SearchEngineWatch.com, Feb. 29, 2000
NOTE: Please see the Company Name Test home page for more information about the results described here.
A letter grade was given for each part of the test, resulting in an overall grade for each search engine. A full explanation of scoring is below the chart. The test was conducted on Feb. 28, 2000. The Major Search Engines page has links to the search engines listed on the chart.
|Search Engine||george bush||john mccain||al |
The overall goal of this test was to see if direct links to candidate web sites were present in the first page of each search engine's results, when searching for the candidate by name (the chart above shows the exact text used -- searches were done in lowercase, as this is how most people commonly search).
A perfect score would be given if a search engine managed to make the candidate web sites appear at the very top of its listings. Unfortunately, each search engine presents its results differently and many commonly provide information from different sources. This makes scoring incredibly difficult.
When testing, I focused on what I considered to be the "main" results for each search engine, the area that I assumed most users will gravitate to. I encourage you to read the Illustrated Guide To Scores page, to better understand what I considered to be the main areas for each search engine.
In general, letter scores were assigned following the criteria below, modified with + or - as I felt appropriate. The Illustrated Guide To Scores page also provides further information.
|A+||Campaign site was listed at the very top of the main search results.|
|A||Campaign site was listed at or near the very top of the main search results, but other factors caused the score to be reduced.|
|B||Campaign site was listed on the first page of results, but not especially near the top. Other factors may also have reduced an A grade.|
|C+||Campaign site not listed, but an alternative route to finding it was prominently available.|
|D||Campaign site could be located through some alternative method, but it was not found to be especially effective.|
|F||Campaign site was not listed at all.|
Please don't try to average out the letter grades using criteria such as A = 4 points; B = 3 points, etc. I did use consistent numerical values to guide me in assigning letter scores, but I also used a curve to score each search engine against the group as a whole.