Roadmaps explain the future nicely. They don’t deliver it.

I’ve sat in enough meetings to know how convincing they can be. Early on in my career, mostly. The plan is clear. The sequence is sensible. Everyone agrees it’s the right thing to do.

And I’m there, doing that familiar thing of biting my tongue and sitting on my hands, watching the operational car crash form in slow motion.

You can see it coming from a mile off.

At that stage in your career, you don’t say much. You learn to nod. You tell yourself there must be context you don’t have yet. You assume the straight line must know something you don’t.

It rarely does.

The thing that eventually fixes the mess almost never follows that plan. It turns up from the side, usually after enough people get tired of pretending the plan is still working.

That’s how progress tends to arrive. Sideways, leaving the process blinking slightly, wondering what just happened.

Most large organisations, and career-focused people, are very good at straight lines.

• Budgets repeat last year.
• Processes smooth out variation.
• Governance exists to keep surprises in check.
• Once you’re on the road, you just keep going straight. All of that is sensible. You need it to keep the place standing up.

The trouble starts when the work changes and nobody redraws the line.

At that point, anything genuinely useful looks wrong on arrival. Incomplete. Inefficient. Difficult to justify in advance because there’s nothing tidy to point at yet.

So it waits. Or it’s ignored.

Eventually, someone intervenes.

These days, more often than not, that someone is me. I come in later, once the crash has already happened, and do the thing I used to sit quietly resisting. I cut across the roadmap. I drop work that no longer earns its keep. I pick up work no one officially owns.

This tends to irritate people who’ve invested a lot of time defending the straight line.

That reaction is familiar too.

The changes themselves are rarely dramatic.

• A workaround because waiting six months helps nobody.
• A report quietly killed because nobody reads it anyway.
• A step skipped because it adds effort but no value.
• A customer problem reframed because the categories stopped matching reality some time ago.
• A small experimental diversion because failing small is cheaper than driving off a cliff. At the time, it doesn’t look clever. It looks disruptive. Untidy. Slightly annoying.

Then it works.

Friction drops. Fewer things bounce back. Decisions start to stick. And only then does the language change. The same behaviour gets a better name. Judgement. Leadership. Culture.

The skid marks are painted over and everyone pretends it was always meant to happen this way.

I’m not pretending this always goes well. Everyone’s seen people charge in and make things worse. But the useful interventions tend to share a pattern.

• The aim is to remove friction, not to be noticed.
• The downside is kept small.
• Something real appears quickly.
• The result gives you evidence rather than opinions. Most progress comes from deciding which bits of work no longer earn their place, and quietly picking up the bits that do, even if doing so upsets the neatness of the plan.

As organisations push harder on efficiency, standardisation, automation, more decisions get frozen into systems. Assumptions harden. Variation becomes a fault rather than a signal.

What gets missed are the early signs that something has stopped working.

I used to sit in those meetings and assume the straight line knew something I didn’t. These days, I’ve learned it usually just knows how to look convincing.

The uncommon sense, stripped back

• Straight lines are comforting, especially when you’re junior.
• Progress usually arrives when someone cuts across them.
• The real risk isn’t that this annoys people. It’s that waiting politely costs more.