The Workplan Failed. The Data Didn't.

In God we trust; all others bring data.

W. Edwards Deming, pioneer of modern quality management

The real constraint was never the plan

Why does a carefully designed strategic initiative fail in a young institution, while something as unglamorous as a daily dashboard quietly succeeds? I used to assume the answer was about the quality of the plan itself. It wasn't. It was about how much organizational capability each approach demanded before it could produce any value at all — and we had far less of that capability, in our first years, than any of our planning documents assumed.

A work plan that looked complete on paper

Twenty-five initiatives. Five segments — revenue growth, cost control, people growth, quality systems, and better commercial-operations integration. Every initiative had an owner, a target, and a deadline. It was headquarters' strategic priorities, carefully translated into a structure that should have turned intention into execution.

Most of it didn't move.

Where the workplan stalled

The first problem was experience. Our leaders had never delivered a formal workplan before, and the gap between "here is your target" and "here is what you actually do on Monday morning" turned out to be much wider than anyone had planned for. Many were overwhelmed simply trying to break a target into actionable steps. The second problem was capability, not headcount — we had plenty of managers, but too few who could genuinely drive an initiative to completion, and grouping several managers onto the same initiative didn't fix it. A group of people who each struggle to take initiative doesn't add up to a team that can. The third problem was momentum. Progress required constant pushing — recurring monitoring meetings just to keep initiatives inching forward, and in several cases, telling leaders exactly which dashboard to build rather than waiting for them to identify what they needed to track. It felt, at times, like pushing a square wheel: it moved only for as long as you kept pushing, and stopped the moment you didn't.

Of the 25 initiatives, only five to seven ultimately succeeded — a pricing book revamp, tighter discount controls, a Kaizen culture built around a Yellow Belt program, and specialist performance tracking among them. Tellingly, the initiatives that survived weren't necessarily the most strategically important ones on the original list. They were the ones simple enough for the leaders we actually had, at the capability they actually had, to execute without constant intervention.

Meanwhile, something quieter was working

While the workplan struggled, our data was working better than expected. Early on, we'd been deliberate about building analytics into daily operations rather than treating it as a reporting afterthought. One example became central to how we actually ran the hospital day to day: an automated dashboard, built using AI tools, that reports patient figures from the last five days alongside the day's planned bookings. It gives us a live sense of expected workload across outpatient and inpatient services before the day even starts — and it directly informs decisions that used to run on instinct. If ICU, with its five available beds, is already full, we can see that clearly enough in the morning to decide whether to reallocate nursing manpower elsewhere for the day, rather than discovering the imbalance mid-afternoon. On days when outpatient demand is expected to spike, departments can prepare ahead of time rather than react once queues have already formed. This dashboard now anchors our daily operations review. It didn't require a workplan, an owner, or a monitoring meeting to keep it alive. It simply told us something true, every single day, and people used it because it was useful.

The capability gap, made visible

The workplan asked leaders to do something they'd never done before — set direction, break it into steps, and drive it without oversight — and most of them, reasonably, couldn't yet. The dashboard asked for something much smaller: look at what's true today, and make one sensible call based on it. That's the actual difference. It was never that data is more persuasive than a plan. It's that one approach demanded a capability we hadn't yet built in our people, and the other didn't require that capability to exist at all — it just needed to be accurate and visible, and it got used.

That distinction has quietly reshaped how we think about driving change here. A formal workplan assumes an organization that can already execute against ambition. A good dashboard doesn't assume anything about the organization — it just has to be true. The lesson isn't that workplans are unnecessary; institutions still need direction, accountability, and long-term goals. It's a lesson about sequencing: we tried to strengthen execution through planning before we'd strengthened it through measurement, when it should probably have gone the other way. Planning tells people where to go. Measurement helps them understand where they actually are. In a young institution, an organization has to be capable of the second before the first can do much good.

What the dashboard made possible next

We've since built out that same instinct into a handful of other tools — small, AI-assisted builds that estimate a patient's likely bill size before treatment, track waiting times across departments, and help process insurance claims faster. What made this practical now, in a way it wouldn't have been a few years ago, is how much faster these tools are to build: what used to require competing for scarce IT resources and months of development can often be prototyped in days. None of these began as workplan initiatives. Each started with a much smaller question — would it help to see this more clearly? — and the answer was usually yes.

The workplan told people what to do. The data told them what was true. In the end, it was the latter that people kept coming back to.