Start with Technology, End with Trust

The central economic imperative of the industrial age was to increase productivity. Every aspect of an industrial firm--from its machines to its organizational structure--was tailored to enhance the efficiency of economic production. But today productivity is a nearly meaningless byproduct in the network economy.

The central economic imperative of the network economy is to amplify relationships.

Every aspect of a networked firm--from its hardware to its distributed organization--is created to increase the quantity and quality of economic relationships.


The network is a structure...

...to generate relationships. Networks haul relations the way rivers once hauled freight. When everything is connected to everything else, relationships are rampant. Each variety of connection in a network begets a relationship. Between firms and other firms. Between firms and customers. Between customers and the government. Between customers and other customers. Between employees and other firm's employees. Between customers and machines. Between machines and machines, objects and objects, objects and customers. There is no end to the complexity and subtlety of relationships spawned in a network economy.

Each of these types of relationship has its own specific dynamics and quirks. And each is nurtured by a particular type of technology. The technologies of jelly bean chip and boundless bandwidth are, in the end, relationship technologies. "We need to shift away from the notion of technology managing information and toward the idea of technology as a medium of relationships," writes Michael Schrage in Shared Minds, a book about the new technologies of collaboration. Despite the billions of bits that information hardware can process in a second, the only matter of consequence silicon produces are relationships.

Of course reputation and trust have been essential in all economies of the past, so what's new? Only two things:

  • With the decreased importance of productivity, relationships and their allies become the main economic event.
  • Telecommunications and globalism are intensifying, increasing, and transforming the ordinary state of relationships into an excited state of hyperrelations--over long distances, all the time, all places, all ways. It's not Kansas anymore; it's Oz.

Relationships among more than two people can be structured as hierarchies or as networks. In hierarchies, members are ranked in privilege relative to one another; in networks, members relate as peers--counterparts of similar power and opportunity. In previous ages the most intelligent way to construct a complex organization in the absence of plentiful information was to build a hierarchy. Rank is a clever and workable substitute for ubiquitous real-time information. When information is scarce, follow orders.


When information is plentiful...

...peers take over.

In fact, as reliable information becomes common, almost nothing can stop peers from taking over. As computers and communications unloose a million bits of information in every dimension, we see peerages form in every dimension. Email and voice mail have brought peerage pressure to corporations. The flattening effect of network technologies and the subsequent turmoil in the organization of business firms is well recognized. But in many ways the emerging peerlike relationship between boss and staff is probably the least interesting and least important of all the relational changes now taking place.

More consequential is the relation between customer and firm, which is yielding to the peer effect. More important still is the relation between firm and firm, which is shifting rapidly to a web of overlapping nets. Still more vital is the lateral relation between customer and customer, which is just beginning to brew. Finally, the elevated relation between customers (rather than citizens) and the rest of society, a relation that is just now being defined, may be the most important of all, as economics elbows its way into every activity. As an example of expanding relationships, consider the traditional relationship between customer and a firm, roles that have been around forever. In the network economy the separation between customers and a firm's employees often vanishes.


When you pump your own gas at the filling station...

...are you working for the gas station or for yourself? Are all those people waiting in line behind the ATM machine more highly evolved bank customers or just nonpaid bank tellers? When you take a pregnancy test at home, are you a savvy self-helper, or part of the HMO's plan to reduce costs? The answer, of course, is both. When everyone is linked into a web, it's impossible to tell which side you are on.

Web sites and 800 numbers can invite customers into the internal knowledge banks of a company to almost the same degree of "inside" that employees stationed on the other side of the line enjoy. Many technical companies post the same technical information and diagnostic guidelines on their help sites that their own support professionals work from when you call their hotline. You can have someone trained to look up and then read troubleshooting answers for you, or if you are in a hurry, you can try to find it yourself. Who's working for whom?

At the same time the complexity of an employee contract, particularly in high-tech fields, is quickly approaching the complexity of a contract with an outside vendor. Stock options, vestment periods, a thousand insurance and benefit combinations, severance clauses, noncompete agreements, performance goals--each one uniquely negotiated for each person. A highly paid technical employee becomes in essence a permanent consultant. He or she is an outsider on staff.


Outsiders act as employees,...

...employees act as outsiders. New relationships blur the roles of employees and customers to the point of unity. They reveal the customer and company as one.

