How Ciena Turned Its Data Catalog Into an AI-Powered Business Resource and Curated 6,000 Assets in Two Weeks

Introduction

Ciena is the global leader in high-speed connectivity,¹ building the world’s most advanced networks to support exponential growth in bandwidth demand.² Ciena provides the infrastructure behind ISPs like Verizon and AT&T that keeps the internet running. In a business this data-intensive, the accuracy and accessibility of data isn't a technical footnote; it's operational.

In 2025, Matt Law joined Ciena's data governance team with responsibility for its data catalog and data quality, a dual mandate – in a role providing clear visibility into the gap between where Ciena's data capabilities stood and where they needed to be.

Having worked with Alation in a previous role, Law arrived knowing the platform well. What he found at Ciena was a chance to put Alation's newer AI-driven curation capabilities to work at a scale his small team could never reach manually, turning the catalog from a place to store documentation into the operational backbone of a governance program.

The challenge

When Law arrived, he perceived three initial challenges – reinforcing each other and making progress on any single front difficult.

• The catalog served the few. It was a technical resource, heavy on database-speak, built for report developers and engineers, and largely invisible to the person in finance or HR. Documentation was written for people who already understood the data, not for those trying to learn it.

• Curation was slow, manual, and hard to sustain. Despite onboarding sessions, documentation workshops, and dedicated training, getting stewards to curate was, in Law's words, like pulling teeth, and even when engagement happened, progress was glacial against the 17 business groups he eventually needed to bring on.

• There was no shared definition of data quality. The organization knew it mattered, especially with its AI ambitions growing, but had no consistent framework. Even the term "critical data element" (CDE) could make eyes glaze over.

The cost of that gap was concrete: among the first things Law was asked to examine were a number of open orders from previous years: records no one was waiting on, but that no process had ever resolved.

Objectives

To turn the catalog from a documentation layer into a resource the whole organization could trust, Law set three priorities, each chosen because it unblocked the next.

• Automate curation at scale to remove the dependency on manual steward effort that had stalled every prior push.

• Transform the catalog into an accessible business resource, legible to non-technical users across every function, not just the technical core.

• Build a scalable, AI-assisted data quality program grounded in clearly identified critical data elements.

Curation at scale was the precondition for everything else: once assets were richly described, that output could feed directly into data quality work, and an accessible catalog gave business teams a reason to engage with both. The question was how to get the first domino to fall.

Implementation

Law's starting problem was raw material: Alation's Curation Automation could generate documentation at a scale no human team could match, but it needed clean, structured, company-specific input to work from.³

First, he built a lightweight prep step using Microsoft Copilot, briefing it on Ciena's environment until it produced a solid first pass on one table. He then codified what worked into a six-to-eight-page protocol covering steward groups, domains, and security classifications. Feeding that protocol into Curation Automation turned rough drafts into complete, governed documentation at scale, with business context applied uniformly across thousands of assets at once. The result: roughly 6,000 assets curated in two weeks, up from a handful of manually curated tables.

Speed only matters if the output can be trusted. Stewards in each domain reviewed a sample of curated assets before rollout, and Law tracked correction patterns to tighten the protocol — since Curation Automation applied it consistently, one fix improved a whole batch rather than requiring row-by-row rework.

"Everybody's looking for an AI win at this stage," Law says. He proceeded by rolling out domain by domain rather than by mandate: supply chain first, the team that "lives and breathes by data," followed by finance, HR, and legal.

He widened what the catalog holds, too, surfacing a machine learning model repository, an index of business processes, and an information asset management section for unstructured data retention through Alation's Doc Hub. 

For Law, the transformation is best captured in a simple analogy: "It's not just the card catalog in the library, it's like the whole reference desk, because you can ask questions of it now." From static index to intelligent guide: at Ciena, Alation has become a resource that meets users where they are, surfaces what they need, and helps even a newcomer navigate with confidence.

Results

The impact arrived fast and ahead of schedule.

A multi-year rollout plan, compressed dramatically. Law had scoped catalog onboarding at 1.5–2 years across 16–17 groups, based on the pace of manual steward engagement. Once Curation Automation had structured input, it curated the same scope — over 6,000 assets — in about two weeks. The comparison isn't apples-to-apples (manual rollout vs. an automated run built on upfront protocol work), but the effect was the same: years of work arrived in weeks, and the roadmap had to be rebuilt.

Adoption tripled, on word of mouth. Users grew from roughly 300–400 to over 1,100, with high-frequency repeat visitors up about 20%, all without a formal internal marketing push.⁴

Every team now gets a head start in AI innovation instead of starting from scratch. Law used to spend the first meeting explaining the basics of CDEs and data quality. Now he arrives with a baseline already assembled — critical data identified, monitors and controls suggested, draft SQL in hand, sliceable by business process, department, or team — and the conversation becomes review and refinement rather than instruction.

He's running the same playbook team by team, giving each domain the same head start on its own AI push rather than building from zero. "I walk into that first meeting with almost an entire solution already built, instead of a blank page," he says.

Conclusion

The momentum shows no sign of slowing. Law's curation protocol is being refined for reports next, with column-level data quality rules on the roadmap after that.

But the development he's most excited about is the compounding effect of curated data feeding directly into data quality work — and the jumpstart it gives even a small team. A full analytical journey that once required months of stakeholder alignment now starts from a baseline Curation Automation already assembled, so teams can review and refine rather than build from scratch.

"It gets us five steps ahead," Law said. "It gets that jumpstart." The catalog that once served a handful of specialists is becoming the connective tissue of Ciena's data organization — and as Law sees it, that's just the beginning.

Matt Law is the Data Catalog and Data Quality Leader at Ciena. Connect with him on LinkedIn.