The Startup Way

Synopsis

To thrive in the coming century, every organization needs the ability to experiment rapidly with new products and new business models; to empower their most creative people, and to engage repeatedly in a process of innovation in order to unlock new sources of growth and productivity. A modern company must give every employee the opportunity to be an entrepreneur, recognizing entrepreneurship as a core discipline. The lessons of the lean startup movement of Silicon Valley are the basis of The Startup Way, a series of tools and techniques that will unleash this entrepreneurial approach in any organization. The core of the approach is the small, internal startup team, a cross-functional group that is focused on testing leap-of-faith assumptions about potential new products with the use of minimum viable products. A repeated cycle of testing leads to an iterative process where failure becomes a key component in validated learning. Making use of growth boards and metered funding, the teams learn how to pivot or persevere on each project. Embedding the entrepreneurial function in the organizational structure, creating career paths and assessment processes that value innovation, and making use of innovation accounting, ultimately leads to an organization that embodies the concept of continuous transformation.

Preface

The Startup Way — released in early-October 2017 — is the continuation of the award-winning The Lean Startup — both written by Eric Ries. The Lean Startup introduced "lean manufacturing" techniques into the innovation community and is credited with institutionalizing "agile methodologies," "lean processes," and "A/B testing."

The Startup Way builds on the same topics and introduces new techniques to identify and develop cutting-edge products within large organizations. The book introduces: a "Metered Funding" model to add resources and funding to successful pilots, a model to create an entrepreneurial culture within an organization based on meritocracy, among other techniques. These techniques will shift your organization's culture towards one that is focused on innovation and positive financial outcomes.

Summary

The five principles

The principles of entrepreneurial management described in "The Lean Startup" (published in 2011) can be applied in any industry, size of company, or sector of the economy. Whether it is a large, established company like GE or a tech startup experiencing hyper-growth that wants to scale beyond their first, successful innovation, any organization can follow the same set of principles to find new sources of sustainable growth.

The "Startup Way" is a management system based on five principles:

1. Continuous innovation that repeatedly finds new breakthroughs.
2. The startup as an atomic unit of work.
3. Entrepreneurship as the missing function in the organization.
4. Unleashing entrepreneurship as a kind of second founding.
5. Continuous transformation that rewrites the company's DNA.

Creating a modern company

To thrive in the coming century, every organization needs the ability to experiment rapidly with new products and new business models; to empower their most creative people; and to engage repeatedly in an innovation process in order to unlock new sources of growth and productivity.

1. Old-fashioned vs. modern companies

The world has become an incredibly unpredictable place and today all organizations operate in a marketplace of uncertainty. Time and again, business leaders and managers voice concerns about new global competition, the speed with which automation and information technology are rendering products and processes obsolete, and the onslaught of potential high-growth startups impacting every industry. In addition to these external sources of uncertainty, today's managers also face the pressure of having to constantly launch new products, seek new sources of growth, or enter new markets.

At the same time, most organizations are operating a system of accountability that was designed in a very different time and context, with the goal of producing high-quality products on time, on budget, and at scale. The buzzwords were things like "standardization," "mass production," and "lean manufacturing." In this old system, it was deemed important to meet a pre-decided forecast and failure was not an option. In fact, failure could be avoided with careful planning and proper execution.

A modern company, however, must have both the capacity to produce quality and reliable products and the ability to discover what new products to produce. It must give every employee the opportunity to be an entrepreneur.

An old-fashioned company is founded on steady growth, has experts in specialized functional silos, and operates huge programs. It uses internal functions like legal, IT, and finance to mitigate risk through compliance with detailed procedures. An old-fashioned company prioritizes all projects based on return on investment, traditional accounting, and market share. It is full of multitasking and meetings, with lots of middle managers, and a hierarchy of managers and subordinates. The company tends to pursue big projects, everyone is busy all the time, and "failure is not an option." Barriers to entry protect an old-fashioned company from competition.

A modern company is founded on sustained impact via continuous innovation, and operates rapid experiments. It uses internal functions like legal, IT, and finance to help its employees meet their goals of serving customers. A modern company tries to maximize the probability and scale of its future impact and uses innovation accounting. It uses the internal startup, where a small number of passionate believers are dedicated to one project at a time, and is organized around leaders and the entrepreneurs they empower. The company pursues a portfolio of smart experiments, where efficiency means figuring out the right thing to do for customers, and "productive failures" are rewarded. A modern company leaves competitors in the dust through continuous innovation.

