You’re sitting across the table from someone you just met, a person clever and articulate, appreciative and gracious. You realize that the two of you went to similar schools, played the same sports, and enjoy the same books. It’s going great, you think to yourself; you’re really hitting it off. You could even travel together; being stuck in an airport wouldn’t be quite so awful if you were with this person.

Is this is a date? Or is it a job interview?

Hard to tell the difference.

The research of Lauren Rivera, associate professor at Northwestern’s Kellogg School of Management, suggests that people making hires often confuse one with the other.¹

It’s counterintuitively intuitive: if you’re hiring someone and your organization doesn’t have an explicit rubric for merit, you’ll use your own criteria. You’ll define merit in your own way, probably defaulting to how much this person is like you. Perhaps you and the candidate had similar educations, extracurricular activities, and hometowns. Or maybe you just click with a potential employee. Isn’t that reason enough to make a hire?

Often that’s the similarity-based case, at least in the consultancies, investment banks, and law firms Rivera researched. Consciously or not, these firms have developed a process of cultural matching in which companies replicate themselves because the people they hire have the same cultural characteristics. How does that work? When you’re looking to hire someone on the basis of an ambiguous notion of merit and the merit is based on being like you, you’re going to hire someone just like you at a deep level, even if she’s of a different sex or ethnicity.

“There’s an assumption that if you’re different in sex or race or nationality, you’ll bring different ideas to the table,” Rivera said to us in an interview,² “but if you look at people who get into the firm, you’re screening on people who participated in varsity athletics in one of ten schools, almost all your firm comes from one of three zip codes in the country, there’s diversity [left] to be harnessed.”

Why does this matter? First, there’s the humanistic, democratic ideal that progress is measured by more people being able to reach their fullest potential. But even if we didn’t care about humanity, we’d have to recognize that diversity is a predictor of innovation. In conversation, Rivera gave a Myers-Briggs example: if you find that ENFPs (Extraversion, Intuition, Feeling, Perceiving) are the ideal workers and hire a bunch of them, you’re going to miss out on developing a project made for ENTJs (Extraversion, Intuition, Thinking, Judging). In other words, the less diverse an organization is—whether in the sense of race, class, gender, or personality type—the less it can empathize with people outside of it and the fewer partnerships it can form.

We are not the first to make such an argument. In an essay published in Fortune in 2013,³ Warren Buffett argued that after 1776, the United States accomplished great things operating at only half capacity, as the country didn’t include women in the economy. If women did insist on working, the Sage of Omaha noted, they were limited to the professions of teaching, nursing, and secretarial work but little beyond that.

This not a new argument: it echoes the work of John Stuart Mill, a principal thinker in the British Enlightenment of the nineteenth century. Mill was a utilitarian philosopher: he thought that the just action was the one that created the most benefit for the most people. Relatedly, he introduced the theory of the marketplace of ideas. Just as an open economy has a marketplace of goods—readily apparent from the souk to the farmer’s market—there’s a similar though less tangible market of ideas you can select from. But what makes for the finest market? The selection of goods, as the growth of department stores, shopping malls, and Amazon suggest. Similarly, a society will offer the best “products”—ideas, opportunities, and ways of life—when it includes the widest selection of ideas coming from the widest selection of suppliers.

If we are indeed in the innovation economy, if indeed the uncovering of new, disruptive products and processes is as important as the optimization of existing ones, we have an incentive to make our organizations as prone to innovation as possible. The more varied the backgrounds that contribute to that organizational marketplace, the greater the capacity of the organization and the greater the empathy we can have across cohorts of potential users. And the greater the diversity of disciplines we bring together, the more hands we have on the elephant.

