Assess First: A Practical Guide to Org Design

Welcome back to Science of Leading. We've spent two episodes on why structure matters and what a failing one looks like. Today I want the actual process. If a leader accepts they have a design problem, what do they do Monday morning?

Monday morning, they do not touch the org chart. That's the mistake almost everyone makes. They redraw boxes before they know what's actually broken. The process starts with facts, not a redesign.

So step one is assessment.

Step one is assessment, and it's specific. Where does work slow down, cycle time, backlog, rework. Where does quality drop, defects, customer complaints. Where is talent risk rising, turnover, internal transfers, burnout signals. And where do decisions stall, approval queues, meeting load. You're not guessing at this. You're using validated behavioral data and quick surveys to find the actual pattern by function, by layer, by team.

And step two builds on that.

Step two, you map decision rights. What are the key decisions that actually drive outcomes, who currently decides them, who should decide them based on who's closest to the information, and what are the real reporting lines and escalation rules. This is the step where most leaders discover their real structure isn't the org chart at all. It's the influence network.

That's such an interesting way to put it, the org chart tells you the fiction and the influence network tells you the truth.

Exactly. Once you have both maps, the fiction and the truth, step three is fixing role design before you touch a single box. Simplify role descriptions. Remove overlap. Define authority and accountability explicitly. Set decision rules for the choices that come up again and again, so people aren't relitigating the same call every month.

And that alone creates lift, before anyone's even reorganized anything.

Immediate lift, and it's nearly free compared to a reorg. Step four is where discipline matters most. Pilot the new design in one area. One department, where the workflow is measurable, the leader is genuinely committed, and you can compare before and after with real numbers. You do not roll out a new design company wide unless chaos is the goal.

Deloitte's research on this actually backs up how rare the discipline is. They found plenty of leaders say redesign is a priority, but far fewer feel confident they can actually execute it well.

That confidence gap is almost always a cadence gap. People try to do a bold, one time reorg instead of a disciplined pilot. Which brings us to step five, collect the feedback, both kinds. Quant is KPI movement, cycle time, quality, engagement and retention signals. Qual is where confusion increased, where decisions got faster or slower, what got easier or harder for the people actually doing the work.

And step six.

Step six, you treat it as a cadence, not an event. Quarterly review for friction and KPI drift. Semiannual review for structural changes. Annual review for strategy alignment and capacity planning. Across OAD's own data, ten million assessments, more than forty countries, the organizations that treat this as an ongoing rhythm outperform the ones that treat it as a once a year event, every time.

Okay, so let's talk about the KPIs themselves, because step five said "collect feedback," but what are you actually measuring on an ongoing basis, not just during a pilot?

Three layers, and if you're missing one, you're flying with half the instruments dark. Layer one is business outcomes, revenue growth or margin, customer satisfaction and retention, quality metrics like defects and escalations, and time to delivery. That's the layer everyone already tracks.

And layer two is the one people skip.

Layer two is the execution system itself. Decision cycle time for the decisions that actually matter. Handoff failure rate, rework, reopened tickets. Meeting load and approval queues. Time to fill and time to productivity for critical roles. This layer tells you why layer one is moving, not just that it moved.

And layer three?

Layer three is people and capability. Engagement, regrettable turnover, internal mobility and promotions, manager effectiveness signals like span health and coaching cadence. Gallup's meta analysis is the reminder here, engagement correlates with productivity, profitability, and turnover directly. It's not a separate HR report nobody reads. It belongs in the same review as revenue.

So the leaders who get this right are looking at all three layers together, not picking their favorite one and ignoring the rest.

That's the whole discipline. Business outcomes tell you what happened. Execution indicators tell you why. People indicators tell you whether it's sustainable. Miss any one of the three and you're managing blind on that dimension.

If you want to see where your own team sits across those layers before you run a pilot of your own, OAD has a free Top Performer Profile, real behavioral data on your people in about seven minutes, results in twenty four hours, no credit card. That's O-A-D dot A-I.

Assess first. Redesign second. In that order, always.

Perfect place to leave it. Thanks.

Always.