This article was published by McKinsey & Company in honor of the Skoll World Forum.

In a recent speech, World Bank President Jim Yong Kim noted that delivery is a major challenge for leaders across the world.

We need a science of delivery, he argued. No leader of the World Bank has ever made a more important case. And the good news is that the science he wants is ready to emerge. This is the story of how it began.

In June 2001, Tony Blair was elected to a second term as the British prime minister. A few days later, he asked me to set up the Prime Minister’s Delivery Unit (PMDU), tasked with securing the implementation and delivery of his domestic policy priorities. Delivery meant more than passing laws and writing speeches. It meant changing the facts on the ground and ensuring that citizens could see and feel the difference. Our task was to translate reform into results.

Over the next few weeks, a handful of colleagues and I developed a set of processes that, with refinement over the years, turned out to be a major innovation not just in Britain but globally. By 2005, most of Blair’s domestic-policy priorities, ranging from healthcare improvements and crime reduction to railway performance, had either been delivered or were heading firmly in the right direction. Other governments began to take notice. Since then, there have been numerous attempts to emulate the work of the PMDU. Some have succeeded, while others have failed.

What were our techniques and processes? Among the attempts at emulation, what has differentiated success from failure? What PMDU put in place was deceptively simple. We asked five questions, repeatedly and persistently, until we received satisfactory answers.

Question 1: What are you trying to do?

We asked this question to establish clear priorities (and therefore also to clarify what was less important) and ensure that each priority was paired with a clear definition of success. Our goals were intentionally ambitious; after all, Blair wanted “step change.” We called them “targets,” a word that became controversial. The word is unimportant—what matters is the clear, measurable definition of success. We also decided to make the targets public, which is not necessary but is desirable in an era of transparency.

The result was a set of specific targets in the categories of health, education, crime, and transportation. They included reducing wait times for routine operations, improving the punctuality of trains, boosting literacy among 11 year olds, and many more.

Question 2: How are you trying to do it?

Once the targets were agreed upon, we asked the relevant departments to prepare delivery plans setting out how they intended to meet the targets. This alone was a revolutionary act; white papers, beautifully written and then shelved, were then the norm. We wanted real, coffee-stained plans that drove action. Above all, we wanted a trajectory. How would they measure progress over the period of time that would elapse between setting the target and hitting it? This simple request required officials to think systematically about the link between proposed actions and their impact.

Of course, even the best-planned trajectories don’t always turn out to be right. They do allow progress to be monitored. And when reality deviates from the prediction, they enable lessons to be learned.

Question 3: How, at any given moment, will you know you are on track?

Crucially, we established routine ways to monitor progress. We insisted on regular data collection so that officials knew whether progress was being made in close to real time. People often say that this sort of ongoing performance evaluation is not too difficult. This is never true. And yet the provincial government of Punjab, in Pakistan, collects data on performance from 60,000 schools every month, in a low-tech, affordable way. As a result, it has transformed performance.

We introduced quarterly monitoring meetings, which we called “stocktakes,” between the relevant ministers and Blair, advised by my team and me. In each stocktake, we examined the data, had an honest conversation, and made decisions, which again was revolutionary. Most governments spend their time running to catch up with crises and events. These routines changed all that. They put data rather than spasm at the heart of decision-making. Punjab’s Chief Minister Shahbaz Sharif, with whom I work now, has done the same, to good effect.

Question 4: If you are not on track, what are you going to do about it?

Once you have proper routines in place and working, problems are identified before they become crises. In my experience, problems can always be solved. Some problems are relatively simple to fix; others are much harder. For the latter, what matters is that you try something—and if that doesn’t work, try something else, and keep trying until you get a result.

When there is a problem, the first instinct in government (encouraged by the media) is too often to allocate blame rather than solve the problem. Always solve the problem first; if blame needs to be allocated, that can be done later.

Question 5: Can we help?

The PMDU didn’t just monitor the performance of government. It also rolled up its sleeves and helped solve problems. When it succeeded, it congratulated the relevant department rather than taking credit for itself. We never yelled at people, West Wing style. Instead we built trusting relationships. We took the view that we shared responsibility for the outcomes.

And we developed techniques that could help solve problems—rapid reviews and delivery-chain analysis, for example. Crucially, we were persistent; we wouldn’t go away until a problem was solved. We were ambitious, too, however tough the present might have looked.

If successful delivery is so simple, why is it hard to replicate? Those that fail tend to see delivery as a passing management fad. They don’t make the commitment to change facts on the ground. They don’t make the routines work. They might listen to “experts” who use the buzzwords and promote the form of a delivery unit, but these people generally don’t understand the philosophy or discipline on which success depends. Successful leaders, such as Former Ontario Premier Dalton McGuinty, Malaysia Prime Minister Najib Razak, and Punjab Chief Minister Shahbaz Sharif, take a very different approach. They prioritize, they persist, and they appoint talented people to focus on delivery. They understand that while getting the policy right is hard, it is only 10 percent of the challenge. The other 90 percent is the blood, sweat, and tears of relentless implementation. They understand that delivering results is not about setting up a delivery unit; it’s about fundamentally changing how you do business.

As Jim Yong Kim says, delivery can become a new science. As more governments see the urgency of delivering results that go beyond the incremental, we can learn quickly what works and what doesn’t and unleash the science that will change millions of lives for the better.

Editor's Note: This collection of articles is part of “The Art and Science of Delivery,” an anthology of essays published by McKinsey & Company in honor of the 10th Anniversary of the Skoll World Forum. It is the most recent installment of McKinsey's ongoing series, Voices on Society, which convenes leading thinkers on social topics. (Copyright (c) 2013 McKinsey & Company. All rights reserved. Reprinted by permission)