status. Lineage tracking reveals how data flows across systems, showing source inputs, transformation logic, intermediate stages, and downstream usage. This transparency enables faster root cause analysis and impact analysis, helping teams understand both where a data issue originates and which reports, models, or regulatory submissions may be affected. These capabilities are essential for governance, auditing, and meeting regulatory expectations around traceability and accountability.

Centralized business glossary

Helps to unify the definitions of key terms, metrics, and business rules across the organization. This reduces ambiguity in risk reports and ensures that teams use consistent terminology when interpreting or communicating data.

These capabilities are closely aligned with principles articulated within BCBS 239, in particular risk data aggregation and governance. Furthermore, SelectZero’s observability model offers financial institutions a way to get beyond checkbox compliance to actual, operational resilience. By bringing observability to their data culture, organizations can significantly improve their ability to meet regulatory expectations.

The cost of non-compliance

Non-compliance with BCBS 239 isn’t just a regulatory headache, as it can have severe business consequences. Not only does poor risk data aggregation block decision-making, slow response times in crises, and undermine stakeholder confidence, but organisations that fall short may now also face regulatory consequences like enforcement actions, capital add-ons, public disapproval, and fines. As one example, global investment bank Citigroup, was fined $400 million in October 2020, and again $136 million in 2024, by US banking regulators for failing to address “long‑standing deficiencies” in enterprise-wide risk management, data governance, internal controls, and compliance risk management – all issues closely related to BCBS 239.

Turning compliance into competitive advantage

Regulatory compliance is often viewed as a burden, but BCBS 239 presents an opportunity for forward-looking financial institutions. By adopting data observability as a foundational discipline, organizations can:

• Respond faster to market stress
• Identify emerging risks earlier
• Support strategic decision-making with reliable data
• Improve auditability and transparency

As the ECB and other supervisory bodies increase their scrutiny, those who invest in observability now will be better equipped not only to comply, but to thrive.

How SelectZero is used for regulatory reporting

Banks and financial institutions in SelectZero’s portfolio are currently working on improving their compliance with BCBS 239 using the platform. One key factor for this is the previously mentioned data lineage tracking, which enables our clients to trace every transformation and movement of risk data with clarity. This supports faster root cause analysis, smoother regulatory reviews, and more reliable end-to-end reporting. The second key factor is automated data validations, which help teams detect issues early in the process, reducing manual workload and strengthening confidence in the reports they submit to supervisors. As a result, institutions using SelectZero are not only meeting regulatory expectations with greater ease, but are also building more efficient and resilient reporting workflows that scale as their data landscape grows.

SelectZero now allows users to create fully customizable data quality dashboards with flexible layouts, visualizations, and content. Dashboards can include Test Suites, Reports, Business rules, or individual Data Objects, and are role-based to ensure relevant views for each team or business domain. Users can set favourites for quick access and explore created dashboards they have permission to view.

Note: Dashboard prior to 2025.10 has been deprecated.

Reports now support filtering by Test Case custom fields, enabling faster and easier analysis of data quality results. This enhancement allows users to create custom data quality metrics, such as Availability, Uniqueness, or Accuracy. Combined with Personalised Dashboards, teams can now easily showcase targeted data quality insights to stakeholders.

SelectZero’s automated lineage detection is now smarter and more comprehensive. The new Smart Lineage option accounts for temporary tables created and dropped within SQL scripts, ensuring more accurate lineage extraction. Lineage can now also be detected directly from procedures, with discovered relations automatically added to the platform. In addition, users can view the exact SQL statement behind each relation, including the precise source context within the script.

The scheduling engine now supports CRON-based scheduling, providing greater flexibility and precision in automating data quality processes. Users can configure Test Cases, Test Suites, and Profiling tasks to run at highly specific times and intervals, enabling more dynamic and fine-tuned execution schedules.

Custom fields now support a Required attribute, ensuring key information is always provided when creating or editing entities such as Test Cases, Catalog Attributes, or Terms. Additionally, default values can be defined to prefill fields during entity creation, streamlining data entry and improving consistency across the platform.

Catalog Datasets now support custom fields, enabling richer and more flexible metadata management. This allows users to define additional attributes and classifications, creating a more comprehensive data catalog by organizing and describing physical data assets within customized datasets.

The Catalog Object view has been redesigned for greater clarity and efficiency. Dedicated tabs for Attributes, Lineage, Test Cases, and Profilings make it easier to explore, manage, and monitor each object’s metadata and quality in one structured view.

You can now connect Tableau and Power BI (Power BI Service) as datasources, enabling access to report metadata and automatically mapping lineage from reports to their underlying datasources.

It is now possible to filter test cases by a specific date. This makes it easier to review and analyze results as they appeared at a chosen point in time.

Test suite reports are now stored and remain accessible for historical results. Reports generated for previous executions can still be opened and reviewed directly from the UI.

The glossary now remembers opened hierarchies and allows setting all terms to always stay open. Editing a term no longer refreshes the entire list, making navigation smoother.

You now have more control over how glossary graphs are displayed with persistent filters, depth limits, compact view, and directional layouts. These options make it easier to navigate and analyze even the most complex business term relationships.

• All graph filters are now persistent for individual users
• Depth – You can set a maximum level of graph nodes to display
• Compact view – With this enabled, every term is only expanded once. Disabling this will expand all terms in all branches.
• Directional – With this enabled, all term relations flow from left to right (instead of root term fixed on the left)

You can now manage all lineage relations between catalog objects in a single view. This includes visibility into their origin and creation dates, making large-scale lineage management more efficient.

It is now even more simple to automate catalog import through API. All catalog importable objects can now be added through SelectZero API.

In this short video we will walk through how to create data reconciliation checks between multiple databases. With SelectZero’s data quality platform, you can define and run data comparisons to detect discrepancies like missing records or mismatched totals that can often go unnoticed, leading to faulty reports, compliance risks, and poor decision-making.