Technology & Innovation

Crowd wisdom

January 23, 2008

Global

January 23, 2008

Global
Our Editors

The Economist Intelligence Unit

_____________________

User-centric innovation

Talent for innovation: Getting noticed in a global market incorporates case studies of 39 companies selected as Technology Pioneers in biotechnology/health, energy/environmental technology, and information technology.

Is it a bold and promising new approach to innovation, or just the application of new buzzwords to existing practices? The idea of involving customers directly in the development and deployment of new products and services, variously known as "user-centric innovation", "outside innovation", "mass collaboration", "wikinomics" or "crowdsourcing", certainly sounds novel. In some ways, however, it is not that different in practice from established practices such as open-source software development, in which volunteer programmers collaborate over the internet, or the even older practice of designing new products that take into account the wishes of particularly demanding customers. And even Scottish philosopher Adam Smith praised workmen who developed "easier and readier" tools for their own use, noting that many machines involved in early industrial processes were invented by end-users who improved upon earlier versions.

But what is certainly new is that companies are now recognising that customers can play a valuable role in devising and distributing new ideas — and are taking a greater interest in tapping into a source of inspiration and innovation that may only have been used informally and occasionally in the past. This is bringing into being new products and services, by distributing tasks among large groups, exploiting large data sets and economies of scale, and harnessing the enthusiasm of volunteers, enthusiasts and tinkerers.

Much of this has been made possible by the internet, of course, and its famed ability to allow groups of people with a shared interest, no matter how obscure, to connect with each other. But it is also a recognition by companies that their customers constitute a valuable supply of ideas, manpower, skills and information. The old model, in which a company develops a product and then puts it on the market for customers to buy, has given way in many cases to a new model in which customers participate in the development of new products, and help to create, improve, distribute and promote them. Companies generally have far more customers than they do employees, after all, so why not make use of their input?

A classic example, cited by Eric von Hippel, a guru of user-centric innovation and the author of Democratizing Innovation, is that of the Lego Group, Europe's largest toymaker, and its Mindstorms product. This combined Lego's usual snap-together plastic bricks with a special control unit into which simple programs could be downloaded from a computer, thus making possible the construction of robots and other complex models. The software inside the Mindstorms brick was developed over a period of many months by a small team of Lego engineers. Within three weeks of the original product's release in 1998, however, an internet user group had reverse-engineered the software and written their own, more efficient version. The user group quickly amassed hundreds of members — far outnumbering Lego's original team of programmers. At first the company was unsure how to respond, but eventually it decided to embrace the contributions of its customers, opening up its products for further customisation and including their ideas in subsequent versions of Mindstorms. The number of potential innovators outside the company, its managers realised, was far greater than the number within the company.

There are many other cases. Haier, China's leading manufacturer of household appliances, found that some of its customers were using its washing machines to clean vegetables as well as clothes, so it made new models with wider outlet pipes to prevent them from getting clogged up with bits of peel. Boeing and Airbus design new airliners after holding detailed discussions with their main customers, and also work closely with their own suppliers of machine tools on the design of new production gear. Manufacturers of sports equipment, from mountain bikes to snowboards, have learned to pay attention to the ways in which advanced users customise their products. One survey found that 38% of members of specialised "extreme" sports clubs had developed or modified their own equipment, for example. These "lead users" can provide valuable ideas -— what Dr von Hippel calls a "product feedstock" — for future models.

Manufacturers and users each have access to different information, which prompts them to innovate in different ways. In the case of scientific instruments, for example, users tend to modify them in order to get them to do entirely new things, whereas manufacturers' innovations tend to be about making existing products easier to use, more convenient, or more reliable. In general, Dr von Hippel points out, manufacturers tend to look for problems that can be solved using whatever technologies they have to hand (a new material, an improved process, and so on), whereas users tend to look for new solutions to problems they really need to solve. That makes user-driven innovations more valuable and, studies suggest, more profitable as well when harnessed by manufacturers. And even when they are not directly involved in the development of a new product, users can play a valuable role in improving and deploying it, and their willingness to get involved is, once again, a recognition of the product's appeal and value. The involvement of users in various ways now underpins innovation in a number of fields, as this year's Technology Pioneers demonstrate.

Look it up on Wikipedia

Perhaps the single best example is the Wikimedia Foundation, the organisation behind Wikipedia, the open-source, user-generated encyclopedia, and several other related projects. The basic idea of Wikipedia is that of an encyclopedia in which every page is a wiki — a web page with an "edit this page" button. This allows any reader to add to, correct and interlink the encyclopedia's entries. ("Wiki" is a Hawaiian word that means "fast", since the wiki-based approach to creating and editing web pages is much faster than using traditional authoring software.) Launched in January 2001, Wikipedia had 20,000 entries in English by the end of that year, 100,000 a year later, and then grew rapidly, amassing over 2m entries by October 2007 and making it the largest encyclopedia ever compiled.

