In response to the question of what should the business model for badges/widgets look like (asked by David Beisel ) here is one possibility:
Vanity Fair’s MyStuff can be turned into a web service.
(Fasten your seat belts, this is a bit rough but I thought I would begin to put it down in writing and then ask for your feedback.)
Let’s call the new service “BrandRank” for the sake of illustration (the URL is already taken).
Users create an anonymous account. No real names. (Username and password) An email address would be required to verify the user but would not be available publicly (The issue of folks try to game the system by creating hundreds/thousands of phantom accounts would need to be addressed but I suspect there are solutions).
Just anonymous demographic facts. Zip code. Birthday. Male/Female. Weight. Height. Percentile of Annual Salary as broken down by the IRS (Top 10%, Top 25%, Top 50%, etc), Level of education (High School, Undergraduate, Graduate, PHD), any affiliations you wish to share (member of Museum of Fine Arts in Boston, civic groups, volunteer groups, schools, etc). The amount of demographic data is up to the user and can be added to or reduced later.
There would be 10 questions to start (We wouldn’t want the account creation process to take longer than 5 minutes). Each question would ask about which products you use. Car, television, cell phone, etc. That would get your brand portfolio started. The database would have thousands of product types listed and the product types could be created by users themselves.
Users would be able to “groom” their brand portfolios as much or as little as they like. Some folks might want to have a brand portfolio which includes thousands of things they own and others just a couple.
For each product type there would be pre-populated choices for brands and specific products (So you might have multiple cell phones from Motorola listed). Each item in your brand portfolio would be an HTML, Flash, or javascript badge which could be displayed anywhere such badges were accepted. If a user encountered a product category which did not list his/her brand of choice, s/he could use a javascritp bookmarklet to go to any URL and create their own badge. Each time a user added a new brand/product, that item would then be available for others to use.
Not only would the brand portfolio enable users to create items for products they currently own but items could be marked with the “Wishlist” tag or the “Owned Before” tag.
Take a look at THIS WEBSITE to get an idea of the power of user generated data provided by the book cataloging website LibraryThing.com
Brand companies could custom create badges which would be available for people to use. The companies would pay for the right to populate the database with “certified badges.” Badges might distinguish between actual owners, who would be required to register with that company, and folks who simply like the brand/product.
Each user account would be open (people could export their data) and anonymous. The anonymity is important because it will remove any fear that the user will be spammed in the future. Users would be able to export their data and then attach their real name to the list. But within the BrandRank website they would only be known by username.
Users brand portfolios display pages could be customized using GUI page generator tools. Users could choose to display only images or images with text. The thumbnail images could be large or small. Widgets would be available to display the portfolio of brands on one’s blog, web page, etc. BrandRank would be more of a virtual storage/display service which could be used much like folks use Amazon’s API. An API would be available.
Each user would have the ability to create an OPML file consisting of rss feeds for each product/brand in their portfolio. Companies would pay BrandRank to access the users through RSS/Atom. Users who have registered badges will receive semi-customized RSS/Atom feeds from the companies. Companies will be able to offer promotions directly to a known group of customers. The OPML file can be imported into any feed reader.
Each product category would have a real time ranking of brand popularity. Brand popularity would be searchable using Boolean search tools. For instance, I might want to search what’s the most popular hotel in London among people who own Honda Accords and drink Pete’s Coffee. Bad example, but you get the idea.
What’s the most popular brand for sneakers among 35 to 40 year old in metro-Boston? There would be “League Tables” of brand rankings on the front page of BrandRank. (Think DIGG.com)
Brands themselves will be powerless to buy ranking points. Rankings will be based on the number of people who choose that brand for their brand portfolio.
Features not covered but which would be present: tagging and group creation
Please send me your thoughts. This is a super rough draft. What got me thinking along these terms last night was the competition factor present in the Diggsig badge. I think people will be very curious to see how their brand portfolio’s compare next to others.
Revenue Streams:
- Certified Badges made available by the brands themselves. Essentially, each badge is a mini-endorsement
- RSS/Atom feeds to establish dialog and offer promotions to specific groups
- Affiliate Marketing opportunities for users who put brand badges on their web pages, blog, MySpace pages, etc
