Predicting Search Engine Algorithm Changes
by John Metzler
Predicting Search Engine Algorithm Changes
With moderate search engine optimization knowledge, some common sense, and a resourceful and imaginative mind, one can be able to keep his or her web site in good standing with search engines even through the most significant algorithm changes. The recent Google update of October/November 2005, dubbed “Jaggerâ€, is what inspired me to write this, as I saw some web sites that previously ranked in the top 20 results for extremely competitive keywords suddenly drop down to the 70th page. Yes, the ebb and flow of search engine rankings is nothing to write home about, but when a web site doesn’t regain many ranking spots after such a drop it can tell us that the SEO done on the site may have had some long-term flaws. In this case, the SEO team has not done a good job predicting the direction a search engine would take with its algorithm.
Impossible to predict, you say? Not quite. The ideas behind Google’s algorithm come from the minds of fellow humans, not supercomputers. I’m not suggesting that it’s easy to “crack the code†so to speak because the actual math behind it is extremely complicated. However, it is possible to understand the general direction that a search engine algorithm will take by keeping in mind that any component of SEO which is possible to manipulate to an abnormal extent will eventually be weighted less and finally rendered obsolete.
One of the first such areas of a web site that started to get abused by webmasters trying to raise their rankings was the keywords meta tag. The tag allows a webmaster to list the web site’s most important keywords so the search engine knows when to display that site as a result for a matching search. It was only a matter of time until people started stuffing the tag with irrelevant words that were searched for more frequently than relevant words in an attempt to fool the algorithm. And they did fool it, but not for long. The keywords meta tag was identified as an area that was too susceptible to misuse and was subsequently de-valued to the point where the Google algorithm today doesn’t even recognize it when scanning a web page.
Another early tactic which is all but obsolete is repeating keywords at the bottom of a web page and hiding them by changing the color of the text to match the background color. Search engines noticed that this text was not relevant to the visitor and red- flagged sites that employed this method of SEO.
This information is quite basic, but the idea behind the aforementioned algorithm shifts several years ago is still relevant today. With the Jagger update in full swing, people in the SEO world are taking notice that reciprocal links may very well be going the way of the keywords meta tag. (I.e. extinct) Webmasters across the world have long been obsessed with link exchanges and many profitable web sites have existed offering services that help webmasters swap links with ease. But with a little foresight, one could see that link trading had its days numbered, as web sites would obtain thousands of incoming links from webmasters who may have never even viewed the web site they were trading with. In other words, a web site’s popularity was manipulated by excessively and unnaturally using an SEO method.
So with keyword meta tags, keyword stuffing within content, and now link exchanges simply a part of SEO history, what will be targeted in the future? Well, let’s start with what search engines currently look at when ranking a web site and go from there:
On-page Textual Content.
In the future, look for search engines to utilize ontological analysis of text. In other words, not only your main keywords will play a factor in your rankings, but also words that relate to them. For example, someone trying to sell NFL jerseys online would naturally mention the names of teams and star players. In the past, algorithms might have skipped over those names, deemed them irrelevant to a search for “NFL jerseys.†But in the future, search engines will reward those web sites with a higher ranking than those that excessively repeat just “NFL jerseys.†With ontological analysis, web sites that speak of not only the main keywords but other relevant words can expect higher rankings.
The conclusion: Write your web site content for your visitors, not search engines. The more naturally written sites can expect to see the better results in the future.
Offering Large Amounts of Content.
This can frequently take the form of dynamic pages. Even now, search engines can have a difficult time with dynamic content on web sites. These page usually have lengthy URLs consisting of numbers and characters such as &, =, and ? The common problem is that the content changes so frequently on these dynamic pages and the page becomes “old†in the search engine’s database, thus leaving the search users seeing results that contain old information. Since many dynamic pages are created by web sites displaying hundreds or thousands of products they sell, and the number of people selling items on the Internet will obviously increase in the coming years, you can expect that search engines will improve their technology and do a better job indexing dynamic content in the future.
The conclusion: Put yourself ahead of the game if you are selling products online and invest in database and shopping cart software that is SEO-friendly.
Incoming Links.
Once thought to be a very difficult thing to manipulate, incoming links to one’s web site have been abused by crafty SEOs and webmasters the world over. It is finally at a point where Google is doing a revamp of what constitutes a “vote from [one site to another]†as they explain it in their webmaster resources section. Link exchanges are worth significantly less now than ever to the point where the only real value in obtaining them is to make sure a new web site gets crawled by search engine spiders.
Over the years, many web sites reached top spot for competitive keywords by flexing their financial muscle and buying thousands of text links pointing to their site with keywords in the anchor text. Usually these links would appear like advertisements along sidebars or navigation areas of web sites. Essentially this was an indirect way of paying for high Google rankings, something which Google is no doubt trying to combat with each passing algorithm update. One idea of thought is that different areas of a web page from a visual point of view will be weighted differently. For example, if a web site adds a link to your site within the middle of their page text, that link should count for more than one at the bottom of the site near the copyright information.
This brings up the value of content distribution. By writing articles, giving away free resources, or offering something else of value to people, you can create a significant amount of content on other web sites that will include a link back to your own.