Yet even if we accept the idea that diversity predicts innovation, we may not act in accordance with that understanding. The way we conflate hiring and dating evidences our blind spots. During the unexamined experience of the job interview, we get the positive feedback of the engaging conversation and feel validated for feeling shared interests. It’s intuitive to want to reward a person we’ve clicked with, especially if we’re going to be spending untold hours and unseen flight delays together, though that may be a decision born of individual rather than organizational interest. What’s more, the replicating that follows hiring a person who shares all one’s interests hems in diversity and thus the market for new ideas. Yet it feels so natural to reward someone with whom we click.

Awareness that unhealthy habits such as firm replication can happen unconsciously—can even feel like the natural thing to do—demonstrates the necessity of being vigilant in how we treat the way we work. The usual in ‘business as usual’ needs to be examined. In this particular case, the replication problem arose partially because people defaulted to “they’re like me” in their validation process rather than following an established, rigorous rubric. Part of the solution, then, is to usher more of the implicit action of the hiring process into explicit prescriptions. In addition, Rivera suggests turning the hiring process into something quantifiable: we could rate each candidate across a range of skills and then track that performance over time.

As we discussed in Chapter 2, a person can plot a more informed trajectory after inquiring into his or her identity. This is also true of organizations, as we discussed in Chapter 4. But just as one must become intimate with one’s mind as well as one’s body in the path of growth, the organization must know not only its motivations and goals but its structure, for as we shall see, the structure of an organization predicts the products it creates. Architecture, then, is the topic of this chapter.

CONWAY’S LAW AND YAMMER

Melvin Conway was an early computer scientist. In 1967 he submitted a prescient essay to the Harvard Business Review, a paper called “How Do Committees Invent?” The paper was rejected; in a blog post on the subject, Conway says that it was turned down on the grounds that he did not prove his thesis. He then submitted the paper to an information technology magazine called Datamation, where it saw publication in 1968. The paper provides a crucial bridge between management and product, between groups of people and the things they make.⁴ Condensed for clarity, the thesis is as follows:

 

Any organization that designs a system . . . will inevitably produce a design whose structure is a copy of the organization’s communication structure.

 

In other, more grandiose words, we can restate Conway’s Law as saying that companies are destined to produce products that mirror their organizational structures. Even more extremely and poetically, we could say that the organization implies the product or, more simply, that the organization is the product.

How can we get to such lofty claims? Because when we think of a product, we tend to think of a single, static event or item: a magazine is released in an issue, a mop is delivered to a store, and a doctor sees a patient. But now that every industry is being pressured to release more and more products, we have more physical touch points with the people interacting with those products. Beyond that, the social web has brought us to a point where humans on either side of the organization-user divide are continuously conversing with one another (and if it’s a good conversation, someone will be listening deeply). It’s most obvious in software and the culture of iteration that has matured with the social web: customers are continually signaling what they do and do not want, and companies are responding to those spoken or measured needs as quickly and effectively as they can. In other words, the customer is interfacing with the organization nearly as much as she is interacting with the product.

BLOCKBUSTER, YAMMER, AND PETRIFIED ORGANIZATIONS

It’s become a truism that we live in an age of disruption: incumbent companies become outmoded, outstripped, and outperformed by younger companies. Why? Because, the standard definition goes, insurgents provide a product at a much lower cost than the incumbents. But as Adam Pisoni, the chief technology officer of Yammer, the social enterprise company, said to us, disruption may have its roots in organizational structure.⁵

He uses a now-classic example: the entwined fates of Netflix and Blockbuster. Blockbuster, as American readers of a certain age will remember, was a video rental service that allowed people to watch movies at home. The medium by which it delivered that value? Physical storage in the form of videocassettes and, later, DVDs. Pisoni observes that Blockbuster was built around physical storage.

“They couldn’t change,” Pisoni says, “because all they could think about was how to improve the thing they did, not the value they offered.”

In this way, Blockbuster’s organizational entrenchment made the company brittle. Pisoni says that they organized themselves around the way—the medium—by which they provided value, not the value they created for the customer. Job descriptions, organizational structures, and the like, were built around making video rental stores, not helping people watch movies. When that gap between medium and value opens up, Pisoni says, disruption happens. If there’s a ghost of a video rental store in your neighborhood, you know what he means.