Wikipedia is arguably the most impressive item of "user-generated content" on the internet, not only because of its size, but also because of its accuracy. Studies have shown that although its entries vary in quality and contain errors, the frequency of errors is comparable to that of traditional encyclopedias. When an expert in a particular subject spots a mistake in a Wikipedia entry, it is, after all, a simple matter to correct it. Inevitably, some entries (such as those on Islam and George W Bush) have become the focus of controversy and conflict, so that it has become necessary to restrict the editing of some entries to prevent vandalism. But rather than being an indication of the weakness of Wikipedia's approach, this could be seen as a measure of its success and influence. Look up almost anything using a search engine, and the corresponding Wikipedia entry will usually be at or near the top of the list of results.

The wiki-based approach has inspired a host of other related projects in which the users directly participate in the creation and development of intellectual property. The basic idea is "many hands make light work"; useful things can be done by sharing out tasks among a large community of internet users and aggregating the results. This has also been called "wikinomics" and "crowdsourcing" and has much in common with the Web 2.0 movement, which involves harnessing the contributions of users, both in creating content and helping others to navigate it. One popular way to do this is to allow users of a website to apply keywords or "tags" to items of content such as video clips (YouTube), blog postings and news items (Digg, Reddit) or photographs (Flickr). This provides a flexible, bottom-up way of arranging and navigating through large amounts of information.

Polar Rose, a start-up based in Malmo, Sweden, is applying the crowdsourcing approach to photographs with a particular aim: to create a visual index of images of people on the internet. Whenever its users, called "explorers", see photographs of people on the internet, they have the option of adding a tag giving the name of the person shown. As more and more images are tagged, two things become possible: users can use the Polar Rose browser plug-in to ask "who is this" (a small rose icon is added to tagged photos) and can search for images containing a particular person. As with Wikipedia, the users of the service also help to build it.

FON is a Spanish start-up with a novel approach to building a free wireless network: get the network's users to do it, granting them free access in return for sharing their own Wi-Fi access. Members of the FON community, who are known as "Foneros", install a Wi-Fi base station called "La Fonera". This functions like an ordinary Wi-Fi base station, providing wireless access to a broadband internet connection within a small radius, but it provides authorised access to other Foneros as well as its owner. The owner, in turn, gets free coverage when in range of another Fonero's base station. BT, the UK's main telecoms operator, recently joined forces with FON and has invited its 3m broadband customers to become Foneros. It joins other major operators such as Time Warner Cable, Neuf Cegetel and BB Excite. Non-Foneros can use the FON network by paying a small fee, with a small commission going to the Fonero through whose base station this fee is collected. The result, says FON, is the largest Wi-Fi community in the world — built and operated by and for its own users.

This won't hurt a bit

It is hardly surprising that the internet is central to many examples of user-centric innovation. But that does not mean that the same sort of approach does not apply in other fields. It is not just programmers who like to build their own tools; the same is true of doctors, and surgeons in particular. They often face problems that can be solved by inventing a new device or customising or improving upon an old one, and their direct practical experience gives them insights that a theoretical researcher at a medical-instrument manufacturer simply would not have. One survey of surgeons carried out in Germany found that 22% had devised or modified an item of medical equipment for their own use, and that around one-half of these innovations ended up being adopted by medical-equipment makers. The first heart-lung machine, for example, was developed by John Gibbon, a Philadelphia surgeon, and his team in 1953, and was subsequently commercialised.

The CyberKnife, invented by John Adler of Stanford University and now commercialised by his company, Accuray, based in Sunnyvale, California, follows in the same tradition. During a fellowship in Sweden, Dr Adler was inspired when he saw a device called a Gamma Knife, developed by Lars Leksell, a Swedish surgeon. This used 3-D imaging to deliver finely targeted bursts of radiation to destroy otherwise inaccessible brain lesions, but required the patient to be held perfectly still using a system of restraints screwed into the skull. Dr Adler set out to improve on this approach and devised the CyberKnife, which uses robotic technology to monitor and compensate for the patient's movement during treatment. In addition, the CyberKnife can be used to treat hard-to-reach lesions in other parts of the body, not just the brain. In keeping with the tradition of user-led innovation in medical technology, Accuray has set up an online forum to allow surgeons to share their experiences, offer advice to each other and, in the process, steer the development of subsequent versions of the product.