This close coevolution between users and producers is more than poetry. There is a very real sense in which the owners of the phone network sell nothing at all but the opportunity for customers to have conversations among themselves--conversations which the users themselves create. You could say the phone companies cocreate phone service. This blurring between origin and end spills over into the birth of online services, such as AOL, where most of what is now sold is being created by the customers themselves in the form of postings and chat. It took years for AOL to figure this out; they initially wanted to follow industrial logic and sell downloadable information created at great expense by professionals. But once they realized that the customers acted like employees by making the goods themselves, the online companies started making money.

The net continues to break down the old relationships between producers of goods and consumers of services. Now, producers consume and consumers produce.


In the network economy, producing and consuming...

...fuse into a single verb: prosuming.

"Prosumer" is a term coined by Alvin Toffler in 1970 in his still-prescient book Future Shock. (Toffler first found his insights as a futurist while working for the telephone networks.) Today prosumers are everywhere, from restaurants where you assemble your own dinner, to medical self-care arenas, where you serve as doctor and patient.

The future of prosumerism can be seen most clearly online, where some of the very best stuff is produced by the people who consume it. In a multiplayer game like Ultima Online, you get a world with a view and some tools and then you're on your own to make it exciting. You invent your own character, develop his or her clothing or uniform, acquire unique powers, and build the surrounding history. All the other thousands of characters you interact with have to be sculpted by other prosumers. The adventures that unfurl are cocreated entirely by the participants. Like a real small town, the joint experience--which is all that is being sold--is produced by those who experience it.

These eager world makers could be viewed as nonpaid content makers; in fact, they will pay you to let them make things. But the same world could also be viewed as full of customers who have been given tools with which they can complete a product to their own picky specifications. They are rolling their own, just as they like. In the new economy-speak, this is known as mass customization.


The premise of mass customization is simple.

Technology allows us to target the specifications of a product to a smaller and smaller group of people. First we can make Barbie dolls in the millions. Then with more flexible machinery and computer-generated target marketing we can make ethnic Barbies, in the hundreds of thousands. Then with improved market research and advanced communications we can make subculture Barbies, biker and grunge Barbies in the thousands. Eventually, with the right network technology, we can make the personal Barbie, the Barbie of you. In fact there is a company in Littleton, Colorado, that currently makes the "My Twinn" baby doll to look like the doll's owner. The doll's eye and hair color and hair style are matched to a photo of the child who will own it.

The most interesting aspect of prosuming and mass customization--of this new relationship between the customer and the firm--is that because customers have a hand in the creation of the product they are more likely to be satisfied with the final result. They have taught the firm how to please them, and the firm now has a customer with a much fuller relationship with them than before.

But creating a product for "a niche of one" is only a small part of the transformation of the customer relationship. (Detroit car makers learned long ago to create customized cars, but that was all they learned.) Network technologies such as data mining, smart cards, and recommendation engines are escalating the levels of relationships available to customers.


The drive to relate to the consumer intimately,...

...to the point of encouraging prosuming, can be articulated as a series of progressive goals:

1) to create what the customer wants
2) to remember what the customer wants
3) to anticipate what the customer wants
4) finally, to change what the customer wants

Each of the missions elevates the firm's commitment to the customer and raises the customer's involvement with the firm.


To create what the customer wants.

Sometimes this will mean simple customization: You want a vacation experience unlike anyone else's. Sometimes this will mean mass customization: You want a pair of jeans that fit your unusual leg shape at the same price as a regular pair of jeans. Sometimes mass customization is not what you want. The huge fashion industry makes its fortune on people's dependable desire for wearing what everyone else is wearing. Sometimes what you want is semicustomized: You read the New York Times because everyone else is reading it, but you don't read the sports section or the obits. You want not the Daily Me, but the Daily You and Me, the publication your 12 closest friends read.

A huge tide of information and trust must flow between users and creators in order to create exactly what the customer wants. The interface technology must be clear and simple for people to convey their desires. The nightmarish logistics of delivery and production must be managed with exactness. The most difficult aspect of this mission may not be the order form but the manufacturing; anything that involves atoms is much harder to customize than first thought. But any solutions surely involve networked technologies.


To remember what a customer wants.

A majority of the things we do, we do repetitively. We engage in the same tasks every day, or once a week, or every now and then. Things done iteratively have different dynamics from things done once. Little events become important. We bristle at having to remember our password again, or having to recite how we like our coffee one more time, or having to explain again what we don't like about bathing suits. Humans who learn our quirks (and they must be learned) earn our favor. Firms who learn our quirks will also earn our favor.

