Reading Time: 6 minute

You are in a room where time is important.

The screen is filled with numbers and maps. People call out things. Some systems get extra attention because they are the ones you can’t afford to go wrong with.

Every decision relies on one assumption: the information in front of you is trustworthy.

That’s what Artemis II aims to test—a crewed mission launching on April 1, 2026, designed to ensure that the spacecraft’s systems operate as expected in space before NASA moves on to the next phase.

Federal agencies face a similar moment, only without the rockets.

Across government, data now determines how benefits are delivered, how funding is allocated, how risks are assessed, and how AI tools are used. Leaders know the data exists – the question is whether they can take action based on it. Which number is official? Who is responsible for this? And when there is a change, how quickly do other people know about it?

These are practical questions. And answering those questions has pushed data governance into everyday mission work.

What makes governance feel urgent right now

The application of AI in government is growing rapidly, raising the stakes for data quality, traceability and access. In a July 2025 report, GAO found that reported AI use cases at select institutions nearly doubled from 571 in 2023 to 1,110 in 2024, and generative AI use cases jumped about ninefold (32 to 282).

At the same time, expectations for how institutions manage and publish data assets continue to rise. OMB’s Phase 2 Evidence Act Guidance (M-25-05) strengthens requirements for data inventories, metadata, and management practices that support evidence access and development while maintaining appropriate safeguards.

Simply put: more usability, more visibility, and less tolerance for uncertainty.

Flight plan: four governance steps that help institutions stay on track

If governance is to accelerate the mission, it must be visible where the work is done – in decisions about access, shared definitions, accountability, and safeguards that are left behind. These four moves are made for that.

Step 1: Build a data council that makes decisions

Many institutions already have data boards. The difference between a helpful board and a slow board usually comes down to two things: goals and authority.

When a board is formed primarily to review documentation and share the latest information, the same pattern emerges: meetings happen, but decisions don’t; each office has its own priorities; and no one knows what actually changed next.

Boards that move forward are based on clear goals at the institutional level and are led in a way that benefits the entire institution. It brings together mission owners, IT, security, and data owners, and treats data and AI as connected work.

More importantly, it makes decisions that eliminate confusion: which data sets are considered authoritative; who owns it (and what that ownership actually means); what the access is like; and what to prioritize first.

When those decisions are made early, teams stop wasting time arguing about which numbers are real and start working from the same map.

Step 2: Treat inventory as a start — then make the data usable

Supplies required. This is where a lot of momentum dies.

Agencies often complete inventory to meet a requirement, then struggle to convert it into something the mission team can use. The common details are predictable: poor prioritization, limited usability, and a lack of clear ownership when it comes to keeping information up to date.

Catalogs are valuable when they are created for non-specialists, not just data teams. It means:

• A simple language description that explains what the data is and why it exists
• Remove ownership so people know who to contact and who updated it
• An access flag that prevents sensitive data from being distributed carelessly, but can still be found by the people who really need it

Usability also depends on traceability. When teams understand where data comes from, trust in dashboards and AI output increases because accountability is no longer just implied.

Step 3: Focus data quality on what matters most

Data quality is where good intentions disappear if the goal is to improve everything.

A more realistic approach is to focus on high-impact data first – data sets related to outcomes that institutions can’t afford to be wrong about, such as grants, benefits, and eligibility decisions. This focus keeps governance practical and links quality efforts directly to mission outcomes.

This is not a theoretical problem. The 2025 IBM Institute for Business Value survey found that data quality remains a top operational priority, and many organizations expect significant financial losses associated with poor data.

When teams trust the data they use, the impact is immediate. Decisions move more quickly because input is not constantly questioned. Duplicate reporting and manual reconciliation are eliminated because definitions and sources are consistent. Audits become easier to support because the underlying data is easier to explain. And AI adoption feels less risky because teams aren’t providing outdated or unclear input to models.

Step 4: Build security into the process from the start

In a federal environment, it is no longer realistic to separate government from security. Governance determines who can access data, how data is shared, and how sensitive information is protected.

AI makes this inseparability even clearer. People tend to trust AI outputs quickly, and if those outputs are based on poorly understood data, they can spread confusion quickly.

The problem is often time. Controls are implemented after data is shared, after AI tools have produced output, or after audits force a response, causing rework and progress to stall.

Building initial security looks like:

• Classify data and define access rules from day one
• Involve privacy and security teams during development
• Treat security as part of the workflow, not a separate approval layer

This is how agencies protect mission speed over the long term — by mitigating late-stage slowdowns.

Where SHI helps institutions turn intentions into actions

Many institutions are not caught up due to lack of policy. They get stuck because implementation is difficult in real environments — siled departments, unclear ownership, limited time to work through years of accumulated data, and modernization constraints that make large-scale change unrealistic.

SHI helps institutions build momentum without disrupting mission systems by working from what already exists and implementing governance in a practical and gradual manner.

It usually looks like:

• Turn decisions into actions. Defining ownership “swim lanes,” decision-making rights, and a board structure that keeps data, AI, and mission leaders aligned.
• Make data usable. Turn inventory into a navigable catalog with practical metadata and clear search capabilities.
• Prioritize what’s important. Focus on high-impact data sets first so progress is visible and sustainable.
• Build security early. Establish classification and access rules upfront so AI doesn’t amplify sensitive or unclear data.
• Linking governance to AI readiness. Align governance work with what institutions are trying to do today — scale analytics and AI responsibly, with teams moving together.

Data governance does not speed up the mission because it adds to the process. This speeds up missions because it reduces uncertainty.

When teams agree on what data means, who owns it, how it can be used, and how it is protected, leaders spend less time reconciling conflicting answers — and more time taking action on decisions that are still under scrutiny.

NEXT STEP

Want to see what data governance looks like in your institutional environment? We’re ready to help you figure it out — contact our team to start a conversation.

Looking for more information on how federal teams are approaching AI and modernization? Read our latest perspectives.