2. The entrepreneurial function

Established organizations usually lack the ability to act on new ideas, being encumbered with layers of bureaucracy that have built up over many years or even decades. No-one is explicitly in charge of grappling with uncertainty. Even hyper-growth startups can end up with big company structures. Instead, a modern company should recognize entrepreneurship as a core discipline.

Atomic unit of work

Not everything faced by a modern company can be managed by an internal startup unit, but it is the best way to respond to uncertainty. These internal units blend elements of research and development, sales and marketing, and engineering; they have no logical home in a traditional org-chart. The responsibility of the entrepreneurial function is to oversee these internal startups.

The atomic startup unit is a dedicated team that relentlessly pursues new ideas, stays true to the experiment, and is flexible enough to pivot when needed.

A new style of leadership

The entrepreneurial function also supports other functions in the organization in doing their work more effectively. Traditional management tools are focused on planning and forecasting. Identifying and managing entrepreneurs requires a new style of leadership. It is particularly important to realize that 'entrepreneurship' is not some special quality possessed only by a few people. In fact, you never know who the entrepreneurs are going to be; and even the non-entrepreneurs will benefit from this new way of working.

To take advantage of its latent pool of entrepreneurial talent, the organization has to make the entire employee base aware of the possibilities of entrepreneurship as a career path. This means meeting a series of challenges:

1. Creating space for experiments but with liability constraints.
2. Learning to make investments on the basis of evidence, experimentation, and vision, not just ROI forecasts.
3. Creating milestones that can work even when there's no accurate forecast.
4. Providing professional development and coaching to help people get better at being entrepreneurs.
5. Provide internal and external networks so that people know what it means to say, "I'm and entrepreneur."
6. Recognizing that high risk and uncertain projects need a separate and rational way to attract talent.
7. Creating new incentive and advancement systems.

These are significant challenges. Fortunately, the structures and systems of Silicon Valley can provide the answers.

3. Lessons from silicon valley

Silicon Valley can best be described as a state of mind – a shared set of beliefs and values that have taken hold in dozens of startups around the world. The startup movement's beliefs, systems, and structures can be replicated in other organizations.

The importance of the team

Silicon Valley investors make their decisions largely based on the quality of the team, looking first at the people and then at the idea. They see the team's ability to come up with a good plan as an indicator of future success, even if the plan itself changes. What counts is the team's ability to execute.

In addition, small teams are the most powerful, with members forming an intense bond and communicating easily. The team is adaptable; it's almost impossible for bureaucracy to take over when the team only has a handful of members. And, almost by definition, a small team means a paucity of resources, which forces everyone to focus.

Finally, the small teams that create startups are inherently cross-functional. Everyone has to pitch in and solve the problems that come up.

Start with the customer

The team has to start with articulating the problem to be solved from the customer's point of view. Customers don't care about market-share, they just want something that makes their lives better. Silicon Valley knows that the key word here is "better." It's not enough to just give customers a solution to their problem; the goal is to delight customers with a dramatic improvement.

Give employees a stake

Startups are usually for-profit companies. Nevertheless, an important component of the Silicon Valley ethos is to give every employee a stake in the company by offering them equity ownership. This gives employees a direct incentive to learn; it's not a cash bonus, it's a measurement of what the startup has learned about its future profits.

Focus on leading indicators

Concepts like gross profit, ROI (return on investment), and market share are trailing indicators. In contrast, leading indicators predict future success and include customer engagement, unit economics, and repeat usage.

Metered funding from investors

In Silicon Valley, whatever money the startup team raises is theirs to use as they see fit, with minimal oversight. But, without progress there won't be another round of funding. Knowing that a board or group of investors will at some point need a progress report. This report gives the team accountability while also allows them the freedom to pursue their goal. Linked to this is the role of the board, which expects a report not on a fixed schedule, but when there is something to share.

Meritocracy

One of the most widely-held Silicon Valley beliefs is that good ideas can come from anywhere, and that people should be given resources based on their talents, not on their pedigree. Related to this is the notion that, unlike in a traditional company, in a startup not everything has to be "figured out" before you can proceed.

Experiment

A small team, operating on a meritocratic basis and backed with metered funding, can create experiments to try things out, without causing financial ruin for the larger organization. A culture that tolerates failure allows the organization to pursue a diverse range of ideas. Many may be dreadful, but a few will be truly groundbreaking.

