We don’t have a shortage of vision in the public sector. What we’re missing is a playbook for how to actually get things done.
It’s easy to announce reforms, adopt strategies, or promise digital transformation. But implementation—the unglamorous, messy middle—often gets overlooked. And yet that’s where change lives or dies.
If you’re a public sector leader tasked with making things happen, this post is for you. It’s not about theory. It’s about the blind spots effective leaders face—and how to avoid them.
The Problem Isn’t the What — It’s the How
Most initiatives start with big aspirations: a press release, a vision statement, a commitment to improve a service or reform a process. But they stall when it comes time to deliver.
Why? Because vision alone doesn’t deliver results. Execution does.
In the same way you don’t run a marathon by adding it to your annual goals list, you don’t implement reform by approving a strategy. You get to the finish line by creating routines, incentives, and decision processes that sustain effort over time.
As James Clear puts it:
“You don’t rise to the level of your goals. You fall to the level of your systems.”
In government, systems protect the status quo and are not neutral. Changing them requires deliberate effort.
A Field Guide to Getting Things Done in Government
After working with public institutions across Latin America and beyond, I’ve seen the same pattern: the teams that get things done—really done—approach implementation with discipline and realism.
I think of it as Google Maps for implementation: you need to know where you are, where you’re going, and how you’ll get there—before you hit the gas.
Skipping this step doesn’t save time, it creates an exponential cost in the life of a project in terms of political and organizational capital.
Here’s the model I use, blending design thinking, implementation logic, and real-world experience.
I. Start With the Basics: The Google Maps of Implementation
Before you jump into delivery, answer three questions. If you don’t, you’ll spend the rest of the project fixing problems mid-flight.
1. Where are you now?
Systems are not broken, they are in place to serve a specific purpose, which is it? What are users experiencing? What’s actually causing the pain? What constraints—political, administrative, or institutional—shape what’s possible?
2. Where are you going?
What is the goal—and is it clear, measurable, and genuinely shared by all key players? What is getting in the w
3. How will you get there?
Do you have a realistic, sequenced plan that reflects how decisions are actually made in your system? Who are your supporters and who your detractors?
These questions are obvious—and that’s precisely why senior teams skip them. Aiming to save time the assumption is the answers are clear for everyone. They rarely are.
This blind spot creates coordination debt: misunderstandings and misalignments that compound over time and become increasingly costly to unwind.
II. Make Things Move: Where Execution Breaks Down
Once there’s clarity, the real work begins. And this is where public leaders usually lose momentum—not because they lack skill, but because authority, incentives, and routines quietly work against delivery.
1. Start Smart—With Eyes Open
Find your “doers”: people with credibility, energy, and the ability to move work forward. Start small. Prove value early.
But be careful. Early wins often create political exposure faster than teams expect. What looks like progress can quickly attract resistance if leaders underestimate who feels threatened by change.
Startups talk about product–market fit. In government, think problem–system fit. If what you’re doing clashes with how your institution actually functions, you either adapt the intervention—or deliberately engage the levers needed to change the system itself.
A useful reminder: the system is not “broken.” It is producing exactly the results it was designed—formally or informally—to produce. Your job is to understand what purpose it currently serves.
2. Monitor Progress—or Fly Blind
Build feedback loops from day one. You can’t adapt if you don’t know what’s happening.
In Results, Charlie Baker and Steve Kadish argue that effective governments make decisions with data—not assumptions. But when dashboards and KPIs become compliance rituals, disconnected from real decisions, they drain energy instead of creating insight.
Pause and ask:
• Are we tracking what actually informs decisions?
• Are we willing to change course if the data tells us to?
• Or are these indicators simply signaling activity, not progress?
If you are not ready to act differently based on what the data shows, then your monitoring mechanisms are an illusion.
3. Adapt as You Go—Deliberately
Implementation is not a straight line. Plans will shift. Resistance will surface. Roadblocks are normal.
The challenge isn’t adaptation—it’s authorizing it. And in systems that defend the status quo, it has to be explicit.
Too often, teams are told to be agile while operating in systems designed for rigidity. Make sure you build in both the time and the permission to adjust course without losing sight of the goal.
Adaptive delivery isn’t improvisation. It’s disciplined learning under constraint.
III. Build the Systems That Sustain Progress
Execution isn’t just about effort. It’s about systems.
Think in three layers:
• Personal: where you delay decisions, avoid conflict, or over-accommodate.
• Team: how decisions are made, escalated, or quietly deferred.
• Organization: the cadence of delivery, informal networks, veto points, and the sequencing of authority.
Understanding this “plumbing” matters. In Esther Duflo’s words, Policy is like plumbing. You need to understand how things actually flow.
The Real Test of Leadership
Getting things done in government isn’t about better plans. It’s about deciding sooner than feels comfortable where authority must be used—and using it before delay hardens into structure.
Most failures don’t come from lack of effort or good intentions. They come from postponing decisions in the name of alignment, allowing coordination to substitute for accountability, and waiting for permission that will never arrive.
Every day a leader avoids a hard call, the system makes one instead—by default, in favor of the status quo.
That is the difference between announcing change and being responsible for whether it happens.
