Lately, I have been thinking a lot about information arbitrages. What are information arbitrages (AI)? They are instances where the spread between the cost of accessing a piece of information and your ability to resell it is sufficient to be able to make a profit.
A simple example of an AI play in the market today are keyword harvesters like Become.com that purchase a block of "horizontal" search keywords from the Googles of the world and then build a layer of verticalized, or job-specific, value add, such as comparison shopping or product research, on top of these search returns. This layer enables them to build a nicely profitable business.
In fact, in the realm of "eating my own dog food," I am in beta on a stock investing service called Insider Engine that leverages an information arbitrage specific to the buying and selling activities of corporate insiders. Through this arbitrage, I capture the spread between the release of publicly accessible data and when that data is "priced into the market" of a given stock.
When you peel back the onion, what an information arbitrage really is, is a kind of algorithm. Algorithms are akin to recipes, where there is a given set of ingredients, in a known measured amount, applied in a given order using a given set of combinational logic, resulting in some fundamental and predictable outcome.
With recipes in mind, and in the spirit of Thanksgiving, here's one type of information arbitrage which I am baking into my own activities that I call Online Local:
- Start with a "job" that has a very specific context but one where the actual fulfillment of the job takes place at the local level. A simplistic example to get you started is dating, since the parameters of finding a mate (or just a date) are fairly specific and ideally you would like to meet someone who lives in your own backyard.
- Look for instances of this job where 90% of the offering is generic or templated so that the same basic offering is able to be deployed across all geographies and a wide range of markets (the proverbial wide net), but where the final 10% is customized to be geography and/or market specific, and can be applied in a systematic, automated fashion.
- The more complex the configuration, the larger the ASP (average selling price) and/or the more mission critical the "buy decision," the greater the potential for material value add in the arbitrage piece of the equation.
- Your ability to piggyback on existing sales channels increases the more steps in the pre-sales qualification process that you are able to move to an automated (read: cheap-to-free) online environment which have historically been handled in an expensive offline manner. This is pure cost reduction, which drops directly to the bottom line, so long as it does not diminish the the prospective customer's perception of the buying experience.
Put the recipe in the oven, bake it at 375 degrees for 45 minutes, and you have a deliciously profitable contextual mashup that scales into a nice cash flow business. Feel free to ping me if this spurs any tasty ideas. Enjoy. :-)