As Pisoni argues—and Schumpeter told us in the last century— the pace of these radical changes has accelerated. Fifty years ago, paradigm shifts like the ones that did Blockbuster in—the mainstreaming of the Internet, then broadband access, and then the increasingly streaming nature of video—would have come over the course of decades or longer. When the pace of change is slow, optimizing for predictability and efficiency is a sensible model, since the operating context remains reasonably stable. But when the rate of paradigm change increases—for example, when people move from desktops to laptops to smartphones to tablets in turns of five or so years—building a rigidly efficient system is no longer appropriate.

“Suddenly focusing on efficiency and predictability becomes detrimental. Focusing on organizing on value or agility becomes the only game in town,” Pisoni continued. “We’ve clearly crossed the Rubicon where the old, traditional structured companies are no longer effective.”

Founded in 2008, Yammer has emerged as a leader—if not the leader—of enterprise social networks. Founded by Pisoni and David O. Sacks, the start-up grew quickly, eventually raising $142 million in funding. It was acquired by Microsoft in June 2012 for $1.2 billion, and some serious cross-pollination is occurring as Microsoft Office, from what Pisoni says, is integrating with Yammer. This means that work is getting more social. But we are more interested in the way Yammer itself works.

“You have to understand Yammer is a product which was built to make companies better companies,” Pisoni says. “That puts the lens back on us.”

Yammer realized that it was great at iterating its product. Then it decided to turn that ethos inward: Why didn’t the organization become something to be iterated on, with each version rapidly tested for efficacy?

“We created a culture of organizational iteration where it gave us room to experiment and fail, because everybody from employees to managers knew this is just a big experiment,” Pisoni says. “Nothing’s permanent, and anyone can suggest any changes. It’s everyone’s responsibility to think about the system, not just their jobs.”

So what does that look like? You’ll find in Yammer’s office in Redmond, Washington, that the engineering team has something they call the Big Board. Products are listed on one side, and their teams are listed on the right, and the teams are reassigned from week to week.

Yes, they have an organizational chart, but instead of being organized by products and processes, it’s organized by skill set. And your boss doesn’t tell you what to do—exactly.

Leadership pulls people out of the org chart and makes temporal, product-focused teams. One of the engineers is made tech lead; that person runs that project. He or she will be your boss for 2 to 10 weeks, and that gives upper management a chance to see who makes a good manager. Small teams deliver value quickly with incremental A/B testing, then disband. While not specifically intended, this has another effect, Pisoni agrees: instead of people working on single projects for years and getting attached to them, they get moved around, eliminating the ego that gets attached to projects and placing the focus on the outcome.

Yammer has, in essence, created an environment with a swarm of ephemeral start-ups. Since Yammer’s structure is so fluid, Pisoni says, it can keep growing and not slow down.

In so doing, Yammer has developed a loose, organic, rotational, rapidly reincarnating structure for itself. One of the benefits of the rotationality, Pisoni says, is that you get implicit knowledge sharing, as in, you get to know more people who know more things. Why is this so crucial? Since problems are emergent, the best practices for dealing with them are equally emergent: there isn’t a playbook somewhere that can tell you what to do. Rather, it’s your network—the people you can call on, the institutional knowledge you can summon, the partnerships you have—that becomes your playbook. The more that people are interwoven with others in the work they do, the better decisions they can make, the more trust people can have with one another, and the faster you can move. The faster you move, the faster you can produce a new version, the faster you can get feedback, the faster you can improve. All together, this allows you to more quickly adapt to providing the value customers are looking for in your brand. In other words, organizational flexibility lets you become a more stable source of value, which allows you to produce a firmer partnership with users.

