LinkBack URL
About LinkBacks
Google "middle column first css" and you'll find some beauties!
Over the years I've been lead to believe that the first sentence of the body text is more important in SEO than other parts of the body text. Having recently remade one of our sites, I've noticed that the search engine is seeing the left hand panel as the first sentence in the body text. Is there any way to make it look at the middle cell which is where the main content is first? I don't want it to completely ignore the left hand panel, just not pick it up as the main content. I'm sure I've read this is a SEO book before but I can't remember which one for the life of me.
Cheers,
Mike
Google "middle column first css" and you'll find some beauties!
To clarify. The first 1024 characters of indexed content are the most important.
This starts with the TITLE and the other META content in the head. If there are any characters left, the first indexible content in the BODY is taken into account.
Characters are taken in the same order as they appear in the code. If you want other content taken first, you have to put that code before the content you do not want to promote.
Most layouts are created by people who have no idea about search engine algorithms and dump what you want search engines to index down in the bottom quarter of the code without taking into account that the database is of finite size and once it is full the rest of the content is ignored. Recent database upgrades indicate that this limit is now over the old 'about 40 - 80 Mb, keep less than 25 Mb to be safe' limit.
PM me with your domain url and other info (html or script?) and I will have a quick look to see how easily your layout can be tweaked to but the main content higher up in the code sequence.
Where did you get that number from moredial? Do you have a related reference I could read?Originally Posted by moredial
I was going to ask the same thing...
The SEs read the whole page (assuming the crawler stays that long) and indexes the words on that page with frequency and importance - derived by <h1> etc tags, bold, italics, bullet lists and so on.
Yes, but keywords towards the top of the page are marked as having more importance than text lower down the page (and have seen this from my own experiments).
If you have a huge left menu with a load of links, it pushes the content right down the page.
This is one of the problems with table based layouts, and CSS layouts do tend to be much better, as you can put your content first, then the left and right columns, and still have it display in the normal way.
If the page in general isn't that great, then overall it makes little difference. However, if you have the page optimized to the hilt, putting the content first can be ramp your page up the results by more notches.
I can't see a reason why a sophisticated search engine would place more importance on phrases that appear near the top of the page. Writing styles differ from person to person and some people don't place important text near the top. Newspaper bulletin style writing places important information near the top but creative, fictional writing doesn't. In giving more importance to phrases that appear near the top, the latter writing style would suffer. I don't think any competent programmer would code something as clumsy as that.
But, I'm open to changing my mind if I see some good proof
Tell me, how were you able to isolate the influencing variables in your experiment. A difficult task given that you cannot even identify them all the variables that a search engine uses.
Basically, by swapping the layout of the page round in terms of code, putting the center column first, and changing nothing else. The page was then left and watched, and I saw jumps in the rankings.Tell me, how were you able to isolate the influencing variables in your experiment. A difficult task given that you cannot even identify them all the variables that a search engine uses.
It makes sense really if you think about it. Google is trying to make their spider read the page in a more "Humanlike" way.
If you take a look at the page source, then having the minimum of HTML, then solid blocks of your content text, and then left menu links etc gives you the relevant info straight away. The most important part of the page the content is easily accessible at the top and can be read straight away. And before anyone says that humans wouldn't look at the page source, this would be exactly the way it would be seen by a blind person using a screen reader. (Which is how I tend to treat the spiders)
I know that shifting your DIVs around is a good idea to prevent menus / header navigation etc being the 'snippet' in the SERPs, but for your average page (i.e. not one sooo long that the crawler never gets to the end) would it really make a difference?
Devils advocate : So the 'conclulsion' to any report / story / article - the summing up, the authors concise conclusion, is given the least weight?
I don't think that crosses over onto the web. There are billions of articles on the web, and people don't want to have to read the whole of each article, to get to the last paragraph, "the conclusion" to find the info they want. If you had to do that, you`d have to read a hell of a lot of text to find the info you're after.So the 'conclulsion' to any report / story / article - the summing up, the authors concise conclusion, is given the least weight?
One the net, people want the information they are looking for fast! They tend to read the first paragraph, and if that doesn't include the info their looking for, or a very strong indication that if they read on they`ll find it, they`ll leave and read something else!
How many times have you gone to a web page, and the first couple of paragraphs haven given you the info you're after, so you`ve left?
Because of the vast amount of info on the web, the importance is finding what you want quickly.
This is one of the reasons why the text in the first <h1> tag on the page has such a high weighting.
In the end you do what you think is right. I`m convinced that my info is correct, it puts 90% of my sites in the top 3 results in google. If you don't agree with this, you don't have to do it
Heh, i was playing Devils Advocate
I was more interested in the 1024 characters bit from the other poster...
Yes, personally I haven't heard of the 1024 character limit before, I don't see why it would apply as 1K is very small now a days.
I just try and put my content as near to the top as possible, styled where possible with CSS from an external stylesheet, so if you look at the page in source view, its still very readable. If you do that, the search engines seem to love it (as do screen readers). All the rest such as left menu links etc can come underneath tucked out of the way.