As well as developing new tools and techniques, surgeons are also well placed to develop new medical devices and implants. Joshua Ben-Nun, an experienced Israeli eye surgeon, has developed a new kind of lens implant for use in treating cataracts (the gradual yellowing and clouding of the lens that causes loss of vision). The usual treatment is to remove the lens and to replace it with an artificial implant. But such implants have no ability to focus, so that the patient ends up with excellent long-distance vision, but must use glasses for reading. One way around this problem is to use a multifocal lens implant, with concentric, alternating rings that focus on near and distant objects. But these lenses can cause problems with night vision. Dr Ben-Nun's company, NuLens, is one of several firms pursuing a new "accommodative" lens that can refocus just as a natural lens does, by changing its shape. His design is based on a flexible capsule of silicon gel that bulges in response to movement of existing eye muscles. It has been successfully tested in monkeys, and the first human trials are now under way.

The latest twist on user-centric innovation in bio-science comes from 23andMe, a start-up based in California's Silicon Valley. The company, which takes its name from the 23 pairs of chromosomes that make up an individual's genome, is a "personal genetics" outfit that provides detailed analysis of its customers' genomes from a saliva sample. By looking up the results on the company's website, customers can then find out about their ancestry, their inherited traits, and whether they are at risk of developing particular diseases. Several other companies offer similar services, all of them based on genetic analysis using "gene chips", rather than the full sequencing of each customer's genome — something that is still prohibitively expensive. 23andMe's analysis, which looks at around 600,000 regions in the genome where the most important variations between individuals are found, costs US$999; sequencing a full genome would cost US$1m-2m, although this figure is expected to fall a thousand-fold within the next decade. What 23andMe adds, however, are social-networking features akin to those on Facebook or MySpace, allowing users to contact other people with whom they share ancestors or genetic risk factors. According to the company's founders, by pooling the genetic data from thousands of users while maintaining personal privacy, it will eventually become possible to perform new kinds of research and enable users to take part in scientific studies. And as medicine becomes increasingly "personalised", and new drugs emerge that only work well on a specific subset of the population, it will be possible for cancer patients, for example, to identify which treatment is best for them.

Plugging into the innovation grid

What of the field of energy? It is hard to imagine how users can drive innovation in such a capital-intensive industry, even working together. And yet they can. For instance, the Toyota Prius, an iconic petrol-electric hybrid car, captures energy usually lost during braking and uses it to power the vehicle at low speeds, such as in stop-start traffic, and to provide occasional bursts of acceleration when needed, thus boosting the car's fuel economy. This is all very clever, but for some Prius drivers it is simply not green enough. So they have modified their vehicles by fitting them with extra batteries and new control software. This gives the modified Prius a range of 20 miles or so on all-electric power; the petrol engine only kicks in on longer journeys, or at highway speeds. (A standard Prius, in contrast, can travel less than a mile on battery power alone.) The modified version can be charged from the mains overnight, so that anyone with a short commute can use it as a pure electric vehicle. Toyota was initially sceptical about this "plug-in hybrid" approach. The company thinks the simplicity of the original Prius, which never has to be plugged in, has greater consumer appeal. But faced by the groundswell among its most enthusiastic customers as they modified their vehicles, it has conceded that plug-in hybrids might make sense after all, and is now developing its own version, due in 2010.

Consumers are also becoming more influential as utility companies look for new ways to generate electricity and manage demand. One increasingly popular approach is to allow homes and offices that generate their own electricity (from solar panels or wind turbines) to sell excess power back to the grid when needed. Another proposal is that the batteries of electric cars could, in effect, be pooled to provide utilities with a vast energy-storage system. When the cars are plugged in, energy is dumped into their batteries from the grid. If more power is suddenly needed, it can then be borrowed back again, flowing out of the batteries and back onto the grid. Once there are significant numbers of electric (or plug-in hybrid-electric) cars in circulation, a useful proportion of them will be plugged in at any one time. Rather than simply being users of electricity, in short, consumers could soon be involved in generating and storing it, in partnership with utilities.

All of this will require new "smart grid" technologies such as those developed by GridPoint, a start-up based in Washington, DC. It has devised an internet-based control system that gives utilities centralised control of power sources, energy-storage capacity and even individual appliances on customers' premises. This allows a power company to draw on excess power from solar panels on its customers' roofs when needed, for example, or turn down air-conditioning units to reduce demand when necessary. Customers can see what is going on via a website. Smarter grids that allow utilities to balance supply and demand more easily will pave the way for much wider adoption of alternative-energy sources by their customers. Similarly, SilverSpring Networks, based in Redwood City, California, has developed smart electricity meters that enable customers to take an active role in managing their energy consumption. A two-way network connection with the meter, based on internet standards, allows utilities to implement dynamic pricing and enables consumers to regulate demand accordingly. A washing machine, for example, could be set to come on when the electricity price falls below a certain level. As standards for smart grids start to emerge, they could underpin a bottom-up revolution in energy technology, just as the rise of internet standards has unleashed so much innovation in computing. Following the user, the user, the user

