Governance Failure Pattern #2: Trying to “Boil the Ocean”
Last week we were introduced to a well, intentioned organization that fell into the Failure Pattern #1: Governance by Committee, Not by Design.
Governance by committee feels productive.
It creates structure, but what you get is policies that don’t map to business outcomes, roles that exist on paper but not in practice and meetings that recycle the same talking points without driving decisions.
We learned that effective governance doesn’t start with a committee. It starts with a blueprint, a clear articulation of what governance needs to achieve; asking the hard business, first questions up front, defining outcomes and building the governance backward from those outcomes.
Let’s take a look at failure pattern #2…and how you can avoid it.
In the world of enterprise AI & Data Governance, few missteps are as common, or as costly, as the attempt to launch governance everywhere, all at once.
Here is another familiar scene.
Executive leaders of a national health insurance provider have been discussing the need for better data governance, so it is decided. “We need more cow bell”.
A program is launched.
They scope it to every domain, every system, every stakeholder.
They build a massive roadmap.
They hold kickoff meetings with 40+ attendees.
And then… nothing moves.
No decisions.
No trust built.
No measurable impact.
Just a sprawling initiative with no clear wins.
This is the second governance failure pattern:
Trying to Boil the Ocean.
The Paralysis of Overreach
Governance is complex.
But complexity isn’t the enemy, lack of prioritization is.
When everything is a priority, nothing is.
When every domain is in scope, no domain gets traction.
When every stakeholder is invited, no one feels accountable.
The result?
- Endless meetings
- Vague deliverables
- Diluted ownership
- Zero momentum
It’s governance by ambition, not by design.
Why This Happens
Most governance programs start with bad scoping.
They confuse scale with success.
They assume more domains = more credibility.
They believe broad buy, in = faster adoption.
But governance isn’t a popularity contest.
It’s a design challenge.
And design requires focus.
What to Do Instead: Start Small, Win Big
The antidote to boiling the ocean?
Modular governance.
Governance doesn’t need to start big. It needs to start smart.
The most effective programs begin with a modular approach; targeting one high, impact domain where governance can deliver tangible value. Think:
- Clinical data in healthcare (where accuracy and compliance are mission, critical)
- Customer master data in SaaS (where clean records drive revenue and retention)
- Product hierarchy in manufacturing (where consistency fuels supply chain efficiency)
By narrowing the scope, teams can build a pilot that’s not just functional, but transformational.
They can define clear roles, implement policies, measure impact, and showcase results.
This creates momentum. It builds trust. And it gives governance a story worth telling.
Once the pilot succeeds, it becomes the blueprint for scaling.
Other domains follow.
Stakeholders engage.
And governance evolves from a theoretical ideal to a practical, business, aligned capability.
How to Identify Your Starting Domain
Don’t start with what’s easiest.
Start with what’s hurting, help ease the pain and you will inevitably identify an “early adopter” that you can eventually turn into a governance advocate. What you want to do is build momentum, because right now you have none (maybe even less than none).
The fastest way to build momentum is to solve a real business problem, one that’s already costing time, trust, or money due to poor data.
Here’s how to find it:
- Scan for Friction Zones
Look across departments for areas where data issues routinely derail operations.- Are sales teams arguing over customer definitions?
- Are compliance teams scrambling during audits?
- Are AI models getting rejected due to lack of transparency?
- Interview the Business
Talk to domain leaders, not about governance, but about pain.
Ask:- “Where does bad data slow you down?”
- “What decisions are hard to make because the data isn’t trusted?”
- “Where do you spend time fixing instead of executing?”
- Quantify the Impact
Don’t just collect anecdotes, tie them to metrics.- Lost revenue
- Delayed launches
- Failed audits
- Customer churn
These are governance opportunities in disguise.
- Pick a Domain with High Visibility and Low Political Risk
You want a pilot that’s meaningful but manageable.- High enough impact to matter
- Narrow enough scope to deliver
- Clear enough ownership to activate
- Design for a Quick Win
Choose a domain where governance can show results in 90 days.- Cleaner master data
- Faster approvals
- Reduced rework
- Improved model confidence
When governance solves a real business problem, it earns credibility.
And credibility is the currency you’ll need to scale.
Not the easiest. Not the most political.
The one that actually moves the needle.
Examples:
- Clinical data in healthcare
- Customer master data in SaaS
- Risk and compliance data in banking
- Supplier data in manufacturing
Build a pilot.
Define clear outcomes.
Activate roles.
Measure impact.
Then scale, intentionally, iteratively, and visibly.
The Modular Blueprint
Here’s how modular governance works:
| Step | Action | Outcome |
| 1 | Choose a high, impact domain | Business relevance and urgency |
| 2 | Define outcomes | Clarity on what governance must enable |
| 3 | Build backward | Roles, decisions, trust mechanisms |
| 4 | Deliver a pilot | Measurable value and executive proof |
| 5 | Scale iteratively | Expand with credibility and momentum |
This isn’t just agile.
It’s strategic.
It’s governance that earns the right to grow.
Final Thought: Don’t Boil the Ocean. Serve a Dish That Works.
Governance isn’t a buffet.
It’s a curated menu.
Start with one dish.
Make it exceptional.
Then invite others to the table.
Because when governance delivers value, it doesn’t need to be sold.
It gets requested.