Unclear Ownership
Nobody is accountable for a data asset, so quality issues go unresolved and definitions drift apart across teams.
Metadata Repository & Governance Platform
Turn Scattered Data Into Governed, AI-Ready Assets
Trusted Data. Intelligent Decisions.
KRISID is a metadata repository that catalogs your estate, records who answers for what and traces lineage from a report back to the column it came from — running inside your own cloud. Its model is a workbook rather than code, so it speaks your vocabulary from the first week. We build it, and we implement it.
The problem
Most Governance Programmes Fail on Adoption, Not Technology
Organisations invest in catalogs, quality tools and workflow platforms — and months later the business is still working from spreadsheets, email threads and tribal knowledge. Governance was implemented for the business instead of with it.
Data Nobody Can Find
Without a curated catalog and lineage, analysts rebuild the same logic repeatedly and leaders lose confidence in reporting.
AI on Ungoverned Data
AI and agentic initiatives inherit unreliable data and undefined authority — creating risk faster than they create value.
Our product
KRISID — A Metadata Repository Whose Model Is Data
Most catalogs decide in advance what kinds of thing may exist, so the asset your business actually argues about waits for a vendor release. In KRISID the model is rows in a workbook. Add an asset type and every screen follows — with no migration and no code.
Your Vocabulary, Not Ours
Rename a table to a dataset, add a reinsurance treaty or a model card, and the sidebar, the lists, the estate tree, search and every saved view follow from the same upload.
Lineage, Owners and Approvals
The same graph read at column, object and business grain. Responsibility that inherits down the estate. Changes to governed types routed through approvals you declare on a canvas.
Inside Your Own Estate
Your cloud, your Postgres, your network. Metadata about regulated data is often regulated itself, and no extension your platform team has to approve is required.
Any size
Fits the Estate You Have, Not the One a Vendor Assumed
One install, one database, and the same product either way. What changes with the size of your organisation is how much of it you use — not whether you can use it at all.
A Smaller Organisation
Ten thousand catalogued assets is about eighty-five megabytes — a repository that fits on the smallest managed database your cloud sells. There is no cluster to stand up and no Postgres extension for a platform team to approve, which is usually the longest wait in a catalog project.
- Start from the model we ship, rename things to your own words in a spreadsheet
- Connect a source and preview exactly what the first sync would create
- One steward can run it; the licence names the installation, not the number of people reading it
A Large, Federated Estate
Measured at ten million assets rather than estimated: the front page in 1.8 seconds, a filtered search in half a second. Graph walks are bounded by breadth rather than depth, so a diamond is visited once and a cycle terminates instead of hanging.
- Many domains, each deciding for what it owns, with responsibility that inherits
- Approvals declared per asset type, with four eyes where it matters
- Nightly harvests that write nothing on the nights nothing changed
Why the size question usually bites
Fixed-model catalogs make small organisations pay for shapes they do not need and force large ones to bend their business into shapes that do not fit. Because the KRISID metamodel is rows rather than code, a five-person data team and a five-hundred-person one are running the same software with a different number of rows in it.
Pricing follows the same principle. The licence covers the installation, not the seats — so cataloguing more of your estate never costs more, and the people who most need to read the catalog are never the ones priced out of reading it. Governance has been sold as though only the largest organisations deserve it; that is the thing we set out to change.
What we also do
Governance Services, End to End
Twenty years of delivery behind the product — and available on its own, including on the platform you have already bought.
Data Governance Strategy
Operating model, roles, policies, glossary and quality KPIs — plus the adoption, literacy and training programme that makes it stick.
Collibra Implementation
Domains, catalog onboarding, lineage, policies and privacy, Data Marketplace and Edge — plus managed support and upgrades.
Informatica CDGC
Scanners, curation model, glossary, lineage and stewardship — with source integration and automated onboarding.
Microsoft Purview
Data map, scanners, classifications, sensitivity labels, lineage and insights — integrated across your Azure estate.
Atlan
Active metadata, personas and purposes, column-level lineage and governance workflows embedded where teams already work.
Alation
Discovery and stewardship driven by real usage — query log analysis, trust flags, glossary and policy management.
Custom Workflow Development
BPMN-driven Collibra workflows for approvals, stewardship tasks, access requests and issue management — auditable by design.
Custom Integration
Secure connectors and APIs that synchronise metadata, lineage and policies for near-real-time, consistent catalogs.
Industries
Where We Work
Governance obligations differ by sector, and so does the vocabulary — which is exactly what a workbook-driven model absorbs. We work where data is regulated, high-volume, or both, at any size of organisation.
BFSI
Risk data aggregation, regulatory reporting lineage, and the evidence supervisors ask for.
Technology
Product telemetry at scale, domain-owned data products, and governance for AI features.
Telecom
Subscriber data, network analytics, and consent management across very large volumes.
Media
Audience data, rights and licensing metadata, and privacy across syndication chains.
Healthcare
Patient data protection, consent and purpose limitation, and clinical data quality.
Where governance is heading
AI Governance Now Needs an Agent Registry
Enterprises are no longer deploying AI that only generates answers. They are introducing agents that discover information, call tools, access systems and execute business actions.
An agent catalog has to be more than an inventory. For every agent you need to know who owns it, what business purpose it serves, which systems it can reach, what it is allowed to execute, where human approval is required, and how its behaviour is audited.
An Agent That Assists ≠ an Agent That Executes
The two carry very different risk profiles, and your governance model should say so explicitly. We help you define authority tiers, approval gates and audit trails before agents touch a system of record.
- Agent identity and ownership
- Tool and data access boundaries
- Human-in-the-loop approval gates
- Behaviour monitoring and audit
- Model and prompt lineage
How we work
Assess, Prioritise, Implement, Embed
A delivery sequence designed to show value in weeks — not after an eighteen-month programme.
-
Assess
We map your current state across people, process, policy and platform, and identify where accountability is genuinely missing rather than merely undocumented.
-
Prioritise
Quick wins first. We sequence domains and use cases so business teams feel the benefit before the programme asks anything of them.
-
Implement
Configuration, metadata harvesting, lineage, workflows and integrations on Collibra, Purview, CDGC, Atlan or Alation — engineered, not clicked together.
-
Embed
Enablement, data literacy, stewardship operating rhythms and SLA-driven support so governance survives after the consultants leave.
Insights
White Papers From Delivery Work
Written from what we see repeatedly in governance programmes. Free to read, no registration.
Why a Catalog Should Not Decide What an Asset Is
Most catalogs ship with a fixed list of the things that may exist. What that costs — and five questions to put to a vendor.
AI Governance Frameworks in the Age of Regulation
The EU AI Act's high-risk obligations are enforceable now. Why the organisations coping best extended the data governance they already had.
A Data Governance Framework That Survives Contact With an Organisation
Five layers built around accountability, enforcement and evidence — with the diagram, and which layer fails first.
Good Data Doesn't Happen by Accident
It happens through accountability. Tell us where your governance programme is stuck and we'll tell you honestly what it would take.
Start a Conversation