CONNECTIVITY PREDICTS SUCCESS

How crucial are your at-work relationships? Larry Miller is a medical statistician and a cofounder of Activate Networks, a consultancy that uses advances in network science to understand how the relationships within organizations drive success. He has a good answer.⁶

Activate was recently called into a large, innovation-centric engineering firm, the kind with thousands of engineers in a given plant. Half a dozen or so people to a team, all with the primary job of dreaming up new products. To get a read on how the innovation happened, Miller and his team mapped the network of individuals in the organization and then mapped the prescribed metrics for success, such as filing a patent or bringing something to the commercial marketplace. Activate then superimposed those successes onto the network of engineers.

What Miller found was striking: aside from years of experience, the highest correlation for those success metrics was quality relationships—and not for the team as a whole but whether a team had an individual who was highly connected to the total network. The takeaway: If you’re working in a closet somewhere, he says, it doesn’t matter how great your work is.

What makes for a powerful network for an individual? Miller says that you need to have broad connections. The highest performers tend to have cross-departmental links; if all your primary connections are in your own department, you won’t be as effective. Also, you don’t want to have connection only on your hierarchical level, but both up and down the ladder. And, he’s careful to note, it’s not merely that you have as many connections as possible but the strength of those connections—the bandwidth, if you would.

The reason teams with more connected individuals are successful, Miller says, is that they can spread their ideas throughout the organization, gather feedback on those ideas, and gain support for them. There’s something deeply intuitive about that. If you go to a strong connection of yours looking for commentary on your project and that person gives you commentary, there’s a sense of mutual investment, a sense of partnership that will develop from that constructive criticism. At an emotional level, there’s a growth of coinvestment; at a conceptual level, your and that person’s ideas get shaped and recombined. Whether or not you have a blindfold on, you’re getting more hands on the elephant.

We can think of the quality of the relationship as the bandwidth through which knowledge can move. If you work with someone but do not trust that person, you will not be able to exploit the implicit knowledge sharing that a rotational architecture such as Yammer’s is built to promote. In Activate’s case, teams within the engineering firm found success when at least one person was connected with the rest of the organization, allowing the insights of the team to gain exposure and feedback from other groups, disciplines, and perspectives within the organization. If you don’t have high-bandwidth relationships with people, you won’t be able to develop that sense of mutual buy-in.

As Yammer demonstrates, network building isn’t the province of only the outrageously social or the relentlessly networked; since connections are so indicative of the success of projects, organizations can take care to increase the quantity and quality of the relationships inside them. This is why when you’re onboarding employees, setting cross-departmental connections will be of benefit to each individual as well as to each team. Think of a person’s experiences as granting that person a certain amount of surface area: if you want your organization to be expansive, you need to have overlap, but you also need portions of each individual team member to be unreplicated anywhere else. In this way, the uniqueness of the individual informs the uniqueness of the organization. Such is the power of complementary contrasts. Getting people from diverse disciplines and work histories to work together is one of the best ways to architect innovation—a clustering we’ll discuss later in this chapter.

THE ORGANIZATIONAL OPPORTUNITIES OF MATURE COMPANIES

Structural innovation isn’t only for scrappy start-ups; mature companies have unique opportunities of their own. The breadth and depth of large organizations can lead them to intersectional innovations. One can call this combinatorial creativity: what happens when two previously unconnected ideas reveal their relatedness. This is a phenomenon found often in art. When Japan relaxed its isolationism in the late nineteenth century, the wealth of its culture spread across the world. Art historians will tell you how the flowers ornamenting the backgrounds of Van Gogh’s portraits came from the postcards of Japanese art that the painter collected, as did the nontraditional diagonal compositions that marked his later work. But this cross-pollination accelerates the growth not only of painting but of products. Procter & Gamble provides a perfect example.