Driven by the mission

Silicon Valley is full of great visionary founders; they are an essential element in any startup. It is the vision that gives the team its guiding light and purpose, providing a deep sense of motivation and energy. Crucially, it also allows the team to pivot – to change strategy without changing the overall vision.

Entrepreneurship as career

The 'founder mentality' is lauded in Silicon Valley; early employees at successful startups are sought after at other organizations and are given the kinds of opportunity for rapid advancement not typically found elsewhere. The entrepreneur is recognized as someone who can make things happen in a range of situations.

These lessons from Silicon Valley give us a common language to talk about the management practices that are the basis of the modern company. With the language and vocabulary established, we can now talk about the methods that are the basis of the startup way of working.

4. Lean startup tools and processes

What are the basics of the Lean Startup method?

Leap of faith assumptions

Identify the beliefs about what must be true in order for the startup to succeed. In a traditional business, these assumptions are the company's guess as to how its strategy will deliver on its vision. In a startup, these assumptions need to be made explicit, but keep it simple; list only those leap of faith assumptions that will have the most impact on the success or failure of the business plan.

Any startup will have two particular leap of faith assumptions that need to be tested: the value hypothesis (whether a product or service really delights customers once they start using it;) and the growth hypotheses (once the product has some customers, will it be able to get more).

Figure: "Determine Value & Growth Hypothesis" — Page 95.

Minimum Viable Product

Create an experiment to test the leap of faith assumptions as quickly and cheaply as possible. This Minimum Viable Product or MVP needs to be a real-life product that creates the maximum opportunity to be surprised by customer behavior and so allows the team to collect validated learning. The goal is to quickly turn an idea into something real, even if it is imperfect, in order to learn. It is not a first step toward scaling.

MVPs can take many forms, depending on the idea you want to test. It could be as simple as an online landing page designed to elicit a customer response; this can test marketing messages or be used to test how customers would respond to potential new product features. A pop-up shop is another form of MVP; a physical store or booth that allows you to interact directly with potential customers. However, it is very important to brainstorm multiple MVPs for any given project.

Validated learning

Think like a scientist. Treat each experiment as an opportunity to learn what is and is not working. The data gleaned from each MVP should lead to a report that has three components.

1. It should be actionable: demonstrate a clear cause-and-effect that is related to changes in the product itself.
2. It should be accessible: everyone involved in the project should have access to the report and be able to understand it.
3. It should be auditable: in other words, the data should be credible.

Build-Measure-Learn

Now, take what is learned from each experiment and start the loop over again. Building an MVP is not a one-off event. Once completed, the data will show where the idea has traction and where it doesn't. With this information in hand, build the next MVP and keep learning. In this way, instead of a quest for perfection the focus shifts to a willingness to experiment and adapt the original idea, which eventually will lead to a better product.

Figure: "Build-Measure-Learn Loop" — Page 105.

Pivot or persevere

Testing assumptions and learning from MVPs lay the groundwork for the key final step in the startup process. On a regular schedule, make a decision about whether to make a change in strategy – pivot – or stay the course – persevere. The decision to pivot may mean aiming at a different market for the product, or developing a different feature of the product, but it does not change the overall vision for the product. Each pivot creates a new set of assumptions to test, renewing the process all over again. It is important to schedule the pivot or persevere decision meeting in advance, say once every six weeks, so that everyone is focused on asking themselves, "Is our current strategy taking us closer to our vision?"

Nearly every successful startup has pivoted somewhere along the way, but not only startups can pivot. Netflix went from an established DVD mail rental service to streaming content. PayPal started life as a money transfer mechanism for palm pilots and is now a world-wide web-based payment system.

5. Management for innovation

Entrepreneurial management does not replace traditional management; rather, it is a leadership framework designed for twenty-first century uncertainty. Although innovation is decentralized and unpredictable, it can still be managed. Doing so just requires different tools and safeguards than found in a traditional setting.

Accountability, Process, Culture, People

The systems, rewards, and incentives that drive employees make up the organization's accountability; in other words, what are employees compensated for, rewarded for, and celebrated for? An entrepreneurial mindset must be recognized and rewarded.

Process means the tools and practices that employees use every day to get their work done. These habits and ways of working become the organization's culture, its institutional muscle-memory. And, that culture in turn attracts a particular kind of person.