The technology of tracking and interpreting our whims heightens the relationships between firm and consumer. The firm must expend great effort to remember your preferences, but you also expend effort in teaching them so they can remember. And the remembering must be intelligent. You order the same espresso every day, except when it's cold out, and then you order a latte. The relationship tech has to be robust enough to be taught these distinctions.

Don Peppers and Martha Rogers, authors of the amazingly insightful Enterprise One to One, state: "A Learning Relationship between a customer and an enterprise gets smarter and smarter with every individual interaction, defining in ever more detail the customer's own individual needs and tastes. Every time a customer orders her groceries by calling up last week's list and updating it, for instance, she is in effect 'teaching' the service more about the products she buys and the rate at which she consumes them." In reward for the firm's effort at being taught, the firm and the customer develop a committed relationship. Peppers and Rogers continue: "The shopping service will develop a knowledge of this particular customer that is virtually impossible for a competitive shopping service to duplicate, providing an impregnable lock on the customer's loyalty." At the same time, the customer has invested so much in the relationship that the cost of switching to another vendor gets steeper by the day. Peppers and Rogers: "When the florist sends a note reminding you of your mother's birthday, and offers to deliver flowers again this year to the same address and charged against the same credit card you used with the florist last year, what are the chances that you will pick up the phone and try to find a cheaper florist?"


Since a relationship involves two members...

investing in it, its value increases twice as fast as one's investment.

The cost of switching relationships is high. Leaving, you surrender twice. You give up all that the other has put into the relationship, and you give up your own investment. In other words, the cost of loyalty is low. Thus we see the huge success of frequent flyer and frequent buyer programs, made possible by the coinvestment that airlines and supermarkets put into them. Affiliation cards are another example of the relationship extension; the costs of tracking purchases are so low compared to the value of belonging--for both sides--that it pays to invent other ways to spread the idea. And the phone companies' attempts at "friends of friends" calling circles are likewise clever experiments in exploiting networked relationships.

Smarter relationship technology, or "R-tech" as economist Albert Bressand calls it, will bind the connections between customers and firms more tightly still. An emerging standard called P3P offers a uniform way to store an individual's profile containing name, address, and so forth as well as preferences, including preferences of what they will reveal. If you shop a lot you will carry a "passport profile" based on the P3P protocol (or one similar) encased in your smart card or online in a browser. You exchange it with the vendor during a commercial transaction. The passport technology will help firms remember you as you teach them how to serve you and earn your favor.

The portability of preferences is a big deal. As the net creeps into yet more aspects of commerce, the ability to track identities and desires across different systems will be key. The Ritz-Carlton Hotel is justifiably proud of its ability to customize rooms for you anywhere in its thirty-one-hotel chain, without having to ask you. Some airlines can do the same. That still leaves a lot of room for success in creating relationships in the network economy as a whole.


To anticipate what a customer wants.

Creating tailored products for people is the first step of R-tech. The second is recalling their preferences intelligently. The third step is anticipating what they'll want even before they articulate it. That's a measure of any great relationship. You can boast you really know someone when you can say, "I know she'll love this book!"

It's remarkable how well this simple system works. I eerily recommended great albums that I liked. There are many refinements to increase its power. I can "teach" the system by grading the results it gave me. Perhaps it recommended Pete Seeger because I named Bob Dylan as a favorite. But say I happen to already know Seeger's work and can't stand him, so I tell it to forget Seeger (and thus Seeger-likes). It's now smarter. I can further locate my space with more precision by rating as many albums as I wish, indicating my love or hate of them. (A strong negative rating is just as useful as a strong positive rating.) Because it is the web, I also have the option of listening to music selections to refresh my memory or evaluate recommended candidates.

People who share small preferences for particular books or movies in a single "taste space" can use thier collaborative sorting to aid them in future purchases.


The real power of this system...

...lies not in mere recommendation, but in its ability to create relationships among its 3 million registered users. It allows members to link up with their taste-neighbors. All the fans of ambient music, or early Seattle grunge, are encouraged to strike up conversations in "venues," or start mail lists, or simply introduce themselves. Out of this technology is born yet another relationship: self-identity.

Most listeners don't have easily classifiable tastes. They're fans of Nirvana, U2, The Beatles, Joni Mitchell, and Nine Inch Nails. They'll have neighbors in an obscure unnamed space--the Beatles/U2/NineInchNails space. Through Firefly, these folks can identify their tastes by the microcommunity of like-minded folks they create for themselves. What Firefly can do with music, it can also do with books. And movies. And web pages. (Firefly recently spun each of these domains out to separate partners.) They are rated in the same way, with equally useful results. But now the combined media space is tremendously potent. Weird subcultures can be detected long before they have a name. Readers of Anne Rice vampire novels who like country and western music and Woody Allen movies suddenly realize they are a group! Self-recognition is the first step toward influence.


