Idea a Day #1

Like many people I have lots of ideas. Unlike most people I know that I won't be able to follow through on over 99% of them, which is also the hardest and most important part, so I might as well put them out there and hope that someone picks them up so that one day I can use the product I imagined existing. I'm considering doing this everyday for a month just because.

This is the most plausible one I came up with today.

I was selling something for a family member on eBay and going through the usual routine. I looked at the sale price of similar items sold and then averaged a few of them using a spreadsheet in Google Docs.
A simple spreadsheet I made in Google Docs that averaged recently sold similar items.
That'd be really cool if this was automated... for everything. What if there was a website that was a bluebook for everything? You could start by taking prices (automatically of course) from sold eBay actions and ranking them by lots of independant variables like condition, location, etc. You could do the same with prices from Amazon. Eventually you could pull prices from everywhere. I'm aware that some services already do something similar but let me expound.

Because of my (paltry) background in economics eBay has long fascinated me. In my opinion it operates about as close to Perfect Competition as anything; a few other real world examples would be commodities like corn, soybeans, etc. What defines perfect competition is very basically:
  • No barriers to entry or exit - on eBay all you have to do is set up an account
  • Perfect information - information can be found on just about any good via Google
  • Homogenous product - it's usually the case that there are already dozens of other items on eBay just like yours
So why not treat it like the market that it is? Something I find frustrating on eBay or Craigslist is people trying to sell items for near what they cost new. They will repost the item over and over never realizing that it's simply priced too high (same goes for real estate). This is known to psychologists as the endowment effect. Basically, people value something they own greater than what they would be willing to pay for the same object owned by someone else. Not logical but it's real, and the studies on it are fascinating. At the end of an art class participants would bid on eachothers art. Inevitably everyone would buy their own stuff.

Items on eBay and CL, regardless of format tend to fetch roughly same final price (with some caveats1), so why beat around the bush? Describe your item to said service, a price is generated, decide if it's worth it to you to sell it, and all the guesswork is gone. No more thinking your old laptop is worth 90% of what you paid for it.

Create a webpage where one can lookup the current market value of anything, new or used, available for purchase online by crawling the internet looking for the best sources for prices whether it be Amazon or completed listings on eBay. Treat it as a regression analysis where your model predicts prices, then compare them to real world data - refine your regression model over time. Someone else can figure out how to make money off of it. I just see it as a useful tool for providing more perfect information to buyers and sellers.

1 - Pennies from Ebay: The Determinates of Price in Online Auctions