Disclaimer: Readers are warned that this article contains language and ideas that may cause offense to those with traditional views: reader discretion is advised. That disclaimer is important since we will be rejecting some concepts you may have accepted without question. If you can suspend disbelief for a time, welcome.
Home Pages Suck
In a prior article, Home Pages Suck, it was noted that blog Home Pages tend to attract all the hyperlinks from other websites that may be referring to the given website. In this sense, the Home Page sucks the inlinks or back links preferentially rather than such links pointing to other web pages on the domain.
This arrangement may create somewhat difficult consequences for the Google PageRank concepts and algorithms as we will discuss here. Weigh carefully what is being said. Surprisingly, no one seems to have remarked on this PageRank problem prior to this article. If anyone knows of such discussions, perhaps they could mention them in the comments.
The Typical Blog Home Page
The field of discussion relates entirely to Home Pages that change over time. The websites produced by blogging software are of this nature. Originally a blog was a personal journal, where the author could from time to time write down an entry covering what was of interest at that time. More recent items were posted at the top and more ancient items were pushed off the bottom. A very typical arrangement was to have say 10 items on this Home Page.
Such blogs usually also produced a RSS news feed. This alerted search engines that the blog now had new content. Rather than this new content being found by spiders or robots crawling the Internet and finding it eventually, there was instant signaling of the change. The search engine could update its register on what was now featured on the Home Page at the URL for the blog.
The Typical New Blog Post Web Page
In parallel with that changing Home Page, the blogging software produces a web page for each item that is added to the blog. This new page may not be fully indexed immediately and is unlikely to get much respect for two reasons:
- the content is also found on the Home Page for a fairly long time, and
- there are only a few links directly to the blog post web page, unless it becomes an item that a large number of other authors are commenting on.
For any keyword phrases that may be found on the blog post web page, the blog Home Page is likely to rank much higher in search engine keyword searches at least in the early weeks.
It should be noted that people may develop their perceptions about blog post web pages from the very few that get the attention and thus their own back links. The vast majority of blog post web pages (probably 99.999%) are not like this. They tend to get lost, particularly early on.
Blogs Score Well With Google
Blogs score well with Google for two reasons. The first as we have mentioned is those news feed pings that the blog issues when new content is added. The other reason is that the blog Home Page is content rich. If each post averages 300 words, then the Home Page may contain 3,000 words. This gives many possible opportunities for matches with searchers who are now with increasing knowledge doing those 'long tail' searches. These feature a combination of keywords and in some cases can get complex. Only a blog Home Page may be able to display such complex phrases.
Blogs also have good interlinkage, both internally and externally. Many bloggers use tagging services such as delicious.com or Technorati that provide many more ways of finding the content with associated links. In parallel with this, the news feed aggregation services such as Feedburner are also providing links to content. Blogs provide prime real estate for the search engines to explore and catalogue.
The Format Of SERPs
SERPs or Search Engine Results Pages display a number of items (10/25/100) as requested by the searcher in response to a keyword query. The format of these items is typically an initial title, a short description or snippet and in some cases a URL. In Bing currently a list of single items is produced. In both Google and Yahoo currently, occasionally from a given domain a pair of items may be displayed in the list. The second item in such a case is indented below the first.
In all cases the search engines are displaying the results in order of relevance as they estimate this for the given keyword query. Presumably they limit the items shown from any domain to two in order to avoid a particular domain dominating a given SERP. In this way, searchers are given more variety in the list, which is more likely to cover the item they may be interested in.
For blogs you will often find that one of the items in many cases is the actual URL for the blog Home Page. This is not surprising given that this blog Home Page has a much higher PageRank than any individual blog post web page.
As an illustration, here are two sets of blog results for keyword searches for which they were irrelevant. The first is for some words from a Seth Godin blog post, 'So quilt. Spend hours every day'.
In this case, you will notice that the first item is describing the actual blog post in which the keywords were used. The second item is the Home Page for the blog.
The second illustration is for a blog post on the CopyBlogger blog. In this case the first item is for the blog and the second item is for the blog post. The keywords in the search were 'Junior Copywriter ad agency purgatory'.
