This guide is aimed at intermediate users of the Actian Data Intelligence Platform. It’s not designed for beginners or seasoned metadata gurus.
As an intermediate user, you already understand basic data governance and data catalogs, but you want to level up by using the Actian platform. In this guide, we’ll walk through the architecture and core concepts, deployment and setup, day-to-day workflows, advanced uses, and best practices. You’ll be able to optimize the platform to discover your data, trust its quality, and activate data assets.
Platform Overview and Core Concepts
First, let’s clarify what the Actian Data Intelligence Platform is, how it differs from other tools, and the key concepts you’ll be working with.
What is the Platform?
Actian Data Intelligence Platform is a cloud-native, enterprise-scale solution for metadata management, data cataloging, data lineage, data governance, and a data marketplace. It’s powered by a federated knowledge graph and uses a Model Context Protocol (MCP) Server to map relationships across data assets, simplify data discovery, and give context to data based on the state of your business.
How is it Different From a Basic Data Catalog?
Unlike traditional data catalogs, the Actian platform:
- Combines metadata, data quality and observability, governance, business context, and discovery in one solution.
- Uses a knowledge-graph foundation, linking technical metadata such as tables and columns with business context, including glossary terms and KPIs, and mapping relationships between data.
- Supports a data product model, where datasets are published into a data marketplace environment, with contracts, policies, access requests, and other details in place.
- Deploys across hybrid or multi-cloud environments, and can scale across domains and teams.
Terminology to Know
Here are terms you’ll encounter:
- Metadata management. Capturing and managing technical and business metadata from your data ecosystem, including data sources, reports, dashboards, and BI tools.
- Data catalog and data discovery. Enabling users from both technical and business teams to search, explore, and understand datasets.
- Knowledge graph and semantic layer. Representing relationships between data assets, domains, business terms, and lineage. This underpins discovery and context.
- Data governance. Policies, roles, access controls, contracts, and compliance mechanisms that ensure trustworthy data usage.
- Data products and marketplace. Published, governed data assets, with documentation, contracts, and discoverability by business users.
- Lineage and observability. Tracing how data flows from ingestion through transformations to consumption, monitoring data quality and integrity along the journey.
Deployment and Setup Best Practices
Before you dive into using the platform on a daily basis, you need a sound foundation. Here are intermediate-level deployment and configuration considerations:
4 Steps for Architecture and Environmental Planning
- Choose your deployment. Decide whether you’ll use the SaaS/cloud version or an on-premises/hybrid deployment. The platform supports hybrid and multi-cloud environments.
- Map your data ecosystem. Determine what sources exist, such as databases, data lakes, BI tools, and SaaS apps. Also, decide how domains are defined, whether by business unit, region, or capability, and who is responsible for tasks related to data.
- Plan for connectivity. Make sure the platform can connect to your sources for metadata ingestion, scanning, and lineage.
- Define access and security models. These include user roles, separation of duties, federated vs. centralized governance, and domain autonomy vs. enterprise alignment.
7 Steps for Setup and Onboarding
- Install/configure the platform. Configure it for cloud or on-prem environments, set up authentication (LDAP/AD/SAML), and define initial admin users.
- Define domains and structure. Establish data domains. For example, this can be sales, finance, or operations, and then create domain catalog partitions or sub-catalogs as needed. This supports decentralized data ownership.
- Connect your metadata sources. Set up the built-in scanners/API integrations to ingest metadata from your databases, data warehouses, BI tools, files, data lakes, and other sources. The platform will auto-catalog them.
- Define the business glossary and taxonomy. Engage business and domain owners to define key terms, definitions, metrics, and KPIs for your organization. These will be used by the knowledge graph and assist with discovery and understanding.
- Document data products and contracts. Begin publishing high-value datasets as data products. Define metadata, policies, access controls, and contracts. This helps operationalize self-service data discovery for your teams.
- Enable data lineage and observability. Verify that lineage capture works and integrate your data quality and observability pipelines so users can trust your data.
- Onboard users and provide training. Train data stewards, engineers, and business users on how to optimize the platform.
Day-to-Day Workflows
Now we move into how you and your team will use the platform operationally, integrating it into your workflows.
Data Discovery
For business users and data analysts, these data discovery process capabilities are key:
- Search using natural language or keywords across datasets, reports, dashboards, and glossary definitions. Knowledge graph technology provides semantic context.
- Browse information across your organization, just like a data marketplace.
- Evaluate dataset trustworthiness. Look at lineage, quality indicators, ownership, or business glossary context before using the data. This builds confidence and reduces rework.
- Submit access requests for datasets through the data marketplace or a self-service mechanism, streamlining consumption while preserving governance.
Data Stewards and Engineers
The “producer” side is where you, as an intermediate user, will spend time maintaining and enriching the metadata, ensuring governance, and publishing data products. Here are some tasks you’ll perform:
- Use the platform’s scanning tools to auto-ingest metadata, ensure robust data lineage, and classify data.
- Enrich metadata manually at will. Add business terms, definitions, domain tags, sensitivity classifications, owners, and usage notes.
- Define data contracts and service level agreements (SLAs) for data products. The platform supports contract-first approaches.
- Manage governance workflows, including approvals for publishing, dataset access, and metadata changes. This helps enable and maintain audit trails.
- Monitor lineage. Visualize the flow of data from source to report, understand dependencies and impact of changes, and identify potential risky transformations or bottlenecks.
Data Products
One of the platform’s richer features is turning datasets into first-class data products that are published to a marketplace within your organization. This is where the business and technical worlds meet.
- You or your data teams identify high-value datasets, like a monthly sales summary, customer segmentation, or cost-center budget data, and publish them as data products.
- For each product, define:
-
- Metadata. The description, domain, owner, and steward.
