The cadence failures we lived through. The anti-patterns, from experience.

The status-update meeting — where nobody is listening

Before Scaling Up and EOS, our leadership meeting was the default leadership meeting — the one I still see almost everywhere when I coach teams on EOS and other business frameworks. Everyone goes around the table and reports what they did. It feels productive. It sounds like people are doing stuff and working. And almost nothing real happens in the room.

The core problem with a status-update meeting is that no one is really paying attention, because everyone is rehearsing their own update in their head. Then whatever discussion happens spins off whichever topic surfaced last — unprioritized, unconnected to any goals we'd actually set. There were no real, meaningful issue discussions. The metrics, to the extent they showed up at all, were woven into the narrative of each update, where they could be framed however the presenter wanted. It was all feel-good.

The daily standup is this meeting's cousin, and while there's a legitimate place for those in the right context, I'm not a fan of either as the operating heartbeat of a leadership team. What killed the status update for us, once and for all, was implementing a real framework — Scaling Up first, then EOS. The structure of a proper weekly meeting simply leaves no room for it: the scorecard replaces the narrated numbers, the issues list replaces the meandering discussion, and the update shrinks to the headlines. We never went back.

The founders taking all the air time

This one I'd put more honestly as founders, plural, because it was both of us. Before EOS, our leadership team had a real mix of experience levels, and in a room like that, something predictable happens: the room defaults to the co-founders' point of view. Our read of the situation became the read. Our decisions went mostly unchallenged. Nobody was being dishonest — it's just the gravity that founders exert when they've been there longest and talk the most. What you end up with is a team of yes-men and yes-women, and you don't even notice, because everyone is agreeable and the meetings feel smooth.

What changed it was EOS — specifically, our Integrator. She didn't just take over running the meeting; she challenged the co-founders directly, and she challenged everyone else on the leadership team to actively listen and actively participate instead of deferring. And she did something I'd recommend to any team that suspects it has this problem: she actually measured our talk time across several meetings, to make sure the two of us weren't taking up 80% of the air.

That was the moment our leadership culture actually changed — from a team that nodded along to the founders toward an open, honest, problem-solving decision-making team. Not because anyone gave a speech about candor, but because someone with the standing to do it enforced the floor time and called out the deference. If your meetings feel smooth and your decisions never get challenged, I'd bet money you have this failure right now. Put a clock on yourself for three meetings and find out.

More metrics, less control

There's a tendency, when you're building a scorecard, to keep putting more on the board. Every metric you add feels like another instrument on the dashboard — more visibility, more control over the business. That feeling is false. What you actually get is an overwhelming board where the few numbers that matter are buried among the many that don't, and the weekly review turns into a reading exercise instead of a focusing one.

We lived this — our real scorecard grew to 23 metrics before we understood that the ideal version was 13. The principle that experience taught us is the oldest one there is: keep it simple, stupid. Less is more. The rule I'd give any leadership team now is to aim for ten or fewer metrics on the main leadership scorecard. Anything beyond that doesn't belong at the leadership level — push it down to the team scorecards, where the people closest to those numbers can watch them properly. The leadership scorecard isn't supposed to be a census of everything measurable. It's supposed to be the short list of numbers the company lives or dies by this quarter.

The right data, the wrong metric

This failure is sneakier than bad data, because the data was fine. For a decent stretch, we had product usage numbers on the scorecard — DAU and DAU/MAU. They were accurate. They were on time. And they didn't really tell us much of anything — a stickiness ratio on the leadership scorecard wasn't generating issues or changing decisions.

What I wish we'd tracked instead were the adoption metrics: activation rate, adoption rate, and time to value. Those are the numbers that would have told us whether new customers were actually getting wired into the product — the thing that drives everything downstream. That's where I wish we had spent our time. The lesson isn't "track more product metrics." It's that a metric earns its slot on the scorecard by changing what you do. A number that's accurate, timely, and decision-free isn't neutral — it's occupying a slot that a decision-driving number should have. Audit your scorecard for these squatters. We had them for years.

The quiet failure: weak goals

The failure I most wish I'd caught sooner never made a dramatic story, which is exactly why it survived so long. It was the quality of our goals. Every metric had a goal — we held that line — but some of those goals were weak. Some were business as usual dressed up as targets. Some were uninspiring. Some were too vague — not SMART enough, not thinking big enough. The machinery of the cadence ran perfectly around them either way: the scorecard got reviewed, the variances got flagged, the issues got worked. A great process will faithfully execute against mediocre goals and never once complain.

I wish we had spent more time in our planning process defining more robust goals — pressure-testing whether each one was specific, whether it would actually move the business, whether it made anyone sit up. That's the failure mode nobody warns you about, because it doesn't look like failure. It looks like a well-run company hitting its numbers. The question is whether the numbers were worth hitting.

And against that regret, the thing I'm proudest of: our commitment to the process itself. We didn't run EOS as a leadership-team ritual while the rest of the company worked some other way — we rolled it out fully, across all the teams, scorecards everywhere. The cadence became how the whole company operated, not how the executives met on Tuesdays. Every failure on this page was survivable because the system around it was strong enough to eventually surface it. That's the honest case for the cadence: not that it prevents you from operating badly, but that it keeps confronting you with the evidence until you stop.