THE MANY INTERSECTIONS OF PROCTER & GAMBLE

When A. G. Lafley became CEO at P&G in June 2000, the company was falling apart. His predecessor, Durk Jager, had just been thrown out after 17 months of missing earnings projections, and as BusinessWeek noted, its stock dropped $7 during his first week on the job. By the end of the year, the company lost a total of $85 billion in market capitalization. To dig itself out of that hole, it sold off failing brands: Crisco, Jif, and Folgers to name a few.⁷ But beyond getting trimmer, soon P&G became better connected.

As one of the conglomerate’s major brands, the transition that Crest made from dental care to oral care provides a telling example. Back in 2000, Crest was all toothpaste and toothbrushes. In his 2008 book The Game-Changer, Lafley reveals how Crest grew.⁸ He writes about the Corporate Innovation Fund within P&G, essentially the corporation’s in-house venture capitalist fund. That fund provided for a combination of talents and technology between different arms of the company, one of which resulted in Crest Whitestrips.

Bruce Brown, the CTO of P&G, would later unpack the cross-pollinating logic in an interview with Forbes.⁹ “Often [the] connection of seemingly disparate technologies delivers disruptive ideas. The magic in a big company is how to create space for connections, so an idea person can bump into a technology person,” he says, citing Whitestrips as an example of the power of intersectional thinking. The film came from packaging in P&G’s paper products, the bleach from its fabric products, and the glue from another application, he says, combining to create “a novel product delivering a service you could previously only get professionally but now get at home.”

According to a Harvard Business Review article, the product faced some internal naysayers during its development, and so P&G brought in a general manager from the Swiffer sweeper product group to guide the idea into reality. Finally, to win over dentists wary of endorsing themselves out of business, P&G made a Whitestrips Supreme product for dental office use.

The Whitestrips story is an interesting contrast to the fall of Blockbuster. Whereas Yammer CTO Adam Pisoni observed that the video rental service trapped itself within its sprawling structure to better produce a single product, P&G shows that large companies can exploit the existing structures of their businesses as long as they keep an eye on creating a value that their customers want. P&G was in fact able to do some disruptive innovation: Crest’s Whitestrips translated a high-end service into a customer’s domestic setting. This led to great reward: in their first year, the strips generated $200 million in revenue and grabbed close to 90 percent of market share.¹⁰

What the Whitestrips show—beyond a brighter smile—is the power of intersectional thinking: taking the expertise of several departments and combining them will yield new ideas. The fact that the home cleaning department and the oral care department would find a market-shaping collaboration is unexpected; that is precisely why such experimental, combinatorial strategies are so effective.

The structural, platform-growing innovation that P&G exemplified in its turnaround extends mindfulness—an awareness of what’s happening inside and outside a body—to an organizational level by seeing what assets that are already present in an organization can combine and meet market needs in new ways. Since gigantic corporations such as P&G have a wide set of businesses, their potential for combining competencies is far greater than that of their smaller peers. Since their work runs across verticals, you can make novel combinations within the organization, bringing new products—and product lines—to market. But this isn’t just something to be found in organizations: we as individuals can use it too.

To bring intersectional thinking into your day-to-day life, you could keep two books on your nightstand—one fiction and the other nonfiction, one contemporary and the other a classic, one a memoir and the other imaginative—and then find the intersections. This is also, perhaps, one of the great joys of undergraduate learning. Since all academic traditions are studies of humans and the universe, they all have more to do with one another than their departments suggest: one will study perception in art history, cognitive science, and marketing classes, and the studies in one field will inform those of the other. With a little intention, we can bring that scholarly “intersectionality” into our adult lives.

There’s a social element, too. Lunch is as good a time as any to cultivate cross-pollination. Just as a city such as New York or London benefits from the divergent histories that intersect there, you can make your lunch hour, perhaps once a week, an ephemeral city: invite people from different disciplines in and outside of your organization for a meal and a conversation. When you invite someone to a meal, you invite all that person’s experiences and ideas, too. Since we all have a different hand on the elephant, getting together allows us to compare notes on what this beast called working life is and let the insights arise from there.

