Tag Archives: pagerank

Link journalism and the no-follow tag

It appears there is a whole discipline around how journalists use links in posts: link journalism. This is a thorny subject as it involves passing authority onto a third party over who you may have no control.

Ryan Thornburg offers a cheat sheet on curating links to his journalism students. Items covered include verifying the accuracy of any links you refer to, in addition to making sure links are as relevant and specific as possible. Links can give background, technical information, audio, video, the list goes on. This makes them a significant tool in online journalism.

One thing Ryan doesn’t touch on is the importance of the link to search engines. Google helped make search an integral part of the web experience for most of us. Their secret sauce to finding the ideal results to list involved looking at the link structure between pages (often termed pageRank). Pages that have more links flowing in are considered authorities. These pages confer some of that authority down to other sites when they link to them.

This particularly favors large encyclopedic sites. And that seems to work for us. What better source to find information on Neanderthal Man than Wikipedia? (Yes, Wikipedia is the number one link in Google). University sites and news sites also fare well under this system. So if a news site passes a link on to another site, it helps the destination site gain prominence in search (nothing to be scoffed at: I’ve seen traffic increase ten-fold for long periods of time – all due to high rankings in search thanks to a single link from a high authority site.)

In most cases this is all good. The problem comes when you are writing about something on the web in a negative light and want to link to that page to highlight your point. In the world of search, your link is taken as a positive referral – so a damning report in a reputable news source can work wonders for your search engine optimization (SEO).

To get around this (and help combat the growing problem of comment spam), the search engines got together in 2005 and created the ‘no-follow’ attribute. Add this to your link and the search engines would take it that you want to make a link for explanatory sake but you do not want to pass on that all-important link juice to the destination.

Funny thing is, it is still rarely used in the media. Case in point: take the recent gripe from the UK’s Telegraph that the Wikileaks showed political bias in distributing key information on the war in Afghanistan to to the left-leaning press (including its competitor, The Guardian):

See the link on the first instance of ‘Wikileaks’? It’s a straight link not including the ‘no-follow’ attribute.

So while Will Heaven, the Deputy Blogs Editor is lambasting Wikileaks, he’s also providing the site with a valuable link that will help it grow in prominence. I’m somehow not convinced this is the desired effect of his vitriol.

I’d call on journalists to make better use of the no-follow tag. New journalists coming through the ranks should be made aware. ‘Think before you link’ should be an integral part of link journalism education. The link is a powerful tool and should be handled with wisely and responsibly to maintain the integrity of the web.

How do you track leads from Twitter posts? The bit.ly URL shortening service is one way

The very sharp Sandy Carter asked a question today that had been playing in the back of my mind but being the weak marketer I am, I never really fully got my teeth into: if you put out a promotion on Twitter, how can you measure the effectiveness of that tweet?

Twitter, like every other web presence, should allow for easy measurement. So if your promo attracts 1,000 clickthroughs from your well-crafted tweet you should know about it. Right? Unfortunately those wonderfully effective URL shortening services that that help you keep your Tweets under 140 characters can get in the way.

For instance if the shortening service uses Javascript for the redirect, this can show odd referral data in your web logs. Another scenario is where you use the same link across different Twitter accounts (say, if you hit that magical pulse and find yourself heavily retweeted).

If you use any of the Twitter clients out there, you’ve probably noticed that many offer a host of shortening services. Each of these has its relative merits. So if you are in the position where you really want to know how many clicks emanate from your wonderfully crafted Tweets, some services work better than others.

I thought I’d ask my diminutive Twitterverse what they thought were the most ‘trackable’ URL shortening services. Richard Barley proved most helpful, pointing me in the direction of bit.ly.

A key advantage of this service is it doesn’t matter whether or not you create an account: you can still have access to the excellent stats, which start off like this:

bitlyThere are further stats on geography and which accounts have used the shortened URL and crucially you can see if there are other bit.ly codes for the long URL.

The advantage of creating an account is that you see all your links collected under one interface (great for comparison). If you sign up for a premium account you get even more features such as detailed referrer stats. Interested? Here’s a few examples of how the service works:

Twhirl and bit.ly

To use bit.ly with the Twhirl desktop client you need to setup a bit.ly account. Then go into the tools section and find the API key. Twhirl will ask you for this information when you try to shorten a URL. This is great because all the URLs you post using Twhirl will be listed in your bit.ly account.

Tweetdeck and bit.ly

Tweetdeck takes a different approach by not asking you to sign in (for the record, bit.ly is the default URL shortening service in Tweetdeck). This makes it easier to shorten URLs right off the bat without having to setup a 3rd party service. But then how do you track links? bit.ly has a nifty use for the ‘+’ sign. Append this symbol to any of the bit.ly URLs to receive stats on that link. Here’s an example:  http://bit.ly/45dHIx+.

Web browsing and bit.ly

You can incorporate bit.ly with any web browser. Once you set up an account, you can drag the boomarklet to your toolbar and use this to shorten the URL, with the option to post directly to Twitter.

There are other uses of the service and it integrates with other clients (and I’m sure the iPhone), so if you have more information, feel free to comment. Now let me drift into speculation on the power of bit.ly and what it means for the web.

Taking on Friendfeed

I’ve been a heavy user of Friendfeed for some time: it’s a great tool for taking content from anywhere (including Delicious links) , aggregating it, and posting it on to Twitter/Facebook (for the uninitiated, here’s an explanation of Friendfeed). However, Friendfeed has gone down the route of concentrating more on engaging discussion around the content and does not yet offer publishers decent stats on which links have attracted most clicks. Is this a potential Achilles heel? It’s making me reconsider recommending the service for business/marketing purposes.

Taking on Digg

Beware Digg, you mighty news aggregation service: there could well be a significant player in town. Gigaom recently reported that bit.ly has received $2 million in funding – interesting given the statistics bit.ly offer could turn it into a compelling Digg-like service. When people use bit.ly to shorten a link, they are indirectly voting for it (in much the same way Google treats a link from one site to another as a vote in the PageRank algorithm). If you pull all this information together, you have a great social bookmarking system. Those links that have created the most stir (eg. through retweeting on Twitter) will be ranked the highest. I haven’t seen in practice and there maybe some kinks to iron out (how do you categorize links?) but the idea is intriguing.

Taking on Google???

OK, so we’ve stated that these links become votes. You could also argue these links are more social than the ones that Google tracks by looking at website relationships. Bit.ly already has a search engine, but this could bloom into a major feature if the service becomes widely adopted. It’s been noted that Twitter is replacing Google for certain queries (I received better information for this article from Twitter than from Google), and as bit.ly is a link between Twitter and the wider web, the service is in a good position to show which web pages are most useful to people. Why doesn’t Google report on these links too? At the moment many of these services fall outside of Google’s domain given that the Javascript tracking they use can render links invisible to the major search engines. This goes for most URL shortening services: not just bit.ly (although some services are beginning to offer more Google-friendly links).

Will Twitter get into the fray?

As a final point there is much speculation floating around about how Twitter will moneytize its service. Offering premium services to corporate customers seems highly likely. Could one of these services be the kind of statistics that bit.ly dishes out, if Twitter gets into the URL shortening business?

I, for one, think the URL shortening space is looking increasingly interesting, and at this point bit.ly is well poised as a major player.