New Models for Manufacturer-Customer Relations

Frankfurt.
The Prosumer Revisited conference begins with a keynote by Frank Piller, who presents the perspective from management research. He begins by describing the story of the ice cream king of upper Manhattan - a small old-fashioned store which sells only a small range of flavours and does not mix them. This is the old market model - where the quality of products means that producers have no need to respond to the needs and interests of consumers. But we've moved away from this, as Chris Anderson's 'long tail' model shows - while such old-fashioned marketing models focussed on extracting profits from the short head of the long tail distribution, stores like Amazon have emerged to cater to the long tail, offering a vast variety of products but selling only a relatively small quantity of each title. What Amazon has managed is to build a sustainable business model from this.

In reality, the long tail is also about selling at a later stage more cheaply those products which didn't find a larger market, though. The challenge here is to access information about user needs in order to plan production quantities and new product development more effectively. In this, product definition is critical to new product success, but challenged by customer and technology undertainty - and even the ever more sophisticated models for market research or the better market intelligence which producers believe they have don't seem to provide eliable answers to this problem.

Indeed, classical methods may not be enough to understand effectively what customers want - new methods are necessary, and marketers can learn important lessons from the iterative trial-and-error (or more precisely, design, build, run, analyse, and repeat) models of software design, for example. Traditionally, concept testing or market research took place outside of the manufacturer organisation, of course, and that process of effectively bridging the manufacturer and consumer domains was very complicated - it was difficult to transfer the tacit and sticky information and knowledge held by consumers to the producer's side.

There are three main models for this: designing for customers (where companies design on behalf of customers, by analysing past demand and on that basis attempting to forecast future needs), designing with customers (where prototypes are presented for feedback to a 'representative' sample of customers), and finally design by customers (where customers are actively integrated in collaborative practices and are able to initiate new design processes). In this latter case, customer participation must be voluntary rather than enforced or specifically invited.

A second fundamental problem in product development, by the way, is the problem of local search bias, where producers tend to seek solutions only within the domains of knowledge within which they already operate - a tunnel vision for solutions which restricts innovation very severely.

One strategy to address this is mass customisation - the kind of customised on-demand production which Toffler described as early as the 1970s. Here, customers may even be able to create their own products, though from a limited selection of predefined elements - and by tracking their demand producers are then also able to fine-tune their product lines and production quantities. In this model, then, the divide between manufacturer and customer domains shifts further in favour of customers - they are now active to some extent also in designing and prototyping new products, not only in testing them and in providing feedback.

This strategy can be exemplified by an (largely unsuccessful) marketing by German electricity company E.ON, which provided MixPower, a system that enabled its users to mix the different energy sources used for generating their electricity, thereby balancing cost and emissions. (In reality, things were more difficult, of course, since it's impossible to provide specific source mixes to each individual consumer.) What this approach did provide, however, was very clear information about consumers' overall price/emissions sensibilities, giving the company clear details on what their ideal price points should be.

Such approaches work especially well where there is a very large 'solution space' - that is, where configuration options are so flexible that they produce a very large range of possible results. Indeed, people are also willing to pay much more for such highly personalised, customised products - not only because of the quality of the product, but also because of the satisfaction of having undertaken this design process themselves. This also has its own costs, however: not least an overload of choices. There is now also some very useful and important information about how to effectively design such user toolkits for product customisation, though.

Another strategy is to involve customers even more deeply in (collaborative) design processes - one example for this is the t-shirt company Threadless, where uploaded designs are evaluated by users of the Threadless site, which leads to greater production runs and even allows the site to pay winning designers some $2000 for winning designs. (It's also an important tool for emerging designers to show their wares, of course.) Frank calls this the collective consumer commitment method - where companies do no research and development of their own, but simply focus on supporting their users in creating value. This community-based model operates by broadcasting creative problems, then, and acts as a very effective screening model for quality content - a form of crowdsourcing.

A third strategy (which we'll touch on only briefly) is von Hippel's idea of the 'lead user', who creates (individually or in communities) new products and ideas in their own right - products which can be taken up later on by commercial manufacturers.

There is, then, a need for a fundamentally different form of dealing with customers. Customers take on a variety of different roles in such processes - as designer, as voter, as community manager, etc. Manufacturers need to think about what roles can be passed on to customers, and what roles need to be retained by the company - or indeed, what new roles (e.g. as community facilitator) they may be able to, may have to, take on.

Technorati : Prosumer Revisited, business, customisation, marketing, produsage, prosumer, user-led
Del.icio.us : Prosumer Revisited, business, customisation, marketing, produsage, prosumer, user-led