How can companies best exploit user-centric innovation? Dr von Hippel suggests three possible responses. The first, and simplest, is to adopt user-developed innovations and produce them commercially, in some cases in co-operation with the users responsible for the innovations. If advanced users of a particular product are modifying it in a particular way, it makes sense for the manufacturer to adopt their suggestions and offer the modified product to a wider customer base, as they are likely to have similar needs. Companies can gain competitive advantage by doing this more efficiently than their rivals, perhaps by forming close links with their lead users; attempting to discourage lead users from modifying products (in other words, innovating) can be counter-productive. A variation of this approach is to offer custom-manufacturing services to advanced users, which can be beneficial to both parties. Users can take advantage of high-quality manufacturing techniques, and manufacturers can learn and benefit from the designs cooked up by their customers. Threadless, for example, is an online T-shirt firm that invites internet users to upload T-shirt designs, vote for their favourites and order them to wear. Each week's winning design wins a cash prize and prompts thousands of sales. The suggestion by the company's founders that this approach could work in the car industry seems far-fetched, given the long lead-times involved, but it could probably work for surfboards or customised mobile phones.

The second approach companies can take is to make life easier for would-be user innovators, by offering design tools or other platforms for their use. Many computer games, for example, now ship with special level-design software that enables gamers to design new levels and accessories for the game, which are often distributed on the internet. Traditionally, game developers kept such tools to themselves. But they have now realised the value of enabling users to modify and extend their products. In some cases, these user-generated add-ons are adopted by the original manufacturer and become products in their own right. A good example is "Counter-Strike", a counter-terrorism combat simulator, which was created by enthusiasts who modified an existing game called "Half Life". Such was the popularity of "Counter-Strike" that it was eventually released as a separate title. This broadened access to the game, since there was no longer any need to apply software patches to "Half Life", a fiddly process that put some people off. Providing tools for users is, in some ways, the opposite of what companies are used to doing. Rather than trying to anticipate the needs of their customers, they are in effect giving up and simply letting the users do what they like. But this can be a rich source of new ideas.

Third, companies can acknowledge user innovations by selling complementary products or services. Many computer-makers, for example, sell machines that are designed to run Linux, an open-source operating system developed and maintained by its users. They do not profit from the software, but from the demand it creates for hardware. Similarly, many software companies have developed commercial products, such as database programs, that run on top of Linux. And a host of firms have sprung up to provide consulting and support services for Linux and other open-source products. They are thus capitalising, albeit indirectly, on user-driven innovation. Some software companies operate "hybrid" open-source models in which they make part, but not all, of their products freely available. Apple does this with its Mac OS X operating system, for example, the basic underpinnings of which are available as an open-source project called Darwin. This allows the company to benefit from improvements to Darwin made by volunteers, while retaining the ability to charge for the full version of its operating system, in which Darwin is overlaid with an advanced graphical interface. Apple also includes many popular pieces of open-source software with its Macintosh computers. It thus benefits from the innovation that goes into those products, while helping to encourage their adoption.

Although some aspects of it have been around for years, it is still early days for the user-centric innovation model, and surprising examples of it continue to emerge. And while it is true that much of the action is currently in software, the same principles and ideas are increasingly being applied in other fields in unexpected ways. It might seem hard to imagine how user-driven innovation could be applied in the pharmaceuticals industry, given the huge costs and regulatory hurdles associated with developing new drugs. How can you possibly trust users to do it? But even here there are innovative approaches emerging from the undergrowth: one is a new online game, akin to a three-dimensional version of Tetris, in which players race to fit molecules together. As players compete, they are actually working their way through a library of drug candidates, looking for possible docking sites with proteins associated with particular diseases. As one researcher in the field points out, people around the world play billions of hours of Solitaire on their computers each year; building the Panama Canal, by comparison, took just 20m man-hours.

Searching for new drugs by getting volunteers to play a computer game? It sounds daft. But the idea of an encyclopedia written by its own readers, and becoming the largest and most comprehensive work of its kind, would have seemed outlandish a decade ago. Of course, user-centric innovation is not for everyone, and works better in some fields than others. That is the nature of innovation: as with Sudoku, the popular Japanese number-placing puzzle, there is no single winning strategy that works, and success depends on combining a number of strategies. User-centric innovation is just one more tool in the innovation toolbox. But even for companies that cannot find a way to embrace this new approach, it provides a valuable lesson: that good ideas can, and increasingly do, come from unexpected or unconventional sources.

Enjoy in-depth insights and expert analysis - subscribe to our Perspectives newsletter, delivered every week