Diagnosis

Almost every engagement I take starts with a stated problem. Ads aren't working. We need a better CRM. Sales won't follow up. We need to hire someone senior.

I write the stated problem down and I take it seriously, because the person telling me has been living with it and they're rarely wrong that something is broken. They're just often wrong about where. The stated problem is the place the pain surfaces. The real problem is usually one or two layers away, in a system that feeds it, and the business has been spending money on the surface because that's the part everyone can see.

Diagnosis is the work of finding that second problem before anyone spends another dollar on the first one.

The pattern underneath most of it

After twenty years of doing this across agencies, startups, and my own companies, the pattern that keeps repeating is pretty simple. Companies fix what they can see, and what they can see is shaped by what they measure.

If your reporting can't tell a good month from a bad one, every problem looks like a campaign problem, because campaigns are the thing with a dashboard. If your CRM only captures what was easy to capture, every question that matters gets answered from memory in a meeting and then thrown away. If a job req has been open for a year, the instinct is to call a better recruiter, when the req itself is a description of a gap in how the product gets delivered.

Nobody involved is short on intelligence, the gap is instrumentation. And it's extremely common: shipping the easy thing feels productive, the way reorganizing the junk drawer without removing anything feels productive. Shipping the right thing takes more time and energy, mostly spent figuring out what the "right" thing actually means.

How I work through it

I often don't start with the thing I was hired to fix. There was a creative agency in Chicago with a social contract that stuck with me. Every client signed a simple agreement: "You tell us what you want. We tell you what you need." In my own work I added "and we'll meet somewhere in the middle," but the principle remains. The order matters more than the instinct.

I start with the plumbing: where the numbers come from, what they can and can't distinguish, and what decisions people are actually making with them. Then I look for the gap between what the business says it cares about and what it actually records. That gap is usually where the real problem is hiding, and sometimes the gap itself is the finding.

The essays in this section are each one version of that move.

Audit the plumbing before you touch the faucet is the closest thing to a field guide. It's about the engagement where I was hired to fix ads and opened the ads last, and the three questions I ask before touching any campaign.

The fields nothing writes to is about the CRM version of the problem. A schema full of empty fields looks like a data hygiene issue. Most of the time it's a list of questions your business still asks out loud and never keeps the answer to.

A job posting open for a year is a roadmap nobody wrote is original research. I read the public job boards of a few thousand funded companies to see which roles stay open, and the answer pointed somewhere most of those companies aren't looking.

What good diagnosis feels like

When it works, the founder usually doesn't feel told. They feel recognized. The thing they've been circling for six months suddenly has a shape, and the shape is smaller and more fixable than the dread they've been carrying around.

That's the part I enjoy most about this work. The fix is often less dramatic than the problem felt. A field that should have existed, a number that should have been split, a question that should have been asked before the budget got approved.

The hard part is seeing it from inside the business, where everything is urgent and every system is somebody's baby. That's why it can be so helpful to have someone from outside check the pipes first. Ideally someone who genuinely enjoys pipe inspections, which narrows the field a bit, but we're out there.