Search engines are still the best directories

A while back I started using Chrome, and it has this clever little feature where if you type in “amazon” and hit tab it will search the Amazon site with Google Search.

I’ve always found Amazon’s search to be really poor. I’m not sure what it should be doing, but whatever it is doing it isn’t giving me what I want. But when I searched Amazon with Google it worked great.

This isn’t by accident: Amazon has put a lot of effort into getting good search results. They’ve built their website for Google. But that seems a little weird, doesn’t it? Why didn’t they improve their own search? They have all sorts of advantages: they know what items aren’t available, what items are popular, what you’ve bought in the past, etc. But from a sales perspective it’s not hard to understand: better on-site search means you are less likely to frustrate someone who wants to buy from Amazon, but better search results on Google mean you can get a sale from someone who is ready to buy from anyone. It requires a certain amount of humility to prioritize someone else’s service over your own, but it’s good for the bottom line.

One of the ideas that comes up frequently when talking about activism, civic involvement, non-profits, etc., is the idea of some new index. Who are the environmental activists in my area? What are the organizations dedicated to bicycle advocacy? What are the government organizations around me? Each of these generally involves finding a whole bunch of organizations and categorizing and indexing them. Sometimes the result looks like a directory (similar to how yahoo.com used to look), sometimes it’s geographic or tag-based.

There’s a lot of investment in creating a new index. You might take the work on yourself and try to fill the index, or you might crowdsource it – let everyone add themselves, spreading the cost around to everyone else.

But does it work? Can you convince people to invest time in it? Can you convince people to use it? Experience would indicate no. Not a definitive no, but a kind of lingering lack of success, where the only death is a slow realization of obscurity.

And it’s not surprising. There are several reasons I think custom indexes don’t work:

  1. People search with a task in mind. The goal isn’t “find an organization.” If you are concerned about a particular environmental issue, you search for the issue. If I’m worried about pollution in the park I live by, I should search for “pollution in Powderhorn Lake.” The results might include organizations, but they might be forum discussions, governmental bodies, or reports that were posted to the web. Organizations dealing with the topic are of interest, but they are not all that is of interest.
  2. You can’t trust any index to be complete. None of them are. Even in a local area, a hand-crafted list of links (the most basic of all indexes) is probably not comprehensive. It’s useful, but not authoritative.
  3. Organizations aren’t great at describing themselves. How can you tell the difference between what an organization does and what it wants to do? Few organizations honestly self-report the difference.
  4. Deep links are the best links. There’s two reasons for this: first, organizations aren’t great at describing themselves, so the best place to start learning about something may not be the front page. Google has put a lot of effort into getting people to the right destination, not just the intended destination. Second, to understand what is happening you want to see something topical. Maybe the organization is considering a particular issue, or is tackling an issue that doesn’t seem like an obvious fit considering its mission. It’s better to start at the center of an issue and then work your way out to understand who is working on the issue.

So here’s what I propose: if you want to help get people involved in issues, if you want to raise awareness of activist organizations, then do what Amazon does: search engine optimization. Make Google (or Bing or whomever) into your target audience.

Unlike a new index you don’t have to convince anyone of future potential to justify SEO; indexes come and go, but SEO works right now and will keep working into the future. It also gives you a chance to communicate with people who didn’t know what they were looking for, or haven’t made up their mind on an issue. And you can turn quantity of communication into quality communication, because people can enter into a topic or organization wherever it best matches their interests.

From a technical perspective, I might go further: if Google custom search doesn’t provide the best search results for your site, then the solution isn’t to provide your own local search, but to fix your content until Google’s search engines do provide the best results. It’s a content bug if search engines can’t give you good results for your site.

But most of all: go where the people are.  It’s search engines now, and it’s search engines as far as we can see.