Online booksellers such as...

...Amazon.com and Barnes and Noble are using similar R-technology to sell more books, and to make customers smarter shoppers. Amazon derives its collaborative recommendations from customers who have a purchasing behavior similar to yours. Based on what you have bought in the past, and what others have bought in the past, Amazon advises: Dear reader, you should like these titles. And, they are usually right. In fact, their recommendations are so handy that they are Amazon's prime marketing mechanism and their chief source of revenue growth. According to company spokespersons, "significant" numbers of users buy additional books--on impulse--because of the co-recommendations that pop up when you inspect a book.

Evan Schwartz, author of Webonomics, goes so far as to suggest that firms such as Amazon should be viewed as primarily selling intangible relationships. "Amazon should not be compared to actual stores selling books. Rather . . . the value that Amazon adds is in the reviews, the recommendations, the advice, the information about new and upcoming releases, the user interface, the community interest around certain subjects. Yes, Amazon will arrange to deliver the book to your door, but you as a customer are really paying them for the information that led to your purchase." When you log on to Amazon you get a relationship generator, one that increasingly knows you better.


The beauty of network logic is that the mechanics...

...of this software does not rely on artificial intelligence, or AI. Rather the collaborative work is done by pooling the teaching that each person would do alone into one distributed base. It's an example of dumb power. Lots of people teaching a dumb program, but all connected together, producing useful intelligence. The strength of the network is built by the slim bits of information that each member is willing to share. Sometimes that's all it takes.

The web is a hotbed of innovations in R-tech. If you had success in a search and are willing for that information to be spread collectively to others, this lateral relationship can improve the search function for everyone. Sometimes called "collaborative filtering" these kinds of social network functions will spread widely within the web itself, as well as within companies and small work groups.


As in other technological evolutions...

...relationship tech will begin its innovation in the avant garde, then work back to the familiar.

R-tech first appears in the world of the web, but will gradually infiltrate the world of canned goods and sports equipment, as well as TV shows and vacation spots. Eventually it reaches the final stage in the progression of customer relations:

To change what a customer wants. The ongoing tango between customer and provider draws them together until their identities disappear at times. This is especially true in frontier arenas, where expertise is usually in short supply. At first there is no authority on what customers want or what providers should deliver--as in these early days of the web and e-commerce. Expertise has to be developed jointly, coevolved. Customers must be trained and educated by the company to teach them what they need, and then the company is trained and educated by the customers. We saw precisely this equation in the pioneer days of online conferencing about a decade ago. When email and chat began, no one knew the difference between great email and okay email, between fabulous chat areas and average chat areas. The best online companies learned all they knew from their first customers. But the customers, too, had little expertise of what to expect and so relied on the visions and vaporware suggested by the companies. Customer and company educated each other on what was possible.


Good products and services are cocreated:

The desires of customers grow out of what is possible, and what is possible is made real by companies following new customer desires. Because creation in a network is a cocreation, a prosumptive act, a multifaceted relationship must exist between the cocreators.

Cocreation and prosumption require an information peerage. Information must flow symmetrically to all nodes. In the industrial society, the balance of information inevitably sided with corporations. They had centralized knowledge while the customer had only their own solo experience divorced from that of all but a few friends. The coming network economy has changed that. Each new layer of complexity and technology shifts the action toward the individual.

The intent of networked technology is to make the customer smarter. This may require sharing previously proprietary knowledge with the customer. It may also be as simple as sharing what the company knows about the customer with the customer herself.

R-tech tries to rebalance the traditional asymmetrical flow of information, so that the customer learns as fast as the firm (and so the firm learns as fast as the customer). At first the idea of focusing on "learning customers" instead of the "learning company" seems misplaced. But it is part of the larger shift away from a view of the firm as a standalone unit and toward a view of the firm as an interacting node in a much larger network--a diffuse node made up of customers as well as employees.


Letting the customer learn with help from the firm...

