BUSINESS IDEA: Instead of giving away personal data to competition, offer them for sale with high granularity

Hi,
watching on of those advertisements for car insurance, I started dreading again the task of selecting another insurer for the family car. Much rather I would offer some relevant information or even a reference from the current insurer and wait for some good offers to come in.

Spinning this thought further it becomes apparent that selling of some personal data is already common place: as subscriptions to mailing lists, detailed searches in company websites or simply tracking cookie. Only, there is no direct profit to the owner of the data as well as no incentive of keeping the data accurate and up-to-date.

The idea works as such: my personal data is stored in a library encrypted together with an access list. A data provider (i.e. the user, person, company, individual, etc.) will have the encryption key -- possibly using asymmetric encryption.

The data provider can then reverse advertise for an offer of goods or services by subscribing to categories and supplying a minimal set of data. For a car insurance perhaps the number plate, age and recent claims history for example.

For solicited access a data consumer (i.e. insurance company, shipping company, market research organisation, etc.) will have to ask the data provider for one-time-access to more of the personal data in order to complete an offer.

The interesting part is the idea of selling some of the information, such as household income statistics in a reverse google adwords fashion. For example the data consumer would put a request out for all data provider with an interest in gadget, a disposable income of X amount and in a certain region. Data providers can then subscribe and offer some more of their information for a guaranteed monetary amount. Of course, just like ebay and google a certain amount of lying can be expected and need to be factored in. Reputation points are on option. Another more drastic possibly proof of location, age and income may be accepted by the website to purchase creditability points in order to help weed out some of the abuse.

Finally, for market research a statistical group subscription could be offered where data provider are classified into groups of age, location or income. An interested market research company could purchase. Again with some small reward to the data provider.

On a grander scale this system can be extended with electronic voting and/or backed by a call center for companies to purchase as sales channel.

The biggest drawback is the privacy issue. Everyone conscious of their data being used against them, such as questions related to health issues or accident history should be rightly concerned. A decent amount of safety can be guaranteed as the data is encrypted to be used at the immediate and temporary discretion of the end user only and all of the statistical and "unsolicited" access is done via a highly granular opt-in. Perhaps 10-30% of the personally sensitive data of random data provider should be withheld as a matter of business rule to avoid the absence of information to have a meaningful value to the data buyer.

Another drawback is data harvesting by companies or company groups sharing this information after purchasing. Unfortunately, there is not much that can be done with the exception to legally require a NDA (non-disclosure-agreement). Interesting options are emitting a significantly insignificant amount of false data, such as misspellings in name or a few days off the age, as tracer to spot companies abusing the NDA. This can only work as deterrent but not as a guarantee. And as the past storage of credit card details has been show some data leakage can be expected even when taken all precautions.

Credibility is the most important asset for this business. Hence it becomes vital to offer to the data provider an audit trail of every access and search specifically to the providers data and details about the encryption.