Great Things

The problem with social search

Greg Sterling at the Kelsey Group pointed me to a New York Times article (reg wall) on the issues with relying on consumer reviews for advice on hotels. What are the issues:
1) The hotels are managing the review process by either offering incentives (like discounts) to consumers who review the property or in some cases manufacturing fake reviews.
2) As a consumer, you really don't know the identity of the reviewer and whether you have anything in common with that person.

These issues are common across all of social search -- not just in the travel space. How do I know the Judy's Book review on a local dentist wasn't submitted by his dental assistant? The real answer is you don't. The Judy's Book team would answer that they are developing an algorhythm that ranks the trustworthiness of the reviewer.

So what is the solution? I like getting advice from two types of people: people I know personally (or friends of friends) or people I share interests with. This solution solves the travel problem by taking hotel advice from others in your social circle who have stayed at the property. This solves the identity issue. Shared interests is a little bit tougher. In the hotel space, I would take advice from travelers who liked properties that I like. For instance, if I like the Hotel de Russie in Rome, despite the fact that I can't afford it, I would be interested in hearing Paris hotel suggestions from other travelers who also liked that hotel.

Data: The New Intel Inside

Tim O'Reilly and John Battelle talk about data as the "Intel Inside" of Web 2.0. This reminds of a version of the golden rule printed on my brother's childhood ruler: "whoever has the gold rules." In the case of web 2.0 companies, whoever has the data rules (or generates outsize returns). Before getting into O'Reilly's core thinking, I want to digress to talk briefly about how much I love information businesses. It is really pretty simple - you aggregate data from as many sources as possible, clean it up and then resell it. Although there is a high fixed cost associated with building the databases, there is virtually no variable cost in running queries against the data. Barriers to entry are only significant when you layer on data layers that you own.

O'Reilly gives two wonderful examples. One of how to build competitive advantage (Amazon) and one of how to let opportunity slip away (Mapquest). Mapquest failed in the mapping space, where failure is defined as not building competitive advantage, because they  were licensing the mapping data and they were not layering on any of their own data. Amazon has been successful since they took the ISBN database and aggressively layered on new sources of data; in the process, they have created enormous value.

What other examples are there? The yellow pages space is interesting. The big yellow pages companies are desperately trying to innovate and become more than just a directory. But the upstarts like Judy's Book are having much more success acquiring user-generated content and layering it on top of existing yellow pages data. Just as Amazon created value doing this, I think that Judy's Book is on the right track here as well.

So my reminder to myself today as I go about trying to build an information business is pretty simple. Aggressively acquire data from as many sources as possible and then layer on user-generated content on top of that to build competitive advantage.