...is not the only way to make the customer smarter. The other way is to reverse the usual flow of information in the market. John Hagel, co-author of Net Gain, says, "Instead of helping your firm capture as much information about the customer as you can, you want the customer to capture as much information about themselves as they can." And you want customers to capture as much information about the firms they are dealing with as well. There are several ways on the web to bias information toward the customer. Among the most exciting innovations are new vendors that send a bot around to comparison shop for you. If thirty music retailers online offer the soundtrack to the movie Titanic for sale, web sites such as Junglee or Jango will collect the offers from each vendor, and rank them for you. But the vendors are calling the shots; they craft the offer, keep the data of requests, and drive the sale.

By reversing the direction of information flow one can create a "reverse market." In a reverse market (already set by a few web sites), the customer dictates the terms of sale. You say, "I'd like to buy a Titanic CD for $10, new." You broadcast your offer into the web, and then the vendors come to you. This works best at first for high-ticket items such as cars, insurance, and mortgages. "I'd like a $120,000 thirty-year mortgage for my house in San Jose. I can pay $1,000 per month. Do I have any takers?" You set the terms, keep the data, and drive the transaction. Technology, of course, means that much of this negotiation happens in the background via agents and so forth; you don't have to do the haggling yourself. But the R-shift moves the capture of information into the hands of customers from those of the vendors. It makes the customer smarter.


And whoever has the smartest customers wins.

The third way to make the customer smarter is by connecting customers into a collective intelligence.

When personal computers first entered the marketplace in the mid 1970s, user groups sprung up everywhere to assist the perplexed. Anyone could attend a monthly meeting and swap useful tips about how to set up a printer, or get an upgrade program to work. It was all informal, and free, and democratic; those who knew, told; those who didn't know, asked questions and took notes. Each specific computer platform spawned local user groups in major cities. There were user groups for "orphan" equipment such as Amigas, and video game consoles, and of course for Macs and DOS-based PCs. Some user groups grew to have tens of thousands of members and some ran their own free software emporiums and had budgets in the millions of dollars.

Firms that encourage customers to talk to each other, to form affinity groups and hobby tribes, will breed smarter and more loyal customers while creating smarter products and services.

User groups were seen by the outside world as evidence of the lousy state of the computer industry. Manuals were horrible, interfaces unfriendly. Critics complained that you didn't need to join a user group to get your TV up and running, or to turn your dishwasher on. Yet for many computer wannabees, the shared knowledge of a user group was essential in starting the journey into computerdom, or later onto the net and the web.

In reality, user groups were not a sign of failure but a sign of intelligence. They were a means of making the consumer smarter. Some computer companies caught on to this reality early and made regular visits to the bigger user groups to answer questions and hear complaints and pick up suggestions. The user group, although independent and nonprofit, became part of the computer companies' extended self.

Today there are still some 2,000 Mac and PC user groups that offer regular meetings in the United States (and an equal number internationally). The Berkeley Mac User Group still boasts 10,000 members, and weekly meetings. Yet most user group action has shifted to the online space. Web sites with attendant conversation areas, FAQ (Frequently Asked Questions) archives, mail lists, and public bulletin boards all keep the distributed exchange of knowledge going.


A user group is a peerage of responsibility.

Group members take education into their own hands, and distribute the job of keeping up among themselves. It's long been appreciated that the best and most useful working knowledge about technical gear comes out of user groups. User groups are now a regular feature for avocations such as scuba diving, bicycling, saltwater aquariums, hot-rod cars, or any hobby where technological change seems to outrun understanding.

The most fanatical of user groups can be thought of as "hobby tribes," a phrase coined by science fiction writer David Brin. Hobby tribes are very informed, very connected, very smart customers. They band their enthusiasms together and become the experts. In some smaller niches they become the market, too.


Expertise now resides in fanatical customers.

The world's best experts on your product or service don't work for your company. They are your customers, or a hobby tribe.

Companies need user groups almost as much as users need them. User groups are better than advertising when customers are happy, and worse than cancer when they are not. Used properly, aficionados can make or break products.

The network economy has the potential to enable a civilization of aficionados. As customers get smarter, the locus of expertise shifts toward affiliates and home-brew groups, and away from large corporations or the solo academic professional. If you really want to know what works, or where to find it, ask a hobby tribe. And not just in the realm of high-tech knowledge. All knowledge is pooling into aficionados. Because of shared obsessions among horse lovers, there are more horseshoers working today than a hundred years ago, in the age of cowboys. There are more blacksmiths making swords and chain mail armor this year than ever worked in the medieval past. A network of aficionados is already here.

