Stop Optimizing What Shouldn't Exist

The First Principles Playbook

The Framework: Understand → Delete → Accelerate → Add Back → Automate

Purpose: To stop optimizing processes that shouldn't exist in the first place — cutting complexity before you ever speed anything up.

Summary: A five-step method for stripping a process to its essentials before you touch it. Adapted from Elon Musk's engineering algorithm and rebuilt for running a business, a team, or a life: understand why something exists, delete what doesn't earn its place, only then accelerate what's left, prove the cut was real by adding some of it back, and automate last — never first.

---

Elon Musk has said it plainly: "The most common mistake of smart engineers is to optimize a thing that should not exist." He built a five-step algorithm to keep himself from making that mistake at Tesla and SpaceX — question every requirement, delete what you can, simplify what's left, accelerate it, then automate it.

Most loan officers skip straight to step four. They buy the faster CRM, add another automated drip campaign, chase a quicker turn time — all without ever asking whether the process underneath deserves to exist. A faster wrong answer is still wrong. It just closes sooner.

The First Principles Playbook adapts that discipline for building a business, a team, or a life: Understand, Delete, Accelerate, Add Back, Automate.

---

Understand

Before you touch anything, know why it's there. Every rule, every step, every recurring task has an origin — usually a person who needed it once, for a reason that may no longer apply. Attach a name to the requirement. If you can't find whose problem it was solving, that's your first clue.

Delete

Remove anything that isn't earning its place. This is the step people resist, because deleting feels like risk and keeping feels like safety. It's the reverse. If you're not deleting enough to occasionally miss something, you weren't cutting — you were tidying.

Accelerate

Only now do you speed up what's left. Not before. Musk's own words: "I mistakenly spent a lot of time accelerating processes that I later realized should have been deleted." Speed applied to the wrong thing is wasted effort with better production values.

Add Back

Musk's own words on this, from the deletion step itself: "You may have to add them back later. In fact, if you do not end up adding back at least 10% of them, then you didn't delete enough." We've made that its own checkpoint rather than an afterthought. If nothing needs adding back, you weren't aggressive enough — you were tidying, not cutting.

Automate

Last, not first. Automation locks in a process — which means it should only lock in a process you've already proven is necessary, minimal, and fast. Automating too early means building infrastructure around something that shouldn't exist. You'll have to tear it back out.

---

Try this week: Pick one recurring step in your process — a report, a checklist, a hand-off. Ask who required it and why. If you can't answer both, delete it and see what breaks. Usually, nothing does.

In the mortgage business, we take for granted what's required. Generally, three, five, or ten years ago, something broke that cost the company money — so a process got built to make sure it never happened again. That process outlives the problem. It slows things down, adds complexity nobody asked for, and quietly adds cost, long after the reason it existed stopped applying. First principles thinking forces you to actually examine why you do what you do, instead of inheriting the answer.

Mortgage is essentially unstructured manufacturing — a production line full of steps nobody has re-examined in years. That's exactly what Lower Complexity, one of our core principles, is built to fix: most of what's slowing you down was never chosen, it was just never questioned. The First Principles Playbook is how you find out which parts of your process you'd actually keep.

---

A note on sourcing: the five-step shape here is adapted from Elon Musk's publicly stated algorithm (question requirements, delete, simplify/optimize, accelerate, automate), not a verbatim copy. We swapped his "simplify and optimize" step for a standalone "Add Back" step — Musk treats adding back as a side effect of deleting too aggressively; we've made it its own deliberate checkpoint. Worth knowing the difference before this goes out under his name.