Sometimes, webmasters are told to submit "bridge" pages or "doorway" pages to search engines to improve their sites' traffic. Doorway pages are created to do well for particular phrases. They are also known as portal pages, jump pages, gateway pages, entry pages and by other names.
Doorway pages are easy to identify in that they have been designed primarily for search engines, not for human. This page explains how these pages are delivered technically, and some of the problems they pose.
There are various ways to deliver doorway pages. The low-tech way is to create and submit a page that is targeted towards a particular phrase and keyword. Some people take this a step further and create a page for each phrase and for each search engine.
One problem which it would caused is that these pages tend to be very common. It's easy for people to copy them, make some changes, and submit the revised page from their own site in the hope of achieving any success. Also, the pages may be so similar to each other that they are considered duplicates and automatically excluded by the search engines from their listings.
Another problem is that users don't arrive at the goal page. Say they did a search for "golf clubs," and the doorway page appears. They click through, but that page probably lacks detail about the clubs you sell. To get them to that content, webmasters usually propel visitors forward with a prominent "Click Here" link or with a fast meta refresh command.
By the way, this gap between the entry and the goal page is where the names "bridge pages" and "jump pages" come from. These pages either "bridge" or "jump" visitors across the gap.
Some search engines don't accept pages any more using fast meta refresh, to curb abuses of doorway pages. To get around that, some webmasters submit a page, then swap it on the server with the "real" page once a position has been achieved.
This is "code-swapping," which is also sometimes done to keep others from learning exactly how the page ranked well. It's also called "bait-and-switch." The downside is that a search engine may revisit at any time, and if it indexes the "real" page, the position may drop.
Another note here: simply taking meta tags from a page ("meta jacking," as Infoseek calls it), does not guarantee a page will do well. In fact, sometimes resubmitting the exact page from another location does not gain the same position as the original page.
There are various reason why this occurs which go beyond this article, but the key thing to understand is that you aren't necessarily finding any "secrets" by viewing source code, nor are you necessarily giving any away.