The pattern
A bad back can hurt in your leg, sometimes all the way down to the ankle.
The pain shows up in the leg; the nerve is in the spine, and rubbing the leg won't fix it. Companies work the same way: what the team can name is where it shows up, and where it's made is further upstream and better hidden.
What looks broken is rarely where it comes from.
A team names a problem that keeps coming back, and everyone in the room agrees something is wrong. Six months later it has wandered from operations to product to hiring to a “culture issue”, and nobody can quite say where it lives anymore.
It travels along the path of least resistance, usually toward the team that can be asked to absorb it without saying no.
By the time anyone names it, it has already moved twice.
- Where it is produced
one decision, made every week
- Operations
- Product
- Hiring
- “Culture”
where it gets treated
we walk back along this line
- Where it is produced
one decision, made every week
- Operations
- Product
- Hiring
- “Culture”
where it gets treated
we walk back along this line
Three things keep pushing it away from where it's made.
Feedback decay
The loops that would tell a company about itself get cut to save money or time, and after a while it can't feel its own pain where it's made.
Role hardening
The people closest to the pain learn to soak it up. Nobody sees that work, and after a while it quietly becomes their job, which is exactly where everyone then tries to treat it.
Attention follows money
People see what they're paid to see. Whatever nobody is paid to see drops out of view within a quarter, however good everyone's intentions are.
A firm paid where the pain is felt works where the pain is felt.
Why consulting usually works on the leg
The seven things people hate about consulting all come from one problem: consulting is paid where the pain is felt, by the hour and for the rescue, so nothing in the deal ever makes a firm walk upstream to where the problem is made.
- “They told us what we already knew.”
- It feels obvious because the team already knows the symptom. What nobody has is a reliable read on the cause.
- “We got a deck.”
- A deck can describe the pain perfectly and change nothing about what keeps causing it. Describing it starts to feel like fixing it.
- “They had nothing at stake.”
- If a firm is paid for activity, nothing in the deal makes it stay until the work holds.
- “The contract kept growing.”
- If the firm needs the engagement to continue, your problem quietly becomes something it can renew.
- “Seniors sold it, juniors did it.”
- The senior person who could see the system sells the job and moves on, and the people doing the repair never saw what they saw.
- “Nobody understood a word.”
- Borrowed jargon makes a firm sound precise while it drifts further from what your team actually lives with.
- “They treated us like everyone else.”
- Benchmarks turn your company into a type. The work is reading this one company, now, with the history it has.
Why it can work this time
There's a bunch of reasons. One is where we come from: all of us grew up in chaotic, bleeding-edge impact projects, running things without money alongside people who were idealistic and frequently allergic to following orders (more on that in Roots). Doing the same work with a budget, inside a conventional organisation, is playing the game on easy mode. Another is the mix of people and the mix of skills we bring together. And every one of us has spent years doing real work in the real world, with our own hands. We are not consultants. We're the opposite of consultants, which, on a site that sells consulting, is a certain irony.
What we do about it
Every piece of work runs the same four moves.
We solve the presenting problem first, fast and hands-on, so people feel the difference. While we fix it, we find out what caused it: who actually decided what happened next (not who should have, who did), where the knowledge lives in practice, and what would still be hard if the loudest bottleneck disappeared tomorrow. The first two moves happen inside every door, and you get the read in a short note you keep.
If you ask us to go after the cause, the people who live with it redesign it with us, and then we build ourselves out: whatever can be systematised and automated is, and we hand it over. The test at the end is simple and slightly brutal: if elegant.work disappeared tomorrow, would the thing keep running, would your people understand it, and could another competent person improve it? If not, we're not finished.
Where AI sits
AI comes after the read.
The first move is never where can we put AI? The first move is what is actually happening here, and what does it need? AI is in the stack and often the right tool. Used before anyone has looked at how decisions get made, it speeds up whatever is already going wrong.
Start with what looks broken.
We follow it back to where it's made and tell you what we see.