Back againHeh, i was playing Devils Advocate
I was more interested in the 1024 characters bit from the other poster...
I am acquainted with a SE programmer who has been around since the days when students played with scripting search programs. It was an 'open source' program at that time and 1024 is the maximum content that could be contained within a data block at that time in computer development. This helped to set the structure of the databases created when content was cached.
You have to think back to what was available in html2 to understand how the database was structured and what early web content looked like. Markup was all you had to lay out the page and give text size so markup indicated content importance just like a newspaper article.
There is too much content on the page to make a quick search on the whole content so a summary field was added to the cache - this is where the 1024 characters comes from. Into this field was added the content of the title tag, meta description, meta keywords and as much body content as possible. In the early days, sites did not have layouts with menus so it was highly probable that the next tag after the body tag was the h1 tag and under that was a brief summary of what the article was about. Which means that this was a very logical way of gleaning page relevance very quickly and within a few characters.
You may recall that a few years back everyone was putting as many 'popular' words in the meta keywords as they could in the hope that people searching for popular searches would find their site and be interested in staying for more than a few seconds.
Over time, algorithms change. The general consensus now varies between meta data being so ignored by search engines as not to be needed (an incorrect assumption as these tags have purposes other than for search engine indexing), stuffed with as many terms as the author can think of or using this area to provide a summary of the page content (what the markup was designed for). Basically there are 2 types of search engine - the meta search engines (MSN and some directories) and the rest. Meta directories only request the head (title and meta) info from the server and use that content to present relevant results. There is every indication that Google and Yahoo still index the meta content to the extent that Google will use the meta content when the page is 'empty' although generally the meta is ignored unless the content is also found in the visible page content - evidence that all that keyword stuffing and irrelevant results got search engine users looking for search engines that offered better results.
1024 is not an exact number as within that count are markers to indicate if the content came from the title tag, meta tags or body tags. You can see in the SERPs how quicky this one field is used to generate the results - title tag on the first line and meta description to give potential visitors a summary of the page, or an extract from the page content where the search terms appear if the meta description does not contain the search term.
Don't start thinking like an ignorant guru, this is only one field out of the database that holds the cache for the whole page content. Other fields contain info on content of inbound link text - you have all seen the cache where you either have the search terms highlighted, each in their own colour, or the phrase "these words only appear in links pointing to this page".
Over time there have been changes to the database structure but you know for yourself how difficult it is to restructure a mature database. MSN/Live experimented with various structures and weightings and now appears stable. Yahoo has had some major changes over the last few years - sometimes it looked like 2 or more different search engines with sites disappearing and then reappearing back at their old positions a few weeks or months later. Google has also been adding in some new features and it is noticeable that some very old ghosts of pages have finally disappeared from the results over the last year, following on from the very deep crawl Google has done over this time. The latest Google change has been to drop the 'supplemental results' tag. Don't be fooled by this - pages are still sitting in the supplemental results, and the only change is the removal of the tag from the SERPs.
The simple formula to SEOing a page is the following:
title - what it says on the label
keywords - prime and secondary page topics
description - a short summary of the page content
h1 - visual confirmation to the visitor that the title seen in the SERPs is what the page is all about.
text under h1 - a longer description and summary of the page content which should expand on the content of the meta description.
Have those elements correct and you will have visitors who stay on you page for longer than the average of 10 seconds.
By putting the h1 and summary text at the top of the body tag content, all you are doing is making it easier for your page to rank as highly relevant for its content.
Put some other content into this area and all it is doing is diluting the page relevance to its title tag - as seen from the SE's point of view.
The sad thing is that businesses spend thousands on getting someone who has no knowledge of the importance of the html markup to design a new site which looks wonderful to the human eye, but completely makes the content irrelevant from the search engines' indexing algorithms. Instead of needing only a few inbound links there is now a whole sister industry of link exchanges to boost the weight of the database field which contains the info on inbound links. And the poor cousin industries who write such spammy content that no one can bear to read more than the first few sentences - if they even get so far.
Better stop now, I feel like I have written a book this evening.
Quick addition
If you design your page so that a screen reader can make sense of it when reading from the top to the bottom of the code, then you also know that a search engine will be able to make sense of it.
To see how well you have done, use Opera without javascript,images, styles and layout and you have the screen reader / search engine view of your page.
Now bearing in mind I have a lot to learn, I am using this product called "Good Keywords" which is freeware that I downloaded from Download Good Keywords - the main reason I use this is that it has a section "Web Page Explorer" that lets me view a web page as "plain text" which I have assumed (I hope correctly) is what you are talking about above. Be really chuffed if someone could take a look at this product and re assure me that this is the case....Quick addition
If you design your page so that a screen reader can make sense of it when reading from the top to the bottom of the code, then you also know that a search engine will be able to make sense of it.
To see how well you have done, use Opera without javascript,images, styles and layout and you have the screen reader / search engine view of your page.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
Bookmarks