Dev48
Language
  • About
  • Services
  • Industries
  • Technologies
  • Articles
  • Contacts
Book a call
    Home/Articles/Governance by design write it check it approve it
Dev48
info@dev48.org+375 29 320-40-96WhatsApp@isky_by

© 2026 · All rights reserved.

Governance by design: write it, check it, approve it

Source: Collibra•November 5, 2026 at 01:02 PM

Governance by design: write it, check it, approve it

In February 2025, Gartner predicted that "through 2026, organizations will abandon 60% of AI projects unsupported by AI-ready data." 2026 is nearly over, and most catalogs still decide "ready" the old way. A steward writes a description when there's time. Someone reviews it next quarter. The…

In February 2025, Gartner predicted that "through 2026, organizations will abandon 60% of AI projects unsupported by AI-ready data." 2026 is nearly over, and most catalogs still decide "ready" the old way. A steward writes a description when there's time. Someone reviews it next quarter. The asset gets published once enough people say it looks fine. Collibra’s governance by design toolkit puts the writing, the checking and the approval on the asset page while the work is happening.

What’s new: Governance by design toolkit

Governance by design brings together three capabilities that are now generally available: the AI Writing Assistant, Smart Checks and Lifecycle Management, so stewards can write, check and approve an asset in one place.

  • The AI Writing Assistant drafts and edits text attributes inside the editor, following a prompt you configure once. Stewards start from a draft, not a blank field.
  • Smart Checks are no-code rules that evaluate an asset every time it changes and show pass or fail next to the field they cover. Gaps show up while the steward is still editing, not weeks later in review.
  • Lifecycle Management defines the stages an asset goes through and lists what each one needs: assessments, sign-offs and suggested Smart Checks. Everyone can see what “ready” requires and who signs off.

How the governance by design toolkit helps

A catalog rarely decays because stewards stop caring. It decays because the work happens in the wrong order. The description starts from a blank field. The standard lives on a wiki page nobody has opened. The review comes weeks later in a bulk pass, when the steward has moved on and the reviewer has to guess what they meant. Meanwhile each asset type has developed its own approval habit. Now, all three steps are on one page in the order the work actually happens.

Problems it solves

  • Empty descriptions: Fields stay blank because drafting from scratch takes time nobody has. The Writing Assistant turns drafting into reviewing.
  • Ten stewards, ten styles: Definitions vary in length, tone and structure depending on who wrote them. One configured prompt sets all three for everyone.
  • Finding gaps too late Stewards learn a field misses the bar after they've moved on. Smart Checks show the gap at the point of edit.
  • Reviews that never happen: Periodic spot-checks are the first thing dropped under deadline pressure. Checks run automatically.
  • "Ready" by consensus: Without a defined path, publication depends on informal agreement, usually living in a spreadsheet nobody remembers, and each asset type has its own. Lifecycle Management names the stages and the people who sign off on a convenient visual tracker.

How the governance by design toolkit works

As a data steward.

  • The lifecycle tracker in the sidebar shows what stands between this stage and the next. You know what to do. The hard gates are the assessment and the sign-off, and the Smart Checks put every gap in plain sight before anyone signs.

Track an asset's expected journey from creation to publication

  • When you open the first attribute in edit mode, the AI Writing Assistant button appears. One click gives you a draft built from what the asset already knows, shaped by your admin’s prompt. Fix a sentence yourself, use Summarize, Make shorter, Make longer or Change tone. Accept and save. That’s where the checks come in.y.

Never face the blank page paralysis again. Quickly draft content for text fields, edit and save.

  • Each check tests a single attribute or relation against the bar your admin set: a description long enough to say something, a link to the data contract, a next review date that’s still in the future. A visual cue next to the field shows pass or fail as soon as it changes. Checks suggest what’s needed but don’t block progress.

Smart checks will help you quickly identify what’s missing before you publish new assets in the catalog.

Once the data owner signs off, the asset moves to Approved. No blank page, no hunting for the standard, no chasing approval by email. The steward’s time goes into reviewing, not drafting and following up.

As the governance manager, you set this up once.

  • Switch on the Writing Assistant for the characteristics you choose. Start from a template prompt, set audience, length, structure, tone and language, and test it on real assets before it reaches stewards.
  • Build Smart Checks without code, one attribute or relation per check.
  • Then map the lifecycle stages for active assets, a retirement phase for archived or rejected ones and the activities each stage needs.

Why you should be excited

By embedding intelligent automation directly into stewardship workflows, these features transform data governance from a manual challenge into a guided, high-efficiency experience for every member of your team.