As we will see, although we are all so used to accepting such results, when we look more carefully at them they are far from ideal.
The Google Dilemma
The problem for Google in presenting these results is that simple URL for the item describing the blog Home Page. In all probability that link will take you to a version of the blog Home Page that is different from what they saw when they last spidered that blog. Unfortunately most often the Home Page will outrank the single blog post web page, particularly if it is a relatively recent posting. This is why in both our examples the blog Home Page is listed.
The problem for Google is that in order to show an item for the blog Home Page, they can only use the Title for the blog as a whole and the description or snippet must be created from the content of that blog Home Page at the time it was spidered. If given that snippet, the searcher does click on the link for the blog Home Page, then they reach a long scrolling Home Page where the location of the desired keyword section may be difficult to spot.
In both cases, displaying the blog Home Page is a disservice to the searcher, when the individual blog web page has been identified as relevant to the keyword search. Showing only the individual blog web page would be much more helpful.
The Google Challenge
This disservice in the formats of the SERPs when dealing with blogs is possibly something that has not been addressed by Google. Google is very much a Product-directed company. It hires very talented people and attempts to do the best job it can according to their own perceptions. Since it is providing a free search service, it might be forgiven for assuming that customers (searchers) should be grateful for whatever service is provided. In fact the service provided by Google is very good on a number of dimensions. It is speedy, in general accurate and fairly complete in the results it provides. That explains why it is so widely used.
However if Google was more Searcher-centric (think customer-centric), it might more often look at these SERP results through the eyes of a typical searcher. As we have seen above, the SERPs include some almost irrelevant content that would be better removed. Why would Google not show only one result from any domain. That seems to be how Bing is now handling the situation. They list all items in the order of relevance and may show only one item or several items from the same domain.
We now move into the area of speculation. If Google showed only one result from any domain, then perhaps the PageRank approach makes this hazardous it does rank the Home Page of a blog very highly relative to other individual blog post web pages. if only a single result was shown for any blog, would it always be showing the Home Page, particularly where keywords were appropriate for recent blog items. If you think a moment about this, showing only blog Home Pages could be an even bigger disservice to searchers. Perhaps with a standard blog, showing two entries is the best way of handling this dilemma. Of course after some time, the individual blog post web pages do establish their own identity and ranking and will then be the only single item to appear in relevant keyword searches.
The Crux Of The Problem And A Solution
The crux of this problem is the changing nature of a blog Home Page. The content is being continually updated yet the Title of the blog and the Meta Description remain the same all the time. Although the blog Home Page may gain a high PageRank, this is attached to content that is in a state of continuing flux.
If a blog is updated on a relatively sedate cycle, such as every 7 to 15 days, then one solution is to have only a single post on the blog Home Page. This means that at least while it stays there, keyword searches will point at the same content on which the keyword relevance was based. This was the approach suggested in an article entitled, Giving Your WordPress Blog An Elegant Face.
Unfortunately this is only a very partial solution to the SERP format problem and will still produce from time to time SERP entries for the blog Home Page that are no longer relevant.
With blogs created in the standard way, it is not at all clear how Google might avoid these irrelevant SERP items. One sure way that bloggers can have their blog posts correctly identified in the SERPs is to follow the LMNHP approach.
Since it is the blog Home Pages that are causing the problems for Google and for searchers, just simply eliminate them. That avoids the duplicate content issues mentioned above and ensures that every blog post web page has only one reference in the SERPs. That seems Win/Win/Win all round.
Thanks Barry – I’m amazed that I’ve never even considered what this means for my blog.
I’d like to see how you’d suggest removing a blog homepage though – just replace the index with a redirect to the latest blog post, or am I overthinking this?
No that’s exactly how you do it, Andy. You then have a blog post web page that has the correct Title and meta description for that specific post, together with all the back links the domain (or perhaps it’s this specific post) is currently attracting.
.-= Barry Welford recently posted: Seth Godin Thinks Bigger Is Better =-.