ORGANIZING WITH TALENT CLUSTERS

Is there a way to organize for these sparks? We think so.

 

Cluster: a loose, cross-functional method for creating and implementing ideas

 

What if you took Yammer’s methodology of adaptability, rapidity, and impermanence for structuring its engineering team and moved it into a general context? What if you made the synergy that P&G engineered to create the Whitestrips more replicable? What if, instead of only trying to create a process within a piece of software or a new oral care product, the multitalented team in question were planning an organization’s future, meeting with users, or running another project? You’d have a taut, time-bound collection of people assembled around a specific task. We call such a group a cluster: our way of systematizing the emergent processes of Yammer, P&G, and other innovators.

Clusters come into being to address a particular challenge and then disperse once that challenge is met, though as the Yammer example points out, the bonds made between those people remain after the cluster disbands. This is also a way, then, of weaving together the total network of an organization, further allowing ideas and feedback to flow between pieces of the whole. The explicit links of the short-lived cluster form implicit, long-lived bonds between the people within them. The trust that arises will accelerate the movement of the organization.

Think of a cluster as a leaner version of a committee, council, or circle. To wrap our minds around the cluster, let’s evaluate what committees and their ilk are usually for: to map and oversee the responsibilities related to a given task, giving us (un)interesting terms such as steering committee. But in the innovative, agile, and creativity-based economy, people do best when they can steer themselves. A cluster is a way of cooperating with the fact that humans crave autonomy. It helps us organize people without encumbering them.

This is not anarchy. Clusters inherit much of the structure of top-down hierarchies—accountability, reward systems, and measurement of performance are all present—while including bottom-up innovations. Those innovations include recruiting the resources and personnel they need within or outside of the organization, an emphasis on self-direction, and aligning their actions to the vision of the whole. It’s all of what you need, as an advertisement might say, and none of what you don’t.

Clusters have the following characteristics:

 

• Tailored agenda. Each cluster has a specific reason for existing, with an agenda of objectives that matches that purpose.

• Time-bound existence. Clusters emerge when needs arise. They disband when their objectives have been completed.

• Evolving membership. Membership in a cluster is not fixed but fluid, with talent and perspectives changing to suit tasks as they develop.

• Self-organizing responsibility. Clusters develop their own structure and operational rules: they alone are responsible for their operation and its results.

• Adaptive ethos. The culture and personality of the cluster align with its purpose: some are explorative, some are directed.

 

There are three primary types of clusters, representing the three phases of development for the life cycle of a product, process, or service. They are visioning, ecological, and implementation.

Visioning Clusters

How does a product begin? Like all living things, it must be conceived. This is the work of the visioning cluster: these people sketch out the potentials of a product. This is more than a leadership committee. Yes, managers should be involved, but it’s also critical that implementation-focused and user-facing team members take part. At this opening stage, it is crucial that the cluster have a diversity of perspectives within the organization. As Professor Lauren Rivera might remind us, if a product is going to be aimed at a range of cohorts, those cohorts need to be represented in the visioning cluster.

The idea here is to take down the velvet rope surrounding leadership. Often leadership is a euphemism for the people who make the most money in an organization, which places the cart before the horse. Rather, leadership, as we noted in Chapter 1, is a way of owning the impact that you or your organization is going to have on the world. This demands foresight of what’s to come and insight on how to get there, which is not necessarily reliant on the amount of money you make. Indeed, as we’ve noted before, if you have only wealthy guys making decisions for an organization that has more than just wealthy guys in it, you’ll be blinding your perspective; you’ll have redundant hands on the elephant.

Instead, bring people with leadership qualities into the visioning cluster: it’s healthy to have a range of experience levels included, since that will create intergenerational bonds within the organization and illuminate blind spots.

