Mathematician Hannah Fry has a wonderful way of explaining "phantom traffic jams," the ones with no accident, no closed lane, no visible cause at all.

They form because the congestion isn't fixed to one spot. It's a wave that travels backward through the traffic. One driver taps the brakes. The driver behind brakes a little harder. That ripples backward until cars a mile back are motionless, for a jam that, by the time they reach it, has already dissolved.

No one caused it. Everyone did.

I've watched phantom jams happen across corporate workflows several times in my career. Employee surveys across many of the organizations and industries I've worked with have one thing that comes up in their feedback consistently - " please improve the way we work by removing friction".

I think about this every time someone asks why a seemingly simple decision at their business takes six weeks and four approvals. Most of the time, it rarely started as a policy., but rather as a single, reasonable caution:

• A finance team adds a second sign-off after one bad expense report, and five years later a $200 purchase needs three approvals nobody can explain.

• A legal team, stung by one contract dispute, adds a review step "just for this kind of deal," and it quietly becomes the default for every deal.

• A security team, after one incident, adds a check, and each team downstream, wanting to look equally careful, adds one more on top.

• A marketing team has one off-message email go out, so every email, social post, and slide now needs sign-off from brand, product marketing, and regional leads. The original mistake is long forgotten. The review stays.

Here's where the traffic metaphor needs updating.

A phantom jam travels backward down one lane: linear, traceable. A business workflow moves through a network, where finance touches legal touches procurement touches security touches the business unit, and a single brake-tap in one node ripples outward through all the others at once.

That's why organizational phantom jams are so much harder to diagnose. The slowdown you feel in week six may have originated two departments away, three reorgs ago, for a reason nobody at the table remembers.

Which brings me to a distinction worth making before you tear down every checkpoint: not all slow is the same.

Good slow is the checkpoint that exists because something real is being protected, like a compliance review that prevents genuine legal exposure, or a second sign-off on a decision that's hard to reverse. Good slow is a deliberate choice, and you can name exactly what it defends against.

Bad slow is the checkpoint nobody can explain anymore. It was a reasonable brake-tap once, for a risk that may no longer exist, inherited by every team downstream because removing it feels riskier than keeping it. Bad slow doesn't protect anything. It just accumulates.

The test is simple: can you name, specifically, what a step protects against? If yes, that's good slow, so keep it and maybe even reinforce it. If the honest answer is "I'm not sure, but it feels safer to leave it," that's bad slow.

I once watched a team go through exactly this.

They wanted campaigns out the door faster, but even a simple email nurture was taking weeks to launch. The easy move would have been to cut approvals across the board. Instead, they mapped every step and asked one question of each: what does this protect us from?

• The legal review of customer claims and statistics protected the company from promises it couldn't prove. Good slow. It stayed, and got faster with a pre-approved library of claims, so reviewers only saw what was genuinely new.

• The brand review of core messaging protected consistency across markets. Good slow. It stayed, but moved to the start of the process as a shared messaging framework.

• The multi-person sign-off on every social post traced back to one off-tone tweet years earlier. Bad slow. It was replaced with clear guidelines and one accountable owner.

• The long campaign brief, inherited from an old reorganization, had fields nobody read. Bad slow. It was cut to the few that drove decisions.

Campaigns got faster not because the guardrails came down, but because the jam was cleared at its source.

That's the lesson for anyone leading change. Transformation isn't about going faster. It's about knowing which slow to keep, which slow to clear, and tracing each one back before you decide.