David Hardtke's Blog

While developing Stinky Teddy, I thought quite a bit about the business of search, from the perspective of the consumer, the search engine, but also from the perspective of the content providers. Most of my thinking of course was about how to build a search engine that was attractive to consumers -- how are the needs and wishes of the consumers not being fully served by Google? The only way to get people to use your product is to give them a compelling reason to do so. We're trying to build a search engine that is better than Google some of the time, as good as Google most of the time, and not ever substantially worse than Google. If we can achieve that, we hope that people we'll trust us with some of their precious attention.

Search, however, is a funny business. The profit margins in search are very high. As with all things software related, the marginal costs of an extra customer are very, very small. Most of the work goes into the development of the software and the buildup of the hardware infrastructure. As a metasearch engine, we don't have to do our own crawl of the complete internet so our fixed capital costs are very low. The portion of our costs that scales with our level of traffic (the extra computing resources, bandwidth, server administration, etc.) are very low (somewhere around one hundreth of a cent per search).

We aren't trying to monetize our traffic to begin with -- we're more concerned with gathering important usage data that will allow us to build a better search engine. Eventually, however, we'll start making money by putting sponsored links next to and (more lucratively) above the search results. The amount of revenue that this will generate depends on many factors (user demographics, types of searches performed, quality of advertisers), but the big boys in general purpose search (Google, Yahoo, Bing, AOL, Ask) generate an average of 10 cents per search for US traffic. A business with unit cost of one hundreth of a cent and unit revenue of 10 cents sounds pretty good, eh?

Typically, businesses with absurdly high profit margins attract competitors who gain market share by undercutting the price. This is where search is a funny business. From a consumer perspective, search is free. You can't undercut a price of free.

I'd argue, however, that search is not really free from a consumer perspective. Huh? There's no Paypal widget on Google. Chris Anderson recently wrote a book called "Free: The Future of a Radical Price" that explained why it is impossible to directly charge for anything on the Internet (disclosure -- I haven't read the book, only reviews). People want stuff to be free, and who can blame them. But free and Free are different. In my mind, something that is "free" is costless to the consumer, but something that is "Free" is paid for by the consumer in non-monetary ways. For the consumer Internet, "Free" most often means that the publisher sells our valuable time and attention to a third party.

We've all made an unconscious and unwritten deal with the search engines. Give us what we want for free most of the time, and we'll let you sell our most valuable moments to the highest bidder. When our attentions turn to valuable activities (i.e. we're about to spend some money), the sponsored links show up above the search results.

People pay differently. Most people claim they never click on sponsored links. For these people, there is a mental cost to identifying and skipping the sponsored links, and this filtering process necessitates absorbing some of the commercial message. Other people don't know the difference between sponsored links and normal results and simply click on the first link they see -- these people are the ones who subsidize the search engine for the rest of us (we owe them our gratitude). These folks pay directly through higher prices for the goods they buy (the money paid for that click has to be built into the price of the good sold).

Our "monetizable moments" are fairly infrequent, so this is a pretty good bargain for the consumer. Most of the time, we get what we want for free. The search engine gets to control our attention when we are truly hot commodities -- they serve as the gatekeepers to our wallet. This system is good for them.

One group, however, gets hosed in this economic system. The people that suffer are the journalists and other original content providers. It's very simple -- they provide much of the content that we consume when we use the search engine for free, but get none of the benefit of our "monetizable moments." Sure, they get to show some banner ads next to the articles we read, but by the time the consumer has reached their page he or she is essentially worthless. The journalists provide a disproportionate share of the material that appears on the search engine, but gets almost none of the revenue. It's kind of like a contractor who builds and sells a house but doesn't pay for the concrete in the foundation.

While thinking about this, I came up with the perfect analogy. Search Engines are like radio stations. With radio, consumers listen to songs for free in exchange for being forced to listen to an occasional advertisement (the "monetizable moment"). Songwriters are the foundation of the whole system. The big, big difference between search engines/journalists and radio stations/songwriters is that radio stations pay the songwriters performance royalties. A performance royalty is a fixed fee paid to the composer of song whenever the song is played. Why not do the same thing on the Internet? Why not pay the content creator whenever an aggregator (e.g. search engine) links to and quotes from an article?

Although this will be bad for Stinky Teddy (added costs), I put together and submitted a proposal to the Knight Foundation News Challenge (Performance Royalty System for Online Journalism). The proposal outlines this idea and proposes to implement the technology and build the needed non-profit organization.

What I'm advocating is a radical change in the economics of the Internet, but quality journalism is the foundation upon which much of the Internet rests and should be kept viable.