Most years, the floor of Death Valley is gravel.
Known as the hottest, lowest, and driest place in North America, the desert holds the record for the hottest air ever measured on Earth.
But this spring, it went yellow.
Fields of gold appeared along Badwater Road in the biggest bloom the park had seen for a decade.
The best part? The seeds had been sitting in the gravel the whole time.
They were waiting for a condition: a storm heavy enough to strip the seed coating, then a cool, wet winter.
Once the conditions changed, everything changed.
Which is why the question I ask about every stuck problem is: what would have to be true for this to move?
Last week Andrew Youn, who co-founded One Acre Fund and later Prevail, argued that donors and programme leaders think too small. He wants Big Goal projects: large in scale, aimed at a defined outcome. Think: end lead poisoning, every child reading by ten.
I agree that this level of ambition is long overdue.
But the recipe is incomplete. Scale and specificity tell you what to aim for. They don't tell you what to do on Monday morning. For that, you need a mechanism — an understanding of what's holding the current outcome in place, so you know what to change.
I call this the Big Lever: the condition that, once moved, changes what becomes possible everywhere else.
Take lead poisoning.
For decades, there was no shortage of blood tests, awareness campaigns or advocacy. All useful. But if the rules still allowed lead in paint, the system kept producing lead exposure.
Change the law, and people behave differently without anyone persuading them one by one. The Big Lever was regulation.
Or take HIV treatment.
Last week I wrote about a year of treatment falling from around $10,000 to $90. The Big Lever was price.
In my own work, 56 countries now pay and accredit community health workers as part of their national health systems, after a century of treating them as volunteers. The Big Lever was policy.
Yet here's the pattern I keep seeing:
When the work isn't moving, we tend to add effort to the existing system, rather than try to change it.
More programmes, more meetings, more messaging, more partnerships. More people trying harder.
But sometimes the system isn't asking for more effort, it's asking for different conditions: different rules, lower price, new forms of accountability.
Yet this is exactly the work most plans dismiss as not tractable, not cost-effective, and not clear enough.
Nobody gets an annual-report photo of a procurement clause that took three years to change. Nobody gets a receipt for a regulation moved by six organisations and one stubborn civil servant.
So we fund what looks like action—and mistake activity for movement.
We call systems work risky because the outcome is uncertain.
Then we pour money into interventions that are safe to describe but cannot change the condition producing the problem.
That's more seed on dry gravel. It looks like progress, but nothing comes up.
Try this
Take the outcome you most want to change.
Don't write the intervention, the campaign or the partnership.
Finish this sentence:
This stays where it is because ______ still allows it.
Then ask:
What could make that condition different?
If the answer is a rule, price, budget line or procurement decision, you've probably found something worth investigating.
If your answer is "because people don't understand," keep going.
What makes it rational for them not to understand?
If your answer is "because nobody has the resources," keep going.
Who controls the budget?
If your answer is "because nobody will agree," keep going.
What rule, incentive or risk makes agreement hard?
Look for the smallest condition that, if changed, would make a lot of other things easier.
🌿 The seeds were already there. Find the condition that lets them grow.