Momentum Over Perfection

There’s a version of this that every team goes through. Someone decides it’s time to get serious about AI. Maybe it’s the CEO after a board meeting, maybe it’s an ops lead who’s been experimenting on their own.

Either way, the conversation starts: where do we begin?

And that’s where most teams get stuck. Not because they lack ideas. Because they have too many. They want to pick the right project. The highest-impact one. The one that will prove the value of the whole initiative. They want a strategy before they start. A roadmap. An AI implementation plan that justifies the investment.

Six weeks later, you might have a few half-started projects or interesting demos, but nobody has built any real value. The team that was excited is now waiting for permission and alignment.

… What now?

This is a common pattern we see. It’s not a technology problem. It’s not a training problem. It’s perfectionism, wearing a planning hat.

Start with small wins.

The teams that build real AI operations don’t start with a strategy. They start with small wins.

And when we say small wins: we don’t mean small wins for some strategically important investment your company is making.

We mean literally anything. In fact, the easier, smaller, and more personal - often the better.

This seems counterintuitive. If you’re going to invest in AI operations, shouldn’t you start with the thing that matters most? The issue is the thing that matters most is almost always too complex, too high-stakes, and too politically loaded to be a good first project. It’s the equivalent of training a new hire on your most complicated process on their first day. You wouldn’t do that with a person. Don’t do it with a new way of working.

AI perfectionism may be hiding in plain sight.

Sign #1: Over analyzing where to start

The most common is the team that’s been debating which process to AI-ify for weeks. They’ve evaluated options, discussed impact, built spreadsheets comparing candidates. They want the first project to be a home run. Meanwhile, the team down the hall that picked their simplest process is on their fifth playbook and compounding.

Sign #2: Strategy decks are on v2

Then there’s the grand strategy version. Leadership wants a full AI roadmap before anyone touches a tool. By the time the strategy is “ready,” the team’s enthusiasm has died down and it’s not even clear if the original use cases are applicable anymore.

Sign #3: Solving integrations and edge cases, rather than the core

This last one is a bit harder to spot, but we see it often. A team picks a real project, starts building, and then keeps expanding it. “While we’re at it, let’s also handle the edge cases. And the integration with the other system because the manual setup here is annoying.”

The first playbook becomes a six-month initiative. If they had focused on the core part of the process, they would have shipped five playbooks in that time and been further ahead.

The real trap is that all of these patterns feel productive while they’re happening.

But momentum isn’t about feeling productive. It’s about actions and outcomes compounding.

What momentum looks like.

Momentum is the compound effect of a team building on what it learned yesterday. Each project gets built a little faster, a little sharper, a little more confident. That is the value that stacks.

When we work with teams, we run the same sequence:

1. Pick something small
2. Get the first playbook done as quick as possible
3. Run it, get everyone to see how it works
4. Improve it
5. And pick the next one

The first project teaches you how to document a process and what good instructions look like. The second teaches you about picking good use cases. By the third, you’re spotting patterns and able to think through bigger problems. By the fifth, you’re faster at building playbooks than you are at debating which one to build. That’s the compounding.

The question we get most often is “but what about ROI?” Teams want to justify each project individually.

• Will this specific playbook save enough time?
• Is this particular process worth documenting?

That framing is another form of perfectionism. The ROI isn’t in any single playbook. It’s across the portfolio. Some playbooks save hours. Some save minutes. Some just make the work clearer. The value compounds across all of them.

Trying to justify each one individually is like asking whether each individual day at the gym was worth the time.

Avoid getting stuck.

This is where most teams are actually stuck, and it’s worth saying plainly: the thing preventing progress usually isn’t a lack of capability or resources. It’s the belief that starting small is somehow not enough. That a real AI initiative needs to be big, visible, strategic.

It’s not. The simple thing is the unlock. The big, strategic, high-impact projects become possible because the team learned how to build on smaller ones. Not the other way around.

Building a full playbook-first operation, where this kind of compounding is built into how the team works, is a bigger transformation. That’s what we help teams do — reach out.

This is a Foundational protocol. It changes the way you work, not just the way you think about work.

The question it teaches you to ask: “How do we just focus on getting momentum?”

Know someone dealing with this? If you know a team that’s been “planning their AI strategy” for months and hasn’t shipped anything yet, send them this. Or someone who keeps saying they need to “find the right project to start with.” The right project is the one they can finish.

Read next: Protocol: Think Like an AI Operator

Figure out your path: Who Is This For?

Stay connected:

I lead a team and want to build an AI-capable organization. [Subscribe to Leading Momentum]

I’m operationally minded and want to build AI operations skills. [Subscribe to the Operator newsletter]