NOTE: Please see the Company Name Test home page for more information about the results described here.
The Honor Role
MSN Search earns top honors in this test. Not only did it list all the candidate web sites at the very top of its main "Web Directory Sites" area, but each site was also well described and clearly identified as the candidate's official home on the web. The listing below is typical of the outstanding work MSN did for all the candidates:
Snap also did an excellent job. As with MSN Search, candidate sites appeared clearly labeled at the top of its main "Top Web Sites" area"
Snap would have tied MSN Search for first place, except that in the case of Bill Bradley, it linked to a page within his site rather than the actual home page. If Bradley were to delete or rename that internal page, users could get an error. Thus, Snap gets an A rather than an A+ for the Bradley search.
Ask Jeeves didn't place links to candidate web sites at the very top of its results, as did MSN Search and Snap. However, the links were so close to the top and so clearly labeled that the service earned a score of A all around. Below is a typical response. I've highlighted the link to Al Gore's official site with a red box:
Netscape Search also turned in an impressive performance, listing three candidate web sites at the very top of its main "Reviewed Web Sites" area and flagging them appropriately as "Editor's Choice," as with this example for Bush:
Netscape only stumbles when it comes to Bradley, failing to list his official campaign web site until near the bottom of its results. That earned it a score of B, for that search, and A+ scores for all the others.
Like Netscape, Google only failed to get a candidate web site to the very top of its results in one case, and that was because it ranked Al Gore's official White House site over his campaign site. I've highlighted the campaign site with a red box, below:
You can see how Google not only lists the White House site first but also "indents" other internal pages from the White House site immediately below the main White House listing. I'd prefer to see Google not list more than one page per web site in its top results, which would have had the effect of moving the campaign site higher. It would also provide more variety to Google's search results. I'd also like to see Google support the meta description tag, because I don't find its automatically generated summaries to be very good.
These are relatively minor concerns. They kept Google from earning A+ scores all around, but the service still did a superb job at listing all the candidate web sites.
Go managed to get all the candidate sites at or very near to the top of its main "Web search results" area. Unfortunately, its score was reduced because of dated descriptions for Bush and McCain that made it seem as if they were still considering running for president, rather than actually doing it. For instance, look at this listing for Bush, which I've highlighted with a red box:
Finally, comes iWon, which got a slight ding for using the bushsux.com address for Bush's web site, as I've highlighted with a red box, below:
Yes, Bush does appear to control bushsux.com. Clicking on this link will redirect you to the official web site. Still, it would be nice for iWon to find a way to list the official address of georgewbush.com, instead.
The B Team
Lycos-owned HotBot succeeded in the test for all candidates but George W. Bush. However, rather than failing the Bush portion of the test entirely, HotBot earned a C+ because it did provide an alternative way to find Bush's web site.
At the top of its results page is a "From The Lycos Network" area where internal content is promoted. The top link there, which I've highlighted with a red box in the picture below, led to a nice summary of Bush's candidacy. This summary, in turn, had a link to his official web site.
The same thing happened at AOL Search, for Bush. His site was not in the top results, but there was a prominent link to internally produced election material, which listed Bush's site. I've highlighted this below, with a red box:
As with HotBot, this alternative route earned AOL Search C+ rather than an F for the Bush portion of the test. AOL Search earned Bs or a B+ for the other searches, because the candidate web sites didn't appear right near the top of its main "Matching Sites" area.
The story again repeated at Yahoo, only this time it was Gore's site that didn't get listed within the first page of the results. Gore's site was accessible only to those who detoured through the Gore category that was top ranked for his name, as I've highlighted with a red box, below:
While the other candidate sites were listed in the top results, Yahoo failed to distinguish them or move them to the top of the list, which really ought to happen. For instance, notice how McCain's official site is buried among the others. I've highlighted it with a red box, below:
Yahoo earned a C+ for the Gore listing and B- scores for the other candidates, because of its failure to make their sites more prominent.
Finally, the situation at Excite followed the story that is now so familiar. Like Yahoo, it failed to list Gore. It also failed to list Bradley. However, similar to the other services earning an overall B grade, Excite did provide prominent links to internal content that in turn led to the missing candidate web sites. For instance, here is the category that appeared for Gore, which I've highlighted with a red box, in the picture below:
By the way, Excite's score for Bush was downgraded to a B- because the listing it displayed was dated from October 1999.
"C" Means Could've Done Better
Lycos leads off the services that scored merely average scores. Only Bradley's site was listed in its top results. The other candidates had to depend on people finding them by clicking through to the internal election content that Lycos promotes at the top of its results pages, such as this for McCain, which I've highlighted with a red box:
Direct Hit lost points for failing to list the Bush site, and while the Gore site did appear in its first page of results, it was at the bottom of the page. Oddly, in the results that Direct Hit provided to HotBot, the Gore site was top ranked.
GoTo managed to list three out of four candidate web sites, but its pay for placement system meant that these sites appeared toward the middle of its listings, only after paid results.
"D" Is For Disappointing Results
AltaVista put in a poor performance, only managing to list the Bush web site within its numbered "Web Pages" results. Moreover, that listing dateed from September 1999, making it sound like Bush was still just thinking about running:
Where AltaVista failed, its RealNames links stepped in to help. They directed people to the right sites for McCain and Bradley, though Gore's RealName link led to his White House site, not his campaign site. An example of the RealNames link for McCain is highlighted with a red box, below:
The RealNames links are prominent in AltaVista's listings, but other search engines such as MSN Search succeeded both with their RealNames links and within their own listings. For this reason, I'm only gave C+ scores to AltaVista in the two cases where it had to rely on RealNames because its own information wasn't up to snuff.
FAST Search came in just behind AltaVista, managing to get the sites of McCain and Bradley to the top of the list but leaving Gore and Bush in limbo.
It was nearly the same at Northern Light, the difference being that the McCain site wasn't at the very top of its results, as was the case at FAST.
Finally, LookSmart brought up the rear of this group, which is amazing when you consider that it was LookSmart information that provided the main listings for the service that took top honors, MSN Search. This shows that it is one thing to have good information and quite another to present the information in a useful manner.
LookSmart failed to list any of the candidate web sites in its top "site matches" results. In the case of Bradley, it did suggest a category for Bradley that in turn listed his campaign site. This category is highlighted with a red box, below:
A category for Bush also appeared, but rather than leading to a single page, it led to several choices, none of which stood out as the place to go to find his campaign web site.
Fortunately, LookSmart could fall back on RealNames links, as did AltaVista. These helped in the cases of Bush and McCain. An example of a link for Bush is below, highlighted with a red box:
Unlike AltaVista, I only credited these RealNames-related successes with D scores. Why? Because the RealNames links did not consistently appear. Sometimes they would be present, then I'd reload the search results or try a few hours later and find that they were gone.
Finally, Gore's web site wasn't listed in LookSmart's top results; the RealNames link led to his White House site, not the campaign site; and no good categories were listed for him. That left LookSmart with one last backup -- the category it displayed for each site listed. For instance, below is the top site that was listed for Al Gore's name on LookSmart. I've highlighted the "home" category of this site with a red box:
If someone were to have clicked on that category link, they would have indeed found the Gore campaign site. However, I think the placement below each web site doesn't promote clickthroughs, so I only credited this with a D- score.
Optimising Digital Marketing Campaigns with Search, Social and Analytics
At SES London (9-11 Feb) you'll get an overview of the latest tools, tips, and tactics in Paid, Owned, Earned, Integrated Media and Business Intelligence to streamline your marketing campaigns in 2015. Register by 31 October to take advantage of Early Bird Rates.