  • Data stewards: Draft definitions are suggested, the gaps show up next to the field and the tracker shows the next step. Less starting from scratch, less back-and-forth after review.
  • Governance managers: Define your standards once (including smart check rules, lifecycle stages and AI prompt instructions) and enforce them automatically. Because every stage transition is backed by completed activities, the governance process becomes traceable, verifiable and completely audit-ready without requiring manual oversight.
  • Data consumers: "Published" now means something specific: the asset went through defined stages and a named person signed off before it reached you.

Use cases

  • Publishing a data product: A domain team registers a new data product. The Writing Assistant drafts the description, checks flag a thin description or a missing data contract link, and the data owner signs off before it moves to Approved. Consumers find it documented and linked to its contract from the start.
  • Reviewing a proposed AI use case: A team proposes a new AI use case. The Writing Assistant turns its technical details into a plain-language purpose and audience description. A check flags a missing risk classification, and the designated AI owner signs off before the use case moves to Approved. Accountability gets settled at the proposal stage, not in a review after launch.
  • Taking a CDE from Candidate to Approved: Build a lifecycle for critical data elements. Checks flag a missing risk classification or a missing link to the governing regulation, and the data owner signs off before the CDE is approved. Status and sign-offs live on the asset, not in a spreadsheet.

Key takeaways

Governance used to happen after the work: draft now, review later, publish when people agree. Now, the standard is on the page while the steward is writing, and the approval path is defined before the asset needs it. The steward's job shifts from drafting and chasing to reviewing and fixing.

  • Write: The Writing Assistant drafts text attributes from the asset's own context, following a prompt your team configures once.
  • Check: Smart Checks show pass or fail next to the field every time it changes, so stewards fix gaps before review.
  • Approve: Lifecycle Management moves the asset through defined stages, with named sign-offs and assessments before publication.

Where to learn more

  • Blog: The new standard for metadata documentation: Meet your new AI writing assistant
  • Maciej RusekMaciej RusekProduct ManagerCollibra

Maciej Rusek

Maciej Rusek

Product Manager

Collibra

  • Volodymyr IliushchenkoVolodymyr IliushchenkoSoftware Engineer Collibra

Volodymyr Iliushchenko

Volodymyr Iliushchenko

Software Engineer

Collibra

  • Marc CampabadalMarc CampabadalSenior Product Marketing ManagerCollibra

Marc Campabadal

Marc Campabadal

Senior Product Marketing Manager

Collibra

In this post:

  • What’s new: Governance by design toolkit
  • How the governance by design toolkit helps
  • How the governance by design toolkit works
  • Why you should be excited
  • Key takeaways

Keep up with the latest from Collibra

I would like to get updates about the latest Collibra content, events and more.

Business e-mail address

There has been an error, please try again

Please enter a valid business e-mail address

By submitting this form, I acknowledge that I may be contacted directly about my interest in Collibra's products and services. Please read Collibra's Privacy Policy.

Thanks for signing up

You'll begin receiving educational materials and invitations to network with our community soon.

← All articles

What this article says

More in Data & Analytics

All →
The Economics of Enterprise AI: How Rising Costs Are Changing AI Strategies
Cloudera

The Economics of Enterprise AI: How Rising Costs Are Changing AI Strategies

Introducing the joint reference architecture for enterprise-grade Redis on VMware Cloud Foundation 9.1
Redis

Introducing the joint reference architecture for enterprise-grade Redis on VMware Cloud Foundation 9.1

Migrate Splunk queries, alerts, and dashboards to Datadog
Datadog

Migrate Splunk queries, alerts, and dashboards to Datadog

Which document-parsing benchmarks can you trust?
Datalab

Which document-parsing benchmarks can you trust?

The next chapter of Redis Iris: what agents need to work in the real world
Redis

The next chapter of Redis Iris: what agents need to work in the real world

Agents Are Software: Introducing the Next Chapter of Redis Iris
Redis

Agents Are Software: Introducing the Next Chapter of Redis Iris

More from Collibra

Collibra Control Tower is here. Your data policies just got a schedule.
Collibra

Collibra Control Tower is here. Your data policies just got a schedule.

Bring SAP S/4HANA metadata into Collibra Platform
Collibra

Bring SAP S/4HANA metadata into Collibra Platform

Bring your own model (BYOM): Collibra's AI, plus your model
Collibra

Bring your own model (BYOM): Collibra's AI, plus your model

Collibra + Trail ML: Bringing AI governance and automation together
Collibra

Collibra + Trail ML: Bringing AI governance and automation together

Context is the new IP: Collibra is where enterprises come to protect it
Collibra

Context is the new IP: Collibra is where enterprises come to protect it

Evolving from data governance to AI decision governance
Collibra

Evolving from data governance to AI decision governance