So You Want To Start A PSS... Part 2

(Previously, Part 1 of the series)

Let's ease back into this by talking about widget storage for your Product Service/Sharing System. If you are going to share things around, you need to figure out what to do with your widgets when they aren't being claimed or used by anyone in the network. At any given time, some of these widgets will be collecting dust somewhere, unless you are extraordinarily lucky and have 100% utilization. (You're a damned liar, by the way. Nobody has 100% utilization.)

Guideline #4: Distribute the storage if you can, centralize it if you must (extended discussion)

So let's say you've got a commercial system which shares art installations between corporate environments on a rotating basis. What do you do with the art that isn't currently in use? That's not too tough -- it's art, it's got some value, you put it in a secure climate-controlled warehouse. I.e. - central control and storage. There is no upside to leaving unused art with the previous corporate client -- the client doesn't want to pay for the art anymore, the client has no incentive to store your art on their dollar and give it adequate security and protection. 

In contrast, what about a group on the Internet that gets together to share Vietnamese-language DVDs amongst themselves? Does it make sense to start by putting all the DVDs in a secure climate-controlled warehouse? Again, that's a pretty easy answer -- not really. If the system was born distributed, the storage can stay distributed, at least for a while. The gross value of all those DVDs isn't enough to justify a centralized storage system, and 500 DVDs don't take up a lot of space. A couple DVDs can stay on each shelf of each person in the group until someone expresses an interest, and then the someone who HAS the dvd agrees to send it to the person who does not. This is an example of the "peer-to-peer" arrangement for a PSS. 

But there are a lot of potential PSSs where the answer is not clear-cut at all. Consider, for example, a group of young men, each of whom has an interest in looking particularly wealthy and fashionable when they show up to work. They get together and decide that while they aren't willing to share suits around, they would each like to make the impression at their job that they have a lot of money to throw around on accessories. So, they want to create a pool of Rolex watches, diamond stud cufflinks, and extremely expensive hand-knotted ties. Now, already I dislike these guys, but I do like it that they're trying to create maximum lifestyle wealth with minimum capital outlay. The question is, who keeps the watches and cufflinks and ties that aren't currently being worn? Do these guys trust each other enough to allow thousands of dollars worth of accessories to sit around in each guy's condo? Maybe so, maybe not. With a small group, when does it make sense to formalize the storage arrangements, and who would be responsible for sending out and then retrieving accessories?

The handbag and jewelry clubs have all been centralized precisely because the answers to these questions are still unclear. It's easier, but more costly in maintenance, to create a corporation (or other formal group) that can hire people to administer and adjudicate fairly. 

Fairness leads us directly to the reputation discussion.

Guideline #5: Establish a reputation system

Your network is going to be made up of people. We're going to assume that more than one person is going to be involved in sharing goods around, and this means trouble. Your PSS must allow for people to build reputation, otherwise there is no trust network. You can get very fancy with this, and implement systems where the good opinion of highly reputable people is worth more than reputation from unknowns or low-rated people. But regardless, you need to give people the option to avoid transactions with jerks, flakes, perpetual procrastinators, poor packagers, and con men. Or con persons.

Guideline #6: Make sure that the signal can be heard above the noise

One of the problems with uncontrolled/peer-to-peer collections of stuff is the tendency of worthless items to overwhelm the smaller number of things people actually want. Using the web to administrate your PSS allows you to add all kinds of signal-promoting features like wishlists, good site search, premium listings and other tools. One of the benefits of doing this so cheaply is that you can allow widgets to be listed for free. Unfortunately, free is also bad. It's not so much that the item may be bad, but that people don't discriminate among their widgets when they have unlimited free entry into the widget-space. 

One compromise is to make sure that listings expire, so that mis-priced items, poorly photographed items, incomprehensibly described items, simply drop out of the system. However, the problem with forcing widgets in a PSS to "expire" is that the good stuff starts to get snapped up first-come first-serve. Maybe that doesn't sound like a problem to you -- after all, it's not a bad thing to expect your users to log in every day, or even every hour, desperately hoping that today will be the day that a Gymboree pink elephant jumper, size 2T, will pop up in the system. But unless you have the kind of tremendous volume that eBay has, all that will happen is that a few hard-core reloaders get all the good stuff (the signal), and the more casual majority are left to pick through the remnants. 

Running a centrally managed and controlled PSS? Easy answer -- feature the good stuff (signal) on the front page, and make everyone queue for it. Running a classified ad service? Make people pay to put the good stuff (signal) on the front page. 

But if you want to run a network as innovative as a PSS, you'll need to innovate or imitate in order to keep items and services in circulation. One way to do this is to use people to weed out worthless items.

Guideline #7: Enter the boutique space with curated collections

One of the finest features of Etsy, and a technique employed by the clothing PSS Swango, is to use humans to pick out the best stuff available. Listings on swango would be removed if they didn't meet the standards of the network. It's labor intensive, but you can reward people for their labor, either by hiring people (the usual way), or by motivating the people in the network to curate the items in the network. On the front page of Etsy, there's a mini-boutique of a dozen delightful, astonishing, charming, coordinating handmade crafts that are as lovely as any department store display. You can do this too -- it's called a curated collection. You give people with good taste the tools to pick out and highlight items that should float to the top, and it keeps people from flooding your PSS with poor quality or uninteresting items.

Next time... wrapping up your PSS design with some infrastructure guidelines!