|
|
 |
 |
 |
|
July 10, 2006Thoughts on Aggregationthinking out loud... In thinking about ways in which sites can be better than other sites -- or to put it in more concrete terms, who actually deserves the top spot when you search for "widgets" -- there is not much agreement on what to value. One old standby is unique content -- the amount of content that has not been duplicated elsewhere. Webmasters get in the most fantastic pissing matches arguing about who has the most unique content, how unique it is, etc. This is one useful thing to consider, but it is not the only way to add value, especially in comparison shopping. In comparison shopping, you can have huge value-add without unique content per se. (example: FatLens ticket search) Simply aggregate information that is available elsewhere, and it is the scale and the style that adds value. Anyone can grasp this concept if you mention an example like Google, but on smaller scales, aggregation does not always get the same respect, and is not viewed as a type of content. Why? Granted, aggregation must be significant to be useful. If one shopping engine lists 45 offers for a product, and another shopping engine lists 50 offers for the same product, no big deal. At some point the law of diminishing returns kicks in for aggregation, limiting the aggregation upside for new general-purpose shopping engines. The thing that amazes me lately is how many new sites there are without unique content like reviews -and- without any signficant aggregation... like, if Shopzilla lists 50 offers for a product, these sites will list maybe 5... or 3... or 1? Now, if you go niche, it is another story, and there are plenty of opportunities to out-aggregate larger competitors. Say, if I'm working an event niche, it is perfectly reasonable to have 3x, 5x, even 10x more listings than craigslist. Then again, if there is an event engine for a city, they might be able to aggregate severalfold the events of my site. I've mentioned aggregation in terms of offers and listings, but there term could just as easily apply to attributes. Across all kinds of products, from common to uncommon, how many attributes remain un-slice-and-dice-able? How cool would the web be if you could slice and dice on any attribute of anything and find all matching items? Although this would come from a Google or Yahoo, in the near-term, I think we are overdue for an explosion in aggregation and attribution from niche sites. The large sites have more tech, but small sites have more manpower, in the aggregate. All they need is a little more tech, multiplied by their manpower, and it is a huge leap forward. Any hypergrowth in aggregation has to include "the edge" to pick up where traditional centralization leaves off. Judging by the amount of cool tech that is trickling down to small sites these days, it is only a matter of time until more aggregating/attributing tech finds its way into the hands of the average webmaster. Like it did with forums. to be continued...
Thoughts on Aggregation
|