Transformation outcomes

As these tools become used throughout the organization, a number of changes will occur. The existence of small startup teams creates more opportunities for leadership, and innovative people will be more likely to stay within the company. There is less waste of time and energy as management figures out in advance the best things to build, without spending significant resources on dead-end projects which can be killed off more quickly. Once failing with honor is seen as a skill, such 'failed' projects can be treated as the groundwork for future successes.

When the ability to experiment, learn, and pivot is embedded in the company culture, problems can be solved more quickly and efficiently, too. Ultimately, all of this will add up to more profit for the company.

How, exactly, to transform the paradigm of the traditional company into the entrepreneurial approach of the modern company is the focus of the next section.

How to transform the company

There are three phases to the company transformation: laying the foundation and creating the critical mass; rapid scaling and deployment; and dealing with the deep systems of the corporation. Each of these phases plays out across the different levels of the organization: the team, the division, and ultimately the enterprise as a whole.

Phase one: critical mass

In Phase One, the overall goal is to build enough critical mass to get senior leadership to buy into rolling the approach out company-wide. At the team level, this will mean figuring out what does and does not work for your particular organization. At the division level, it will mean enlisting a small team of senior management 'champions' to make exceptions to company policy as needed. At the enterprise level, building critical mass means getting agreement with the most senior leaders on what success looks like; focusing on leading indicators; and establishing the criteria to move to Phase Two.

Start with a limited number of projects and build from there. Create dedicated, cross-functional teams to undertake these pilot projects, and a growth board system to make quick, clear decisions about the projects presented to them. Teach these teams how to design startup experiments and how to measure the results. Finally, translate these new concepts into company-specific language and tools.

Start small

The scale of the program will depend on the size of the company, but it is important to start with a limited number of small teams, to create an iterative process that will gradually expand.

Dedicated cross-functional teams

A cross-functional team will harness the energy from various disciplines within the organization. Allow this functional diversity to grow over time so that team members become 'functional ambassadors' who can explain their role in terms that other team members can understand.

The golden sword

Wielding the Golden Sword is another way of describing the role of senior leaders in cutting through the bureaucracy in one stroke. It encourages the team members to ask for what they really need to move forward; usually, in the form of cover and clearing away obstacles.

A good experiment

Designing a good experiment is key. It should have:

• A clear, falsifiable hypothesis
• An obvious next action
• Strict risk containment
• A tie between what is measured and at least one leap of faith assumption

New measurements

For teams to know that they are succeeding, Phase One requires the development of new forms of measurement, using leading indicators that measure validated learning. Leading indicators come in many forms but their common purpose is to track signs that the process is working at the team level. This could include showing development of a faster cycle time, or increased customer satisfaction and engagement.

New metrics will be needed to measure these indicators; the key here is to keep the metrics simple and focused on validated learning. One example, taken from an IT division, might have just four metrics to measure the success of a new project:

• How fast can a team get a new task done?
• How many tasks can a team complete in the course of one work cycle?
• How long does it take a task pulled out of the production backlog to get back into production?
• How long does a task sit in the backlog?

Sponsors

Every startup team needs a sponsor, someone in the company leadership who can resolve the tough problems the team will encounter and clear away obstacles. The sponsor can make exceptions to usual practice and policies, and make sure that progress isn't stymied by conflict and a clash of systems. Having a sponsor is also reassuring to team members and middle managers who may be nervous about the changes they are trying to implement.

Own the process

An outsider pushing an organization to change is doomed to failure. Instead, early in the transformation the company has to make the process its own. This means making sure that a person from within the organization leads the change. It also means finding ways to adapt the tools and techniques described here to fit the specific company, to translate the startup way of working into terms that make sense for the organization.

Phase two: scaling up

The goal of Phase Two is to build organizational clout in order to tackle the difficult problems that will arise in Phase Three. At the team level, scale up the number of teams, build programs and accelerators as needed, and make sure to include all divisions, functions, and regions. At the division level train all senior leaders, event those not directly responsible for the entrepreneurial function, so that they are literate in the new approach. Scaling up at the enterprise level means developing coaches, a company-specific playbook, and new finance and accountability tools.

Scaling up looks different for each organization, but there are some common patterns and tasks that come with this Phase.

Identify The challenges

Review and identify all the challenges faced by Phase One teams and projects. This includes listing all the exceptions that had to be made for the teams to kick off their projects, along with detailed information on why some projects failed.

Widespread rollout

In Phase Two, develop and implement a system for working in the new way. Roll out the system across the company. It is crucial to share information about the new methods throughout the organization.