The conclusion: It all starts with useful content. If you are providing your web site visitors with useful information, chances are many other sites will want to do the same. SEO doesn’t start with trying to cheat the algorithm; it starts with an understanding of what search engines look for in a quality web site.
An expert at organic SEO, John Metzler has held executive positions in the search engine marketing industry since 2001. He is the President of FreshPromo, a Canadian-based SEO firm, and services American clients
=================================================================================
Google Ranking Tips
Google is by far the most popular search engine available today for both ordinary surfers and webmasters alike. Surfers like it because of the highly relevant results it gives and the speed at which it delivers them. This is due to its complex text matching algorithm and of course the Pagerankâ„¢ system that this engine uses. More on the Pagerankâ„¢ system later. Google is popular with webmasters and Internet marketing companies due to the highly workable ranking system it uses.Â
Unlike other engines where information about how the results are obtained are sketchy at best, Google actually publishes information on its site about the results it produces. Hence webmasters have things they can do to produce higher rankings.
What also makes Google popular with webmasters is the speed at which they will spider and list your site. If you’re not listed in Google and submit your URL you’re usually indexed within two weeks. If however your site is already listed in the index Google should reindex once every month, but more frequently if you’ve a high Pagerankâ„¢.
This indexing and reindexing time is much quicker than most other search engines. This allows webmasters to edit their pages properties such as title, first few lines of text, headings, keyword distribution and of course the number of incoming links to their site. They can then discover quickly if the changes they made were successful or not.
It’s because of this popularity that you need to know the workings of the Google search engine. Without knowledge of it you’ll be ranked lower than all other sites that are only slightly familiar with the Google algorithm and hence could lose lots of potential customers.
Google ranking algorithm
Let’s now continue onto the main part of this Google rankings report by indulging ourselves in the Google ranking algorithm. Well there are two main parts to the algorithm Google uses, the first is its text matching system whereby Google tries to find pages relevant to what the searcher has entered in the search box. The second and equally important part of the algorithm is of course the Google patented Pagerankâ„¢ system.
I’ll first go through how to make your pages relevant by discussing the text matching part of the algorithm.
Google gives a lot of “weight” to the title tag when searching for keywords. It is therefore vital to make sure your most important keywords or keyphrases appear within this tag. It seems to work best if you’ve other words in your title tag too after your keywords, but try to remain under 35-40 characters.
I imagine many of you know this already but Google does not use meta tags such as the keywords meta tag or the description meta tag. This is because the text within these tags can’t be seen by visitors to a website. Therefore Google feels these tags will be abused by webmasters placing lots of unrelated words in them in order to get more visitors.
This lack of support for meta tags means that Google creates your description from the first few lines of text on your page. This in turn means that you’ve to have your keywords and phrases right at the top of your webpage, if Google finds them your page becomes more relevant, if however it doesn’t find them the rest of your page has to work harder to become relevant. To see an example of what I mean scroll back to the top of this page and you’ll notice keyword rich wording similar to:
Google submitting tips, ranking high at google.com, Google ranking tips, pagerank algorithm, Google algorithm guide.
The above text includes keywords and keyphrases related to the theme of this page. Now when people search for any of those keywords or keyphrases this page is much more likely to be near the top of the results than a page that doesn’t imply this technique.
Google considers keyword density in the body of a page for determining relevancy too, so make sure your keywords and phrases appear a couple of times throughout the whole page. Don’t go overboard though, a density of 6-10% seems to work best.Â
Google has recently been noticed to give a substantial amount of “weight” to words appearing between the various header tags. These are tags designed to help you split up sections of your page, so this approach by Google seems to make sense. The header tags go from
the smallest to
the biggest, the bigger the heading tag the more relevent your page will become for the words within it. It is for this reason that you should always try to have your most important words within these tags as often as possible throughout your page.
Other advice about making your page relevant would be to make as many keywords appear within bold tags as you can. In the past Google has been known to index text in alt image tags, whether they still do or not I don’t know but it couldn’t hurt to include keywords in these tags anyway.
One final tip on page relevancy is the point on having your keywords and phrases in links which point to your site.
Â
It’s a good idea to have the linking text contain your keywords as Google even says itself in its description of its Pagerank technology that it analyzes pages that links come from.
How much keyword laden links matter is anyone’s guess. I have however noticed a lot of sites which give the HTML code to visitors who want to exchange links do include keywords in the actual linking area. You should do this too on your links page, say something like “if you want to link to this site, please use the following code”. The code would of course have your most important words in the actual link text and your less important words in the accompanying description of your site.
Google Pagerank
In the above section of the article you’ve learned what areas Google uses and looks at when looking for a relevant site, but what method does Google use to determine which site is better, the answer is the Pagerank system.
Pagerank is as the name suggests a ranking system of pages. It works on the basis that if a website ABC.COM has been linked from a website XYZ.COM, abc.com must have some good content and therefore Google will count the link from XYZ.COM as a vote for ABC.COM. You can check your Pagerank on Google by downloading the Google toolbar from http://toolbar.google.com/
The Pagerankâ„¢ scale goes from 1 to 10 on the Google toolbar and from 1 to 7 beside listings in the Google directory. A less important site is of course a site with a PR of 1 and a very very important site is a site with a PR of 7 or 10, in the directory or toolbar respectively.