The net tends to dismantle authority and shift its allegiance to peer groups. The cultural life in a network economy will not emanate from academia, or the cubicle of corporations, or even primetime media. Rather, it will reside in the small communities of interest known as fans, and 'zines, and subcultures. In Future Shock Alvin Toffler sets the stage: "Like a bullet smashing into a pane of glass, industrialism shatters societies, splitting them up into thousands of specialized agencies . . . each subdivided into smaller and still more specialized subunits. A host of subcults spring up; rodeo riders, Black Muslims, motorcyclists, skinheads, and all the rest." That initial shatter is now several thousands of subcultures. For every obsession in the world, there is now a web site. What industrialization began by shattering, the network economy completes by weaving together and serving with great attention. The web of broken shards is now the big picture.


Information shifts toward the peerage of customers...

...so does responsibility for success. The net demands wiser customers.

The advent of relationship technologies on the net creates a larger role for the customer, and it puts more demands on the consumer, too. None of this enlargement of relationships can happen unless there are vast amounts of trust all around. "The new economy begins with technology and ends with trust," says Alan Weber, founder of the new economy business magazine Fast Company.

If you send all your workers home to telecommute, you'll need a whopping lot of trust between you and your workers for that relocation to succeed. If I tell Firefly all the books I read, all the movies I watch, and all the web sites I visit, I will require a high degree of trust from them. If Compaq lets me delve into its expensively compiled knowledge database of known bugs and problems with certain computer parts, it has to trust me.

Trust is a peculiar quality. It can't be bought. It can't be downloaded. It can't be instant--a startling fact in an instant culture. It can only accumulate very slowly, over multiple iterations. But it can disappear in a blink. Alan Weber compares its accretion to a conversation: "The most important work in the new economy is creating conversations. Good conversations are about identity. They reveal who we are to others. And for that reason, they depend on bedrock human qualities: authenticity, character, integrity. In the end, conversation comes down to trust."


A conversation is a pretty good model...

for understanding what is going on in the network economy. Some conversations are short, abrupt exchanges of minimal data; some are antagonistic, some are periodic, some are continuous, some are long-distance, some are face to face. A back-and-forth exchange starts between two people, and then spills over to several people, and as the conversation becomes multipronged and divergent, it gathers in more and more players. Eventually there are conversations between firms and objects as well as people, as more of the world's inanimate artifacts become connected. Increased animation increases the number or times of interaction, and the frequency of conversation. The more interactions, the more important learning becomes, the more essential relationships become, the more trust becomes a factor. Trust becomes what Weber calls "a business imperative."

But for all the talk of the importance of trust, it only comes at a price. It comes slow and it always comes awkwardly. "Trust can be messy, painful, difficult to achieve, and easy to violate," writes Weber. "Trust is tough because it is always linked to vulnerability, conflict, and ambiguity. For managers steeped in rationalism, hierarchies, rule-based decision making, and authority based on titles, this triad of vulnerability, conflict, and ambiguity threatens a loss of control."


The technologies of relationships will not ease...

...this fear or pain. They can strengthen and diversify relationships and trust, but not make them automatic, easy, or instant. At the forefront in the chore to cultivate trust--as a business imperative--stands the rugged hurdle of privacy. No other issue summarizes the unique opportunities and challenges of the network economy as much as privacy does.

Privacy concerns were once exclusively aimed at Big Brother government, but net residents quickly realized that commercial entities--the little brothers on the net--were more worrisome. James Gleick, a technology correspondent for the New York Times put it this way: "Whatever the Government may know about us, it seems that the network itself--that ever-growing complex of connections and computers--will know more. And no matter how much we bristle at the idea, we nevertheless seem to want services that the network can provide only if it knows."

An entire book could be written about the fundamental conversation between what we want to know about others and what we want others and the net itself to know about us. But I will make only a single point about privacy in space of an emerging new economy...


Privacy is a type of conversation.

Firms should view privacy not as some inconvenient obsession of customers that must be snuck around but more as a way to cultivate a genuine relationship.

The standard rejoinder by firms to objections from customers for more personal information is, "The more you tell us, the better we can serve you." This is true, but not sufficient. An individual can't comfortably divulge unless there is trust.

Take the trust many people feel in a small town. The interesting thing about a small town is that the old lady who lived across the street from you knew every move you made. She knew who came to visit you and what time they left. From your routine she knew where you went, and why you were late. Two things kept this knowledge from being offensive: 1) When you were out, she kept an eye on your place, and 2) you knew everything about her. You knew who came to visit her and where she went (and while she was gone you kept an eye on her place). More important, you knew that she knew. You were aware that she kept an eye on you, and she knew that you watched her. There was a symmetry to your joint knowledge. There was a type of understanding, of agreement. She wasn't going to rifle through your mailbox, and neither would you peek in hers, but if you had a party and someone passed out on the porch, you could count on the neighborhood knowing about it the next day. And vice versa. The watchers are watched.


