How I Find the Structure Hiding Inside the Chaos

Chaos is rarely random. Behind most messy projects there is a structure trying to get out: a handful of real priorities, a few people whose input changes everything, and one or two bottlenecks holding the rest hostage.

My job is to find that structure before anyone pours concrete.

Every Seat at the Table, Before the Plan

The loudest voice in the room is not always the most important one. The analyst who touches the data every day, the salesperson who hears the customer complaint, the engineer who knows which system is held together with tape: they see what an executive summary can’t.

So before I scope anything, I gather input from every seat that the work will touch. It is slower for a week and faster for the rest of the project, because effort lands on what actually matters instead of what was easiest to see from one chair.

Weigh Everything, Then Rank It

Gathering input is half the job. The other half is deciding. I weigh every perspective, then rank problems by urgency and impact so the team solves the ones that matter first, not the ones that happen to be visible from the top.

The same discipline applies to growth. At AE Global Solutions, the platform expanded its data product offerings by 120% through partnerships with external providers including S&P and Reuters. Growth like that only holds together when someone decides which integrations come first and how each one fits the data clients already rely on.

Map the Plan So Nobody Is Surprised

A plan only works if the people depending on it can read it. I map timelines, owners, and dependencies with every affected team in mind, so nobody finds out about a delay by living through it.

That map is also where restructuring opportunities show up. At Root Labs, mapping where our data budget actually went revealed that most of it paid for a vendor’s markup on data we could reach directly. Rebuilding that layer as our own ingestion pipeline cut third-party data costs by 80% and gave us control over how the data could be used. The savings came from seeing the structure clearly, not from working harder inside the old one.

Keep Scope Honest

Scope creep is not a rare event. In PMI’s 2018 Pulse of the Profession survey, 52% of projects completed in the prior year experienced scope creep or uncontrolled changes, up from 43% five years earlier.¹

The fix is not saying no to everything. It is building the plan with every voice up front, so late requests can be measured against what everyone already agreed matters. When a new ask arrives, the question becomes simple: what does it replace? Resources stop being wasted rebuilding what an inclusive plan would have caught the first time.

Watch: Build a tower, build a team (Tom Wujec, TED)

Tom Wujec’s marshmallow challenge makes the point in a few minutes: teams that test and iterate their structure early beat teams that plan in the abstract and discover the problem at the deadline.

Solve the Bottleneck First

Every system has one constraint that sets the pace for everything else. Optimize anything other than the bottleneck and you have only made a faster queue.

When I led four developers through a SQL ORM data pipeline in Python, the constraint was not code. It was infrastructure: what we could deploy, where, and how fast. We sequenced the build around those limits instead of pretending they weren’t there, and the project moved from architecture to deployment without stalling.

What You Get on Day One

• A plan built from every seat, so the people who can sink it later get heard now.
• Ruthless prioritization: the most urgent problems first, not the most visible ones.
• A map everyone can read, with owners and dependencies nobody has to guess at.
• Scope that stays honest, because new requests get measured against what everyone agreed matters.

Sources

1. Project Management Institute, Pulse of the Profession 2018, as summarized in PMI’s “Scope Patrol.”
2. Tom Wujec, “Build a tower, build a team”, TED.