Ch2 02: When the Hurricane Won’t Wait for Your Approval Chain#

Hurricane Irma was bearing down on Florida in September 2017. Category five. Winds topping 180 miles an hour. Millions of people were evacuating, and every gas station, grocery store, and highway was slammed.

Tesla owners in the evacuation zone had a specific problem: range. Some of our vehicles carried software-limited battery capacity — the hardware could hold more juice, but the customer had bought a lower-range configuration. Under normal conditions, that made clean business sense. These weren’t normal conditions. People were trying to drive hundreds of miles north with their families, and some of them didn’t have enough range to make it without a charging stop — at Supercharger stations that might have lost power.

Karim Bousta, running Tesla’s service operations in the region, had a call to make. The standard process for unlocking extra battery capacity wound through multiple approval layers. Software configuration decisions weren’t something a regional service manager was cleared to make on his own. Following the chain could eat days.

He didn’t have days. He had hours.

Here’s the mental framework Karim ran, though he probably wouldn’t have described it this way in the moment:

Will this kill someone? No. Unlocking battery capacity carries zero safety risk. If anything, it increases safety by giving customers more range during an evacuation.

Will this break the law? No. The vehicles already had the hardware. The software unlock was a Tesla-internal business call, not a regulatory issue.

Will this cause catastrophic financial loss? No. The cost of temporarily unlocking a few hundred vehicles in a disaster zone was a rounding error against the company’s revenue.

Three gates checked. None tripped. Decision: act now, report later.

Karim unlocked the extra range for every Tesla in the evacuation zone. Customers who’d bought standard range suddenly had access to the full battery. Some of them got out of Florida on that margin.

What Karim did looks like individual heroism. And it was — he showed judgment, guts, and decisiveness under fire. But the deeper lesson isn’t about Karim the person. It’s about building an organization where hundreds of Karims can make the same kind of call without needing to be heroes.

The problem with heroism is that it doesn’t scale. If the right move in a crisis depends on having the right individual in the right spot at the right moment, you’ve got a brittle system. What you need is a decision framework simple enough for anyone to run, under pressure, without access to the C-suite.

That’s what the three-condition filter delivers. It encodes the organization’s risk tolerance into a checklist any frontline employee can clear in thirty seconds:

Will this action risk anyone’s life or physical safety?
Will this action violate any law or regulation?
Will this action create a financial loss the organization can’t absorb?

If all three answers are no, the employee is pre-authorized to act and report afterward. No phone trees. No email chains. No waiting for a VP to check their messages.

The pushback I always hear: “But what if someone makes a bad call?” And the answer is: yes, sometimes they will. The question is whether the cost of occasional missteps under this framework is higher or lower than the cost of systematic paralysis under the traditional approval chain.

In my experience, it’s dramatically lower. The standard process doesn’t just add time — it adds the specific kind of time that matters most in a crisis. The minutes and hours burned waiting for sign-off are the exact minutes and hours when the situation is shifting fastest. By the time approval lands, the context has changed, and the approved action may no longer be the right one.

There’s a subtler cost too: learned helplessness. When employees are trained that every significant decision needs a stamp from above, they stop thinking. They stop reading situations. They stop building judgment. They become execution machines waiting for instructions. And when the instructions don’t come — because the boss is on a plane, or the systems are down, or the crisis is outrunning the approval chain — they freeze.

The three-condition filter does the opposite. It tells employees: you are trusted. You have the authority to act inside these guardrails. And in return, you have the responsibility to think.

Making this work requires two things most organizations are reluctant to give: clarity and trust.

Clarity means spelling out the three conditions in concrete, unambiguous terms tailored to your business. “Catastrophic financial loss” needs a dollar figure. “Safety risk” needs real examples. “Legal violation” needs a bright line between actual statutes and internal policies that feel like law but aren’t. Vagueness kills the framework, because vague boundaries become excuses for inaction — “Well, it could be a big loss, so I’d better escalate.”

Trust means accepting that employees will sometimes make calls you wouldn’t have made. That’s not a flaw. It’s the price of speed. And if you hired well and trained well, the net outcome will be overwhelmingly positive.

Guidance#

Build a crisis decision protocol for your team:

Define the three gates. Under what specific conditions must an employee escalate? Be concrete. Dollar amounts. Named safety scenarios. Specific regulations. The more precise you are, the more confidently your people will move.
State the default. Make it explicit: if none of the three gates are triggered, the default is act now, report later. This is the step most organizations skip. They define the escalation triggers but never clearly declare what happens when no trigger fires. The result is ambiguity, and ambiguity produces paralysis.
Practice. Run tabletop drills. Hand your team scenarios — a customer emergency, a supply chain meltdown, a PR crisis — and have them apply the three-condition filter in real time. Debrief. Calibrate. The goal is to make the filter reflexive, so when a real storm hits, nobody has to wonder whether they’re allowed to act.

The best organizations don’t need heroes. They need systems that turn ordinary people into effective decision-makers. The three-condition filter is one of the simplest and most powerful systems I know.