One of chief chores in the network economy...

...is to restore the symmetry of knowledge.

For trust to bloom, customers need to know who knows about them, and the full details of what they know. They have to have knowledge about the knower equal to what the knower knows about them. I would be a lot more comfortable with what the credit companies knew about me if I knew with great accuracy what they knew about me, how they know it, and who else they told. And I'd be even more at ease if I derived some compensation for the value they get for knowing about me.

Personally, I'm happy for anyone to track all my activities 24 hours a day, as long as I have a full account of where that information goes and I get paid for it. If I know who the watchers are, and they establish a relationship with me (in cash, discounts, useful information, or superior service, or otherwise), then that symmetry becomes an asset to me and to them.


We see the first inklings of this trust machinery...

...in protocols such as Truste. Truste was founded in 1995 as a nonprofit consortium of web sites and privacy advocates to enhance privacy relationships in the online marketspace. They have developed an information standard also called Truste. The first stage is a system of simple badges posted on the front pages of web sites. These seals alert visitors--before they enter--of the site's privacy policies. The badges declare that either:

We keep no records of anyone's visit. Or,
We keep records but only use them ourselves. We know who you are so that when you return we can show you what's new, or tailor content to your desires, or make purchase transactions easier and simplified. Or,
We keep records, which we use ourselves, but we also share knowledge with like-minded firms that you may also like.

Those three broad approaches encompass most transactions; but there are as many subvariations as there are sites. (To post the badges or seal, sites must submit to an audit by Truste, which guarantees to the public that a site does adhere to the policies they post.) But the seals are only labels. The real work happens behind the scenes by means of very sophisticated R-tech.

Here is a hypothetical scenario of a visit to a Truste-approved commerce site a couple of years hence. I visit the Gap clothing store online. They notify me that they are a level 2 site; they remember who I am, my clothes size, and what I bought or even inspected last time I visited--but they don't sell that data. In exchange for information about myself, they offer me a 10% discount. Fine with me! Makes life easier. I visit the site of Raven Maps, the best topographical maps in the world. They let me know that my visit with them is on a level 3 basis--they trade my name and interests, but nothing else, with other travel-related sites, which they conveniently list. In exchange they will throw in one free map per purchase. Since the friends of Raven Map look very intriguing, I say yes. I visit CompUSA. They want to know everything about me, and they will sell everything about me, level 3. In exchange, they will lease me a multimedia computer with all the bells and whistles for free. Okay? Ummm, maybe. Then I visit ABC, the streaming video TV place. They declare that they keep no records whatsoever. Whatever shows I watch, only I know. They keep aggregate knowledge, which they use to lure advertisers, but not specifics. A lot of people are attracted to this level 1 total nonsurvelliance, despite the heavy dose of commercials, and keep coming back.


At the end of the month I get a privacy statement...

...similar in format to a credit card statement. It lists all the deals and relationships I have agreed to that month and what I can expect. It says I agreed to give the Gap particular personal information, but that information should go no further than them. I gave a pretty detailed personal profile to Raven and the three companies they gave it to show up on my statement. Those three have a one-time use of my data. Raven owes me a map. In the end I gave CompUSA my entire profile. I am owed a computer. The nine vendors they sold my info to also show up; they have unlimited use of my profile and CompUSA web site activities. I'll get junk mail from those nine for a while--but my new computer will be able to filter it all out! In addition, I made a deal with the New York Times which lets them keep my reading activities, but nothing else, for a free month's subscription. Also, my statement shows that American Airlines got my address from ABC, when they claimed level 1. I'll have to have my privacy bot contact them and sort that "mistake" out.

Caller ID, unlisted phone numbers, unlisted email address, individual-free aggregates, personally encrypted medical records, passport profiles, temporary pseudonym badges, digital signatures, biometric passwords, and so on. These are all the technologies we'll be using to sort out the messy business of creating relationships and trust in a network economy.

If only we knew precisely what relationships were. Industrial productivity was easy to measure. One could ascertain a clear numerical answer. Relationships, on the other hand, are indefinite, fuzzy, imprecise, complex, innumerate, slippery, multifaceted. Much like the net itself.