The more links or votes a site has the more important it must be and therefore the higher it will rank for search words which it is relevant to, right?, WRONG!.
Google does not simply count the number of incoming links a page has, if that was the case every webmaster from Iceland to Vietnam would try and exchange links to every Tom, Dick and Harry website that would let them. In Googles own words:
“Google looks at more than the sheer volume of votes, or links a page receives; it also analyzes the page that casts the vote. Votes cast by pages that are themselves “important” weigh more heavily and help to make other pages “important.”
Hopefully your beginning to get the idea. The idea is to have your page linked to by as many high quality and high pageranked sites as possible. Right? RIGHT and WRONG.
WRONG BECAUSE, you see the Google Pagerank system also takes into account the number of links the page that has linked to you has. The reasoning for this is that a page X has a certain amount of voting PR, if your site Y is the only link from that page X, then Google feels confident that page X thinks your page Y is the best link it has and will give you more PR. If however page X has 50 links, page X could think your only the 50th best link. Hence the more links a page has the less of a PR boost your site will get.
RIGHT BECAUSE, linking to a site with a 6+ PR will provide a significant boost to your PR in most cases, but in cases where the site also links with 100 other sites the boost will be almost zero. Likewise if a site has a PR of just 2 but you and only one other site are linked from it, then the PR boost would be more than the site with 100 links and a PR of 6.
Google Pagerank formula
It’s beginning to come complex isn’t it, just wait till you see this formula. It looks scary for non math’s people.
First let me explain what the damping factor is. The damping factor is the amount of your PR which you can actually pass on when you vote / link to another site. The damping factor is widely known to be .85, this is a little less then the linking pages own PR.
PR(A) = (1-d) + d(PR(t1)/C(t1) + … + PR(tn)/C(tn))
In layman’s terms PR(A) is the Pagerank boost your page A will get after being linked from someone else’s site (t1). PR(t1) is the pagerank of the page which links to you and C(t1) is the amount of total links that (t1) has. It is important to know that a pages voting power is only .85 of that pages actual PR and this voting power gets spread out evenly between all sites it links to.
Imagine http://www.akamarketing.com was linked by XYZ.COM’s link page which had a PR of 4 and 9 other links, here’s how the formula should look like:
PR(AKA) = (1-.85) + .85*(4/10)
PR(AKA) = .15 + .85*(.4)
PR(AKA) = .15 + .34
PR(AKA) = .49
To sum up my site would get an injection of .49 PR after being linked from a page with a PR of four and 9 other links.
Let’s say I was linked from a site with a PR of 8, double the previous example’s amount, which had 15 other links, a total of 16 outbound links, my boost would be:
PR(AKA) = (1-.85) +.85*(8/16)
PR(AKA) = .15 + .85(.5)
PR(AKA) = .15 + .425
PR(AKA) = .575
The above two worked examples show that not only is the PR of the linking page important but what is also important is how many other sites are also linked to from that page.
I think we’ve had enough maths stuff for now, just remember that the name of the game is to get as many links from pages with high PR and few other links. The more of these links you get the more your PR will grow and the more your rankings will improve for your relevant keywords.
Google Pagerankâ„¢ and web directories
The best thing you can do for your PR seems at the moment seems to be getting listed in Dmoz.org. Dmoz.org is home to the Open Directory Project which is a human compiled directory of websites.
Pagerankâ„¢ is widely known to be biased towards big name directories such as Dmoz.org, Yahoo and Looksmart.
This is true, especially in the case of Dmoz.org. These ODP links are treated like gold by the Pagerankâ„¢ system. It doesn’t even matter what the individual PR of the category page is. I have seen sites gain a large PR boost on the toolbar as soon as Google updates its directory with the latest one from Dmoz.org. This is because Google uses its own version of the ODP for the Google web directory.
Don’t believe ODP links are very important to Pagerankâ„¢?
Don’t believe a listing in the ODP will boost your ranking?
Well they are and it will. Perform a search for almost anything on Google and you’ll discover that 75-80% of the top 10 results are also indexed in the Google directory and therefore also listed on Dmoz.org. The fact of the matter seems to me to be if your not listed with the ODP, you shouldn’t expect much traffic from Google.
Getting a listing is not difficult, it does sometimes take time but it’s not difficult. Just make sure your site has good content and follow the guidelines for adding a URL. Try to get your index page listed at least. I say at least because although ODP claims only to list your index page, there are plenty of sites with 5 – 10 pages listed.
Therefore if your site has very distinctive sections you can submit each section, beware though that this must be done slowly, otherwise you may be banned from the directory altogether for spamming. Once Google updates its directory these listings could do wonders for your sites Pagerankâ„¢. My article entitled Open Directory Project guide is a complete guide to getting into the ODP, I highly recommend you read it.
Regarding Yahoo and Looksmart, Pagerankâ„¢ will usually allocate a more