Executive level champions

Identify and use executive-level champions to reinforce the new methods. Different from a coach or an executive sponsor, the executive-level champion's primary function is to clear obstacles that crop up for teams as the startup way of working spreads through the organization, advocating publicly and effectively for this way of working. Clear executive authority and support will encourage middle managers who might otherwise fear what appears to be a breakdown in traditional processes and procedures.

Train representatives of internal functions

Bring internal functions into the transformation process by training people in the new way of working. The reality is that there will be backlash; make sure that every internal function is involved in the training program, and include participation at the executive level, to minimize this backlash.

In-house coaching

Create an internal coaching program, a cadre of individuals who can help teams to make the mental shift to the new way of thinking. This is a great way to build internal support for the transformation. Whatever the specifics of your coaching program, make sure that the coaches are more than just occasional participants, that they receive rigorous training, and that coaching is elevated to a vital position within the company.

Growth boards and metered funding

The traditional way of funding projects is 'entitlement funding.' Once a project is approved the cash spigot is always turned on, with an assumption that projects will continue to be funded year after year. When teams feel entitled to funding, they are unlikely to operate with energy or focus; it becomes easier to delay the launch of a new product, to make sure there is no chance of failure.

In contrast, metered funding gives teams the freedom to spend the money, with strict criteria that must be met to unlock more, and an emphasis on validated learning. It enforces a scarcity mindset, is conducive to cross-functional collaboration, and reduces middle-management interference. To be most effective, metered funding should be paired with growth boards – groups of people to whom the teams are accountable and that sign off on the next round of funding. This creates a direct relationship between the financing of a project and its progress.

Figure: "Metered Funding" — Page 218.

A growth board is the internal version of a startup board: a group that meets on a regular basis to hear from the team about its progress and to make funding decisions. It is a single point of accountability that pushes team members to think about their progress and question whether they have really achieved validated learning. It also acts as the clearinghouse for information about the startup for the rest of the corporation, and it provides the metered funding for the project.

The only way to give these changes staying power is to use early successes to tackle the deep institutional systems of the company. In Phase Three, the organization grapples with its incentive structure, how people are held accountable, and how resources are allocated.

Phase three: deep systems

In Phase Three, the overall goal is to build an organization capable of continuous transformation. At the team level, "this is how we do it" tools and training are made available throughout the organization, not just for those working on high-uncertainty projects. At the division level, creating deep systems for entrepreneurship means establishing growth boards, innovation accounting, and strict accountability for all senior leaders to allocate resources to the change. At the enterprise level, Phase Three means tackling the company's hardest deep systems: compensation and promotion; finance; resource allocation; supply chain; and legal.

Because every company and organization is unique, every deep system transformation is different and the patterns are not as common. The following stories illustrate what Phase Three may look like.

Always building: Airbnb

Airbnb had a second founding with the launch of its Trips product; but the launch was not immediate. The company knew it had found its next big idea but the project languished until a small team of designers, product people, and engineers went to New York and ran a three-month internal incubator program. They tested numerous ideas and went back to San Francisco headquarters with a proposal. The company formed a cross-functional team to work on the project, with team members hanging out at Fisherman's Wharf to ask questions of their customers. The team spent two years growing the trips technology.

The approach was so successful that the company launched Samara, an in-house innovation and design studio made of up designers and engineers, that will help ensure the company's continued growth and evolution.

Human resources: GE's employee management system

A team at GE was making rapid progress in getting a new product to market but the project ran into a brick wall – the Employee Management System (EMS). Under the EMS every engineer had an annual goal to work toward that was evaluated in EMS based on a functional matrix for the particular position. The new project entailed increased rework, something that would be rated negatively in the EMS. Furthermore, EMS was based around an annual review cycle, something that was at odds with the new approach of experimentation and validated learning. The established human resources system wasn't designed for a method of working that entailed extreme uncertainty.

So, GE launched a new evaluation program, and it did so through testing MVPs and validating learning. The initial assumption was that employees would want to give feedback upward to managers and across to fellow colleagues. However, testing revealed that in practice they were uncomfortable with doing so. The team tasked with creating the new program pivoted and instead focused on the behaviors and culture that would need to be in place to make the new evaluation system work. In two years the team changed the performance management system from a prescriptive, formal, annual process to an approach that provide a framework within which teams can operate. The emphasis now is on learning, honesty, and outcomes as measures of success. They were able to get HR to function like a startup.

