How does Google decide what it shows in its listings?

I am helping build the website MarketsWiki and having trouble getting Google to properly display relevant information about the web site.

When I do a Google for “marketswiki” I get the following text snippet in the results:

“MarketsWiki gets the bulk of the funding it needs from direct sponsors. We have a number of excellent sponsors already (the companies to the left of this …”

That of course is nothing like what the site is about. I noticed however it is about the first static text on the page. The stuff before that are news story links and change daily. My guess was Google latches on to the unchanging portion as the parts it wants. However it grabbed the second sentence of static text and not the first. To confuse things further Yahoo Search grabbed something different (but seems to be operating along the same lines).

So, I was set to arrange for a better description somewhere on the page but noticed the Google entry for Wikipedia includes a text snippet that is nowhere to be found on their front page (nor in the HTML code underlying it).

I am not out to game the system. I just want the relevant bits to display properly. I know Google ignores meta tags and reading their site they seem to pretty much say make your page descriptive. But how? And how do I get it to grab parts that make sense rather than any old piece of random text?

Try looking at your website with a text-only browser, with all scripts & java turned off. That gives you an idea of what Google is seeing when they scan the site.

Then re-design your site, so that the main point, that you want Google to include with the link, is clearly shown on the top.

Hmm, that’s a good question. My first guess would be to use the meta description tag, that may work.

The description Google is using for one of our pages came from our submission of the page to http://www.dmoz.org/, the Open Directory project. But it took many many months for them to add it, I didn’t even realize it had happened until I saw the Google description for our site doesn’t match anything in the HTML code.

Google explicitly ignores nearly all meta tags. They used to read them until unscrupulous web developers (often porn sites) would put anything and everything in there so a search for (say) “catholic churches” might list a porn site.

Consider using Google Webmaster tools to help keep tabs on what Google is doing with your site on their search engine.