Why Your Business Framework Isn’t Working (And What Actually Will)

Your framework isn’t failing you. You’re failing the framework.

That’s not meant to sting. It’s meant to redirect, because almost every leader I talk to who’s frustrated with the results of a business operating framework, EOS, Scaling Up, OKRs, Lean, you name it, is asking the wrong question. They’re asking which framework would work better. The real question is what they would have to do differently to make the framework they already have actually deliver.

In nearly 30 years of working inside operations, I’ve watched companies adopt and abandon framework after framework. Same leadership team. Same operation. Different tool every time. And every time, the explanation for why it didn’t work is some version of “this one wasn’t the right fit for us.”

Sometimes that’s true. Most of the time, it isn’t.

The Pattern That Plays Out Almost Every Time

Here’s the pattern I see across mid-market manufacturing and service operations.

Leadership adopts a framework. There’s energy in the rollout. Training happens. People are paying attention. Initial steps get taken. Maybe a few quick wins land. And then the harder, longer work starts, the part where the framework is supposed to compound over months and years, not weeks.

And that’s where it usually breaks.

Leadership doesn’t see immediate results. The fires are still burning. The pressure is still there. The lagging indicators they’re used to watching, revenue, margin, on-time delivery, customer complaints, haven’t moved yet, or have moved only slightly. From the leadership perspective, it feels like nothing is working.

So they decide to change direction.

That could mean a different tool. A different framework. A pivot to something completely different. The reasoning is almost always the same: “this isn’t producing the results we need.” And the moment that decision gets made, the team on the floor experiences it as exactly what it looks like. Flavor of the month. Another initiative that’s about to get replaced by the next one. So they stop investing real effort in any of it. They wait to see what’s next.

The leader, meanwhile, is convinced the team is the problem. The team isn’t executing. The team isn’t listening. The team isn’t doing enough. In a lot of cases, none of that is true. The team was actually starting to get traction. It was a matter of time. But the leader couldn’t wait for the time, because they couldn’t see the proof.

I’ve seen this cycle play out with EOS implementations that got swapped for Scaling Up. With Lean rollouts that got abandoned after six months. With OKR processes that got modified into uselessness because leadership wanted faster signals. The framework doesn’t matter. The pattern is the same.

The Marble Jar Principle

Brené Brown has a concept I think about often when this comes up. She calls it the marble jar. The idea is that trust is built through small moments, M&Ms or marbles going into a jar one at a time. Each small moment of someone showing up, following through, doing what they said they would, adds one more marble. The full jar, the visible result, is the accumulation of all those small deposits.

Discipline with an operating framework works exactly the same way. Every meeting that runs the way the framework prescribes. Every decision that gets routed through the right process. Every metric that gets reviewed on the cadence it’s supposed to be reviewed on. Every uncomfortable conversation that happens when someone isn’t following the new rhythm. Each of those is a marble in the jar.

The visible result, the operational improvement that everyone’s hoping to see, is what the full jar looks like.

The companies that fail with framework after framework are dumping the jar out every six months and starting over with a new container. They never accumulate enough deposits in any one container to see what a full jar actually looks like. Then they conclude that none of the containers work.

This is the discipline problem, not a framework problem. And it’s why selecting a different framework doesn’t solve it. You’re still going to dump that jar out, too.

The 5S Story

Let me make this concrete with the example I see most often in manufacturing operations.

5S is one of the foundational Lean tools. Sort, Set in Order, Shine, Standardize, Sustain. It’s about creating organized, visual, consistent workspaces. Plenty of companies adopt it. Two very different things happen depending on how it’s adopted.

The surface-level version goes like this. Leadership decides to “implement 5S.” The team gets trained. Workspaces get cleaned up, organized, labeled. There’s a visible difference. People take photos for the newsletter. And then the company moves on to whatever’s next on the priority list, because 5S looks done.

Six months later, the workspaces look exactly the way they did before 5S. Tools are scattered. Materials are out of place. The labels are still on the walls, but nobody’s maintaining what they were labeling. Leadership concludes that “Lean doesn’t work here,” or “5S didn’t stick,” or “the team didn’t take it seriously.”

What actually happened is that 5S was deployed as the answer. The leadership team thought “we’ll do 5S and that will fix things.” But 5S doesn’t fix utilization. It doesn’t fix quality issues to any meaningful degree. What it does is create the conditions, organized workspaces, visible standards, that make every other Lean tool work better. It’s an entry point, not a solution. When it’s treated as a solution, it can’t deliver what was expected of it. And when leadership doesn’t see the operational results they were expecting, they stop reinforcing the practice, and it falls apart.

The discipline-led version starts somewhere completely different. Leadership decides they believe that eliminating waste, in all its forms, is the path to higher utilization and a better customer experience. They commit to that as a long-term direction, not a project. 5S becomes the visible entry point into that work. The team learns 5S, applies it, and sees some early momentum. But it doesn’t stop there.