Company-wide innovation: Intuit

Innovation projects in a large organization can have a high mortality rate; but, the projects that do survive can have a dramatic impact. In 2013 the CEO of Intuit was the host of the annual American Heart Association benefit. He put together a small team of two designers, an engineer, a product manager, and an innovation leader, and asked them to help him host the most successful fund-raising event that AHA had ever held.

The team came up with a mobile app that the volunteers at the event could use to keep track of all the ways that money was coming in. The app was connected to a screen projected in the main room that showed the total amount raised. Whenever anyone gave money, the numbers on the screen ticked up in real time. The goal for the event was to raise $1 million. When the total stood at $947,000 near the end of the event, the auctioneer was able to urge people to contribute more, to reach the goal – people could see their pledges making a difference. By the end of the night, the event had raised $1,170,000 – the most successful ever fundraiser for the AHA. The AHA went on to adopt the app for all of its regional benefits, and the Intuit team made the technology self-service, so that it could be used by any organization free of charge.

Innovation accounting

Traditional companies ensure accountability by requiring managers to meet (or beat) a forecast. This does not work in a system based on testing, repeated failure, and validated learning. Instead, companies will have to adopt a system of innovation accounting – a way to evaluate progress when metrics like revenue, ROI, and market share are effectively zero.

Innovation accounting ties long-term growth and R&D into a system of three levels, providing a clear process of funding for innovation.

Level one: simple dashboard

The first level is the dashboard of metrics that teams agree are important. The dashboard should answer key questions, such as:

• Execution: did we do what we said we were going to do?
• Behavior change: are our people working differently?
• Customer impact: do customers (internal or external) notice an improvement?
• Financial impact: are we unlocking new sources of growth?

The dashboard gives a basic sense of what's working and what isn't.

Level two: business case

Level Two of innovation accounting focuses on the business case and calls for building a more detailed dashboard that represents the complete interaction with the customer, from when they first hear about a product to when they actually use or purchase it. The Level Two dashboard must encompass the value hypothesis – what specific customer behavior indicates delight with the product? – and the growth hypothesis – what specific customer behavior will cause us to acquire more customers?

Level three: Net Present Value

The goal in the final level of innovation accounting is to translate learning into dollars, by rerunning the full business case after each new data point. Take the basic business model spreadsheet, something that shows how certain customer behaviors aggregate over time and result in a positive future impact. Then rerun that initial spreadsheet with new numbers learned from experiments and see how things change. Each new run of the spreadsheet yields a new graph and a new set of projections that can then be translated into net-present-value (NPV) terms using standard finance tools.

Figure: "Level 3 Dashboard" — Page 279.

The global view

The goal of all these tools and techniques is to move the organization from a state of continuous innovation to one of continuous transformation; an ongoing cycle of change that transforms not just a project or a team but the structure of the enterprise itself.

Every organization should have an active program of experimentation in new organizational forms and methods, peopled by those who will become the founders of the next company-wide transformation. These people will need a skill set similar to what is needed to build a new startup from scratch. They need career paths, accountability, and a rigorous training system. This engine of continuous transformation should be seen as a permanent organizational capability, one with responsibility for the entrepreneurship function.

Role of public policy

How can we run policy-making experiments that will help leaders to create the conditions in which the next generation of entrepreneurs can thrive? To foster startup-driven economic growth, we need the pro-business policies usually associated with conservative politics – less regulation, more competition, more public-private partnership – combined with the kinds of pro-worker policies usually associated with the political left – workplace protections, portable health insurance, and sensible immigration. Add into the mix non-partisan policies like patent reform, open data, and more responsive government, and you have a pro-entrepreneurship public policy stance.

Entrepreneurs are motivated by a desire to make the world a better place. Any policy that helps them to take the first steps toward their goals will reap returns. And, given that most experiments will fail, it is also important to cushion people from the worst consequences of that failure, so that the overall rate of entrepreneurship can increase. This might mean a more rational health insurance system that removes uncertainty about coverage; introducing entrepreneurial skills into the school curriculum; and encouraging the influx of global talent through immigration. It might mean trying to create a pro-productivity trades union structure; allowing people to convert unemployment insurance payments into a business loan to help launch a new company; or setting up a government-run microloan system. It may also involve creating a new stock exchange that is focused on the long term, in order to fund a company that will last for generations. In sum, a pro-entrepreneurship public policy should nurture human capital over the long term.