As we create technologies of relationships we keep running into the soft notions of reputation, privacy, loyalty, and trust. Unlike bit or baud, there's no good definition of what these concepts mean exactly, though we have some general ideas. Yet we are busy engineering a network world to transmit and amplify reputations and loyalty and trust. The hottest, hippest frontiers on the net today are the places where these technologies are being developed.


The network economy is founded on technology...

...but can only be built on relationships. It starts with chips and ends with trust.

Ultimately the worth of a technology is judged by how well it facilitates an increase in relational activity. VR pioneer Jaron Lanier has proposed the Connection Test: Does a technology in question connect people together? By his evaluation telephones are good technology, while TV is not. Birth control pills are, while nuclear power is not.

By this measure, network technology is a great deal. It has the potential to link together all kinds of sentient beings in every imaginable way, and more. The imperative of the network economy is to maximize the unique talents of individual beings by means of their relationships with many others.

That means not being connected at times. Silence is often an appropriate response in a conversation. Privacy is often advantageous in a networked world. The dimensions of relationship extend into not knowing as well as into the known. It is one of many mysteries in the human condition that will be wired into the technologies of the network economy.


Make customers as smart as you are.

For every effort a firm makes in educating itself about the customer, it should expend an equal effort in educating the customer. It's a tough job being a consumer these days. Any help will be rewarded by loyalty. If you don't educate your customer, someone else will--most likely someone not even a competitor. Almost any technology that is used to market to customers, such as data mining, or one-to-one techniques, can be flipped around to provide intelligence to the customer. No one is eager for a core dump, but if you can remember my trouser size, or suggest a movie that all my friends loved, or sort out my insurance needs, then you are making me smarter. The rule is simple: Whoever has the smartest customers wins.


Connect customers to customers.

Nothing is as scary to many corporations as the idea of sponsoring dens in which customers can talk to one another. Especially if it is an effective place of communication. Like the web. "You mean," they ask in wonder, "we should pay a million dollars to develop a web site where customers can swap rumors and make a lot of noise? Where complaints will get passed around and the flames of discontent fanned?" Yes, that's right. Often that's what will happen. "Why should we pay our customers to harass us," they ask, "when they will do that on their own?" Because there is no more powerful force in the network economy than a league of connected customers. They will teach you faster than you could learn any other way. They will be your smartest customers, and, to repeat, whoever has the smartest customers wins.

Just recently E-trade, the pioneering online stock broker, took the bold step of setting up an online chat area for its customers. We'll see more smart companies do this. Whatever tools you develop that will aid the creation of relationships between your customers will strengthen the relationship of your customers to you. This effort can also be thought of as Feeding the Web First.


All things being equal, choose technology...

...that connects. Technology tradeoffs are made daily. A device or method cannot be the fastest, cheapest, more reliable, most universal, and smallest all at once. To excel, a tech has to favor some dimensions over others. Now add to that list, most connected. This aspect of technology has increasing importance, at times overshadowing such standbys as speed and price. If you are in doubt about what technology to purchase, get the stuff that will connect the most widely, the most often, and in the most ways. Avoid anything that resembles an island, no matter how well endowed that island is.


Imagine your customers as employees.

It is not a cheap trick to get the customer to do what employees used to do. It's a way to make a better world! I believe that everyone would make their own automobile if it was easy and painless. It's not. But customers at least want to be involved at some level in the creation of what they use--particularly complex things they use often. They can superficially be involved by visiting a factory and watching their car being made. Or they can conveniently order a customized list of options. Or, through network technology, they can be brought into the process at various points. Perhaps they send the car through the line, much as one follows a package through FedEx. Smart companies have finally figured out that the most accurate way to get customer information, such as a simple address, without error, is to have the customer type it themselves right from the first. The trick will be finding where the limits of involvement are. Customers are a lot harder to get rid of than employees! Managing intimate customers requires more grace and skill than managing staff. But these extended relationships are more powerful as well.

The final destiny for the future of the company often seems to be the "virtual corporation"--the corporation as a small nexus with essential functions outsourced to subcontractors. But there is an alternative vision of an ultimate destination--the company that is only staffed by customers. No firm will ever reach that extreme, but the trajectory that leads in that direction is the right one, and any step taken to shift the balance toward relying on the relationships with customers will prove to be an advantage.


Archives - This site operates under a Creative Commons License.

This is a blog version of a book of mine first published in 1998. I am re-issuing it (two posts per week) unaltered on its 10th anniversary. Comments welcomed. More details here.
-- KK