- Policies. Who can access the data product, how often it’s refreshed, and quality metrics.
- Documentation. The business context, typical uses, and known caveats.
- Place the product in the marketplace where business users can easily discover it via search, browsing, or recommendation.
- On the consumption side, users find the product, evaluate its trustworthiness, request access, and start using it.
- Monitor product usage. Track how many users have accessed it, downstream assets built from it, receive feedback, and view the evolution of the product.
Actian Data Intelligence Platform’s Advanced Capabilities
Once you’re comfortable with the standard workflows, you can leverage the deeper, value-adding features of the platform.
Knowledge Graph and Semantic Search
The platform’s knowledge graph links datasets, columns, business terms, owners, domains, transformations, and reports. This, along with the MCP Server, gives you semantic context and richer discovery.
For example, when a business user searches for “customer churn rate,” the platform not only returns datasets named “customer_churn” but also related terms like “attrition” and “customer_retention.” It also shows lineage, domain, and business glossary definitions.
As an intermediate user, you can refine the graph by linking glossary terms to datasets, tagging domain relationships, and building semantic relationships. You can use the graph for impact analysis, and if you modify a dataset or drop a column, the platform will identify all downstream dependencies via graph traversal.
AI Integration and Agentic AI Use Cases
The platform offers a MCP Server, which enables governed, high-quality data to be served into AI assistants built on LLMs.
This means that business users or analysts can ask an AI assistant a natural-language question, such as, “What was our net revenue in Europe last quarter?” The assistant uses the knowledge graph, MCP Server, metadata, and data contracts to generate answers or SQL statements that adhere to governance policies.
Intermediate users should monitor the performance and accuracy of these AI interactions, measure user adoption, and ensure feedback loops to improve metadata alignment.
Data Observability
One of the significant developments in the platform is the introduction of Actian Data Observability. This solution integrates with the data intelligence platform to provide real-time monitoring of data quality and anomalies.
As an intermediate user, you can set up key quality metrics for critical data products, connect them to observability dashboards, and define alerting and workflows for when quality thresholds are breached. Use this to enforce governance by design. With this approach, data products in the data marketplace have built-in quality thresholds and mechanisms for remediation.
Link quality observations back to metadata. If there’s an anomaly in a dataset, ensure it’s recorded in the metadata or case management so users understand past issues.
Best Practices for Intermediate Users
These practical best practices ensure your adoption of the platform delivers long-term, sustainable value:
- Start with domains and high-value datasets. Avoid trying to catalogue everything at once. Start with focused areas and expand.
- Understand that metadata is not “set and forget.” Schedule periodic reviews of owners, stewards, definitions, and dataset usage, and retire stale assets.
- Embed governance into the output, not as a separate process. Publish data products with contracts, policies, and access controls, and automate enforcement where possible.
- Treat business users as first-class citizens. Ensure the user interface is usable, the business glossary is populated, search is intuitive, and business users have the access they need.
- Monitor usage and metrics. Track the most used and least used products, user satisfaction, and search effectiveness. Use this to steer metadata enhancement and retire unused assets.
- Train and communicate. Metadata culture is as much about people as it is technology. Supply training, internal marketing, office hours for data stewards, and feedback loops.
- Ensure scalability. As you add domains, sources, and users, ensure your architecture can scale without performance bottlenecks.
- Plan for change. When underlying data sources or business models change, you’ll need notifications and workflows to update metadata and lineage, as well as to communicate with data consumers.
- Document workflows and responsibilities. Define who owns what, such as the domain owner, steward, metadata curator, and technical connector owner. Clarify roles and responsibilities so nothing gets lost or overlooked.
- Enact security, compliance, and sensitivity identification. Ensure data products classify PII and other sensitive data, enforce access control, and maintain audit trails. The platform supports this.
Implement the Actian Data Intelligence Platform
Actian Data Intelligence Platform is a powerful enterprise metadata, catalog, governance, and marketplace solution built on a knowledge graph, designed to serve both technical and business users. As an intermediate user, you should focus not just on getting the platform up and running, but on embedding it into your organization’s processes.
These processes include defining domains and data products, enriching metadata, and enabling self-service discovery. The platform also offers governance by design, advanced capabilities like lineage and data observability, and AI-ready data. You can use the platform to democratize data access, establish trust, and activate AI.
Ready to get started? Request a live demo today to see how the Actian Data Intelligence Platform transforms workflows and simplifies data management.
FAQ
An enterprise data marketplace is a governed platform that connects data producers and consumers, enabling publication, discovery, licensing, and secure delivery of curated data products, with APIs and connectors to integrate data into analytics and AI workflows.
Review the provider’s quality score, request sample datasets, and verify schema documentation. Many platforms offer trials to assess quality.
Prioritize governance and automated policy enforcement, data quality and lineage verification, intuitive search, flexible licensing and billing, and seamless analytics/AI integrations, plus cloud-native scalability and regulatory compliance automation.
Common challenges include balancing privacy and access, integrating legacy systems, scaling infrastructure, achieving adoption, and managing licensing complexity; these are addressed via change management, training, and phased pilots.
Public marketplaces have open listings for any qualified buyer, while private exchanges restrict access to invited participants and often offer custom agreements.
Data contracts embed schema definitions and quality thresholds that CI/CD pipelines validate on every release, ensuring compliance.
Connect the marketplace’s API to your pipeline, pull schema definitions into version control, and configure automated tests for data quality.
Verify the provider’s data contract for consent documentation and audit logging. Review compliance certifications and data processing agreements.
They combine provider vetting, automated profiling and anomaly detection, lineage tracking, SLAs, and continuous compliance monitoring with detailed audit trails and integration to security and governance systems.