Ecological Clusters

Bringing a product to market is intensely interdependent, relying on suppliers, customers, intermediaries, and partners. The ecological cluster’s primary purpose is to connect with these varied players within the ecosystem, building support across all these spheres.

As Dave Gray observes in The Connected Company, a fundamental characteristic of our era is that the line between organization and customer is increasingly perforated; that is, people outside an organization are increasingly being represented and included within it.¹¹ These ports of exchange are ecological clusters.

Ecological clusters can take many forms. It’s assumed that consumer-facing organizations will have a person with social media responsibilities, but that is an immature model of the ecological cluster. The mature model goes further by including stakeholders in decision making. Suppliers and intermediaries may participate directly in the ecological cluster’s functioning, though if there are parties who don’t care to be involved, we can use a proxy: someone who can accurately represent their point of view.

Although we cannot peer into the inner workings of P&G, we can surmise that something of an ecosystem cluster occurred when it put together the Crest Whitestrips, especially the advanced model for dental professionals. Sensing that the disruptive product would harm preexisting relationships with those critical customers, P&G was canny enough to make the premium product to preserve those partnerships.

A theme that emerges again and again in contemporary life is the increasingly palpable interconnectedness of every organization’s work. It behooves us, then, to have people on hand who can lend us the benefit of context. We want to have many perspectives on what the elephant in the room—our product—looks like. We’ll go into detail about which people to include in Chapter 6.

Implementation Clusters

Ideas don’t realize themselves. They need to be ushered into life by people. Just as the visioning cluster is composed of people who affect an idea’s conception, the implementation cluster is composed of those tasked with instantiation.

They get the work done, joining the prescribed vision with the context described by the ecological cluster.

Members of the implementation cluster are focused on making things. Its membership is more defined than are those of the other clusters: it must contain every element necessary to bring the vision to reality. As in the other clusters, participation varies with the dynamics of the cluster. Unlike the other clusters, however, members’ roles and participation are dictated by the implementation process.

The implementation process happens in three steps: prototyping, validating, and taking to market. This is a generalized process whether you’re building toasters or software, a healthcare workflow or a solar panel. Visioning is concerned with the blueprinting of an idea in the look and feel of a product. A coffee cup, for instance, will have its initial idea and purpose, but its specifications will not start to be laid out until the implementation phase. In this way, the product life cycle is a movement from vagueness to specificity. In the implementation phase, we’re starting to get specific.

Prototyping is different from visioning in that you’re actually starting to gather momentum for this product. If Leonardo da Vinci were making a cannon, he’d have to begin by testing the cannon out, which would require gathering the physical materials and funding necessary to make the first mock-up. The prototyping phase is simply a minimum product: let’s build our first cup of coffee, our first cannon, or our first piece of code and then start tinkering with it.

Next is the validation phase. You could think of this as a mature prototype: you’re testing in more contexts. This is where the ecological cluster plays a crucial role. As we know, the success of every product is contingent on a staggering number of factors, and an organization cannot know all those factors as it prototypes them. We end up learning things as if we were in a dark hotel room: by stumbling over furniture while pawing for the light switch. This is the part where you stub your toe and whack your shin, but in the validating phase, you’re still in a controlled, prerelease situation. This allows you to fail small, fail often, and improve rapidly. It’s important to note that some products never make it out of the validation phase. If, for instance, a critical supplier partner backs out, it’s better to find out in this phase and cancel the project than to roll out a less than excellent product.

Then, after prototyping and validating comes the launch, the mass production. This has different stages of finality that vary with the circumstance. If you’re creating a physical product, it’s not going to be modifiable once it’s out in the world, but if you’re building software or an organizational process, you can edit the product much more readily. Of course, a launch isn’t simply the physical creation of a product but also the way it is welcomed into the world. All of these clusters need to be integrated, taking into account the way all the people interacting with the product, from those within the organization to contingent partners to end users, need to be represented from the beginning. Just how will emerge as we continue on.