The team understands why they’re doing 5S. They see how it connects to something larger than tidier workstations. The leadership team keeps the broader purpose visible in every conversation. They tie the daily 5S practices to leading indicators they actually monitor, like cycle time, work-in-process levels, customer-experience signals. Months in, when the visible operational results start to compound, it doesn’t feel like a sudden breakthrough. It feels like the natural consequence of work that was already being done. The team kept depositing marbles, and the jar filled up.

Same tool. Two completely different outcomes. The difference isn’t the tool. It’s the discipline to keep showing up for it after the initial novelty wears off.

Why Leaders Pull the Plug Too Early

The hardest part of operational discipline isn’t the work itself. It’s the window of time between adoption and visible results, when leadership has no proof that what they’re doing is working.

That window is where most companies break.

The instinct in that window is to look for confirmation. Is this working yet? Are the numbers moving? Should we try something different? Leadership scans for proof, and the proof they’re looking for is operational, the lagging indicators they’ve been watching for years. Revenue. Margin. Throughput. On-time delivery. Defect rate.

But those are lagging indicators by definition. They reflect work that was done weeks or months ago. Watching them in the early stages of a framework rollout is like checking the oven temperature by tasting the cake. You’re going to draw the wrong conclusion almost every time.

The leaders who successfully hold the discipline through that window are doing something different. They’re identifying leading indicators, signals that show up before the operational numbers move, and they’re monitoring those instead.

What Leading Indicators Actually Look Like

This is the piece I want to make concrete, because “track leading indicators” is one of those phrases that sounds obvious and means nothing without specifics.

Stay with the Lean example. If you’re trying to improve customer experience by eliminating waste, here are leading indicators you could start tracking before the operational numbers move:

Total orders in backlog. Are we keeping up with intake, or is the backlog growing? Backlog trend is a real-time signal for whether the work you’re doing is moving the system in the right direction.

Cycle times by work center. Even rough measurement. Are the cycles getting shorter? More predictable? Less variable?

Product quality signals at the source. Defects caught upstream, rework loops, the kinds of issues that traditionally surface at final inspection. If the framework is doing what it’s supposed to do, these should start moving before final defect rates move.

Work-center utilization. Granular, but useful if you can measure it. Where is capacity actually being absorbed?

Some of these depend on data collection capabilities your operation may or may not have. If your data infrastructure can’t support that level of granularity, don’t let that stop you. Keep it simple.

Put a whiteboard on the wall and track daily output. That’s it. If the number on the board is trending up while you’re doing this work, and you know what the number needs to be to clear backlog and improve cycle time, you have your leading indicator. No software required. No dashboard. Just a number on a board that the team writes down every day.

The point isn’t sophistication. The point is having something to watch that gives you confidence you’re on the right track before the lagging numbers move, so you don’t lose your nerve in the window where most companies pull the plug.

(I’ve written more on leading vs. lagging indicators in other pieces. The principle applies far beyond framework rollouts.)

What Operational Discipline Actually Looks Like

If you take one thing from this article, let it be this: operational discipline is the willingness to do the work consistently when you don’t yet have proof that it’s working.

In practice, that looks like a few specific things.

A connected purpose. The team understands why the framework or methodology is being used. Not in a vision-statement way. In a “this is how this connects to what matters” way. When people understand the purpose, they sustain the practice even when leadership isn’t watching.

A weekly cadence that doesn’t slip. The meetings that the framework prescribes actually happen. Every week. Even when there’s pressure. Especially when there’s pressure. The most common failure mode is “we’ll skip the weekly review this week because we’re slammed.” Skip it twice and the discipline is already eroding.

Leading indicators that get tracked. Even simple ones. The team and leadership have something they’re watching that tells them whether the work is going in the right direction, so they’re not waiting for lagging proof.

Quiet execution. The companies that succeed with frameworks talk about them less than the companies that fail. They’re not announcing initiatives. They’re just doing the work, week after week, watching the signals, adjusting based on what they see.

Patience for the marble jar. Leadership understands that the visible results compound. They’re not measuring progress by whether the operational numbers have moved yet. They’re measuring progress by whether the deposits are being made.

When those things are in place, almost any framework will work. EOS works. Scaling Up works. Lean works. OKRs work. When those things aren’t in place, no framework will work, and switching frameworks won’t change that.

Final Thoughts

The hardest part of operational discipline is also the simplest. It’s the willingness to keep doing the work after the novelty wears off and before the proof shows up. It’s making the marble deposits when nobody is checking the jar yet.

Most leaders confuse activity with discipline. They confuse the announcement of a framework with the practice of it. They confuse the early visible wins with sustained operational change. And then when the operational numbers don’t move on the timeline they expected, they conclude the framework was the problem, and start looking for the next one.

The framework isn’t the problem. The willingness to stay with it through the window where you can’t yet see the proof is.

That’s the question worth asking. Not “is this the right framework?” but “are we actually doing the work the framework requires?” If the honest answer is no, the next framework won’t help. If the honest answer is yes, the framework you already have probably will.

That’s it for today.

See you all again next week!

Dave