Cut the Links?

Since the Penguin algorithm changes, SEOs have been worried about a couple of things in particular. Firstly, they are concerned that all the illicit link-building tactics that their client’s previous SEOs engaged in (because, of course, they would never do such things themselves) are now going to have a negative consequence on a site’s ranking. The second major concern is that competitors can take advantage of Google’s scrutinizing of backlink profiles to deliberately create “bad links” and incur a penalty.

To allay some of those fears Google have released a tool that will allow webmasters to disavow those incoming links that might be causing their sites to be flagged by Google as engaging in bad link-building practices.

The Disavow Tool allows webmasters to upload a text file of domains and URLS that Google will then disregard, in much the same way that they disregard nofollow links. This is a somewhat out of character move for Google, who, as acolytes of the algorithm, prefer to rely on machine intelligence to winnow out the chaff. As usual, Google will take these link lists as a strong signal, rather than as an instruction that they are bound to follow. Continue reading →

(Source:Flickr/adactio)

That wily fox Matt Cutts once again set about the clucking SEOs this month with a Twitter announcement that Google were making changes to their algorithm focusing on page layout.

Google uses aspects of page layout as one of the signals that determines SERP ranking. They are especially concerned that, all else being equal, they don’t rank pages highly when the ‘above the fold’ portion of the page does not contain useful content for visitors. What that generally means is that they would rather web site owners didn’t fill the part of the page that first appears to visitors with adverts and bury the content further down the page.

Unfortunately, Google, with their usual lack of clarity, has failed to stipulate exactly what constitutes good content ‘above the fold’, but the common sense approach is usually best. Google tend to attempt to put themselves in the place of their users, and ask what is that user likely to find most useful. They may get that wrong fairly frequently, but absent any better data, this is probably the best approach for website owners too. Continue reading →

(Photo credit: planeta)

Duplicate content is a problem for search engines, and that makes it a problem for SEOs. Google  strongly dislikes including the same content more than once in search results, and if content exists in more than one place on a site, search engine algorithms have trouble determining which version they should include or exclude. They also have difficulty knowing where to assign link juice and authority. This confusion can lead to sites experiencing a loss of traffic and reduced SERP rankings. The “rel=canonical” element is intended to help search engines out by letting them know which page is “the page” when it comes to particular content.

Duplicate content from the perspective of search engine crawlers can be created in a couple of different ways. Sites can actually have the same content on two different pages of their site or they can have links that differ yet point to the same content — these both look the same to the crawlers. Continue reading →

(Photo credit: Wikipedia)

Anyone who hasn’t been living under a rock in recent days will be aware that GoDaddy recently suffered an outage that left millions of its client’s websites inaccessible. While this sort of downtime will obviously affect a site’s traffic and therefore revenue, many people are asking whether it will have an effect on their SEO.

The short answer to that question is no, probably not. Google is generally not happy when they coming knocking and find that a site in their index has apparently disappeared, but they are also aware that problems occur, and so long as they don’t occur regularly then a site’s SERP ranking is unlikely to be degraded. Reliability is important, and a site that is regularly down will take a ranking hit, but a short period of unavailability is an anomaly, not a trend that is useful as a signal.

With that specific case dealt with, it might be useful to consider the more general case of downtime and how it should be handled. It’s occasionally necessary to take a site down for various reasons: making large scale changes to the software or server, for example. How are we to handle this?

Continue reading →