Large Internet companies spend millions on consultants and technology trying to get their sites to rank among the highest results on Google. Everyone else has to rely on the poor man's search-engine optimization: the link exchange.
If you've ever hung up your own shingle on the Web, you've probably gotten an e-mail to this effect at some point: "Dear So-and-so, I believe your site and mine could benefit from exchanging links." We probably get eight to 10 a week in the CNET News general mailbox, mostly from technology-related companies but occasionally from auto-parts suppliers and watch retailers who either have no idea what we do or few moral qualms about spam.
The idea is that if you can coax a link out of a large site like CNET, Google and other search engines will record that link as a vote of confidence in your site's worthiness and improve your ranking in searches for certain topics, thereby boosting traffic to your site. The technique is quite old, dating back even before Google and its PageRank system emerged as the Web's dominant search engine.
But does it still work? And at what point do two or three sites struggling to get off the ground veer off the road from mutual assistance to a full-blown spam operation designed to game the system?
Evan Duffield, for one, thinks it still works. He contacted us trying to get CNET to exchange links with WarpedAI.com, a site he has launched to promote stock-trading tools for day traders, and says he has been able to slowly build up the PageRank of another site he owns using techniques that don't run afoul of Google's Webmaster guidelines.
"It's kind of a vicious circle," he said. "To start a new business you need PageRank, but to get PageRank you need links to your service. You have to get the ball rolling."
PageRank is the currency of the Web. Google's novel approach to site indexing way back when was to evaluate the worthiness of a site based on how many other sites were linking to it, also taking into account the worthiness of the sites passing along the links. This meant, and still does mean, that a link from a site with a high PageRank counts for way more than a link from a site with low PageRank.
But how do you get a link from one of those sites? Google's official advice: "The best way to get other sites to create relevant links to yours is to create unique, relevant content that can quickly gain popularity in the Internet community." That, of course, sounds like something your mother would say.
In a Web as vast as this one, getting attention for a new site, even one with superb content, is a very difficult undertaking. Bloggers can discuss each other's work and help each other build up a following, but if you're selling a product or service it can be much more difficult to climb the ranks of search results for things like "day-trading software" when you're starting from scratch.
So Webmasters like Duffield turn to solicitations for links. Danny Sullivan, who writes about search-engine optimization for Search Engine Land, says "if you're a new site, absolutely you want to be doing link building. But you need to be doing that in a smart fashion."
Duffield says he's very careful to only solicit links from sites that are related to his product: his pitch for exchanging links that somehow wound up at our doorstep was addressed to computer-go@computer-go.org, a mailing list for hobbyists trying to tackle the difficult chore of building a computer AI system for the ancient game of go.
That was a mistake, he said; the result of prematurely hitting send on an e-mail template. Duffield compiles his targets by searching for sites that are related to finance and stock trading, and attempts to contact a general e-mail address to pass along his site's information and offer a link exchange.
"It's not about the actual links so much as it is optimizing search queries," Duffield said. "When I figure out a query I want from Google, I can see the top three positions have this much page rank and this many positions, and try to beat that out."
As long as people like Duffield are exchanging links without offering payment, or crossing obvious lines such as breaking captchas and posting spam links in guestbooks or comment forums, they're following the spirit of Google's Webmaster guidelines.
"Where it tends to get into tricky issues is where people are doing it primarily for payment," Sullivan said. "Search engines would see links as votes. Google does not like that people would simply be buying links to do better.
While paid links are clearly off-limits, Google appears to ban link exchanges in general, saying it does not allow "excessive link exchanging" but failing to define exactly what constitutes "excessive." Other practices that are verboten include links to "bad neighborhoods" on the Web and complicated networks of several Web sites with little content but pages and pages of links amongst themselves that Google can usually identify.
For the most part, however, the practice is rampant enough that only the most egregious violations get snagged. "If you start thinking too much about not getting caught, you're probably doing things you shouldn't be doing," Sullivan said.
In an era where SEO is a budding industry unto itself, link exchanges are perhaps the most basic approach. Far below the realm of those dithering over Google's search index are those like Duffield trying to make something out of literally nothing.
While he needs to build PageRank equity to get started, Duffield acknowledges that at a certain point that Google is right: a site will live or die on its content. Link exchanges only work to get one's name out there: the real boost needed to turn a Web site into a business comes when real people start discussing and linking to a service on blogs, message forums, and social-networking sites.
That's when your search ranking (and therefore traffic) really starts to grow, he said. "If you can make Google see that something is being talked about all over the Internet, what choice do they have?"
John Gruber of the blog Daring Fireball doesn't like what Digg's doing with its DiggBar, and has come up with a relatively simple way to block it on his own site. By making a small change to his site's PHP files, any shortened DiggURL created for one of his pages will automatically take users to a separate page Gruber has created that chides the company.
In Gruber's opinion, the service, which automatically shortens a site's URL and adds some of Digg's features to the top of the source content, is bad for both users and sites. Gruber says it's tainting the purity of a site's URL, which also affects search engine optimization and the capability for users to easily bookmark content. "URLs are the building block of the Web. They tell the user where they are. They give you something to bookmark to go back or to share with others," he says.
But is Gruber right?
As for the SEO, John Quinn--Digg's vice president of engineering, came out on Thursday and said that Digg had been in touch with search engines like Google and traffic monitoring services like Comscore, Compete, Quantcast, and Nielsen prior to launching the DiggBar, and that pre-launch testing had shown that it was not, in fact, changing the accuracy of traffic numbers. In fact, in the week since launching, Quinn said that the DiggBar was giving both Digg and publishers a noticeable boost in traffic.
One area where Gruber has a point though is with user bookmarking and link identification. Many sites use URLs that contain the headline, or certain keywords about the story. For readers this is a quick way to figure out where a link is going. For Digg, part of the problem with this is that its users can completely rewrite the headline and description of a story when submitting it to the site, which means the last way to see what something is--prior to clicking on it, is to check the URL, which is what the DiggBar effectively kills.
While Digg retains the source of the story right in front of the description, along with the full URL on the DiggBar, it's also changing what users are seeing in their address bar, which is yet another place where users are used to figuring out where they are and what they're looking at.
So how does this factor into bookmarking, and more importantly--social bookmarking? For personal bookmarking, Digg is replacing a site's Favicon (yet another identifier) with its own, along with replacing the site's standard URL with a shortened Digg one. Add a few shortened Digg bookmarks to your own personal bookmarks and you'll see where this can hinder the capability to sort, and quickly parse saved links.
For social bookmarking, sites like Delicious merely show the hottest links by page title (something the DiggBar does not alter), however when browsing the URLs alone, yet again it's a sea of Digg.com links.
Where Digg may have to change its tune is in giving publishers a way to opt out of having their site URLs shortened, along with a way to keep the bar from showing up on the top of the page. If Digg were to meet publishers in the middle, and act like any other link shortening service out there (TinyURL, Bit.ly, et al) and convert the Digg URL into the site's normal URL, I think it would go a long way toward preserving the happiness of publishers who want to maintain their site's identity.
While Google bombs and sites that operate as giant search engine optimization hacks can make it difficult to find what you're looking for on some search engines, sometimes it's even harder to get to the page of a product or service you saw a promotion for while out and about.
The sad truth is that nearly all of the good domain names are taken. We see it in the products we look at every day that have slight misspellings, missing vowels, or letters tacked on before, after, or sometimes in between regular words. These names can be incredibly hard to remember when it comes time to market a service, and in Japan companies are starting to realize that.
To compensate, companies are placing resources into making their official home page the top search result instead of going for the easy, and often times, expensive domain name. On a trip to Japan, blogger Cabel Maxfield Sasser noticed a barrage of ads for sites that did away with a long form domain (such as http://www.washingtonpost.com/) and instead put the company name in a search box. Of course part of this is because of Japan's phone-savvy culture, where even on the best phone keypads or mobile browsers, typing in extra letters can be a pain.
Sasser has posted a pictorial of the ads on his blog, many of which feature a mouse cursor on the search button of a made-up search box. In my own experience I've seen enough URLs of pretty great start-ups that end in .net, .us, or other odd .com names that I know would confuse my friends and family if they tried to type it into their browser's navigation box. Telling them to simply search for it, or read our blog post about it ends up being easier.
We don't spend a lot of time talking about SEO on Webware, but clearly it's important for these companies with odd .com domains or spellings if they're trying to reach mass markets. Services such as YouTube are now ingrained in people's heads with that spelling, but because it was so easily confused for UTube.com meant that Universal Tube & Rollform who owned the domain switched businesses entirely last year to cash in. The .tv's and .fm's have helped with video and audio services, but clearly going forward the land grab for domains might not be nearly as important as making your service the top search result.
[via BoingBoing]
There are a few niches of the "new Internet" that are just about full. Social networking that doesn't gear to a particular sub-group is one of them (though I'd argue that there's still ample room for niche social networks, like this one that we looked at a few days ago). So is broadband video. But start-ups in both those categories keep trying to find ways to stand out: one of those, I've seen, is by putting an emphasis on search engine optimization (SEO). In the video category, we saw this with Veotag. And on the social networking side, this is the philosophy of Ziki.
Ziki is, in many ways, a very basic social networking site. It consists pretty much exclusively--though there's no telling whether or not this will expand--of member profiles that aggregate links and content from other social media sites. This way, you can put links to your podcasts, videos, blog entries, and photos on your Ziki profile and have them all accessible from a common "hub." This is nothing new: other sites like OnXIAm are also based on this concept, though Ziki offers more social networking functions, like contact and personal information on profiles and an RSS feed. Plus, profiles can also be created for companies or organizations. But where Ziki really promises to be different is in the aforementioned SEO features. The site is currently, as a temporary promotion, sponsoring links at the top of Google, Yahoo, and MSN searches for free. So, if someone went to Google and typed in my name, the top link would be my Ziki (if I had one). It's not clear whether there are more SEO capabilities to the service, or if it's limited to the sponsored links.
The site is ad-free, at least for the moment, and it looks like Ziki's monetization strategy will be focused on premium content offerings. Beginning next month, members who cough up a few bucks will be able to buy their own search engine ads and customize them to their needs.
And there you have it: Shameless Self-Promotion 2.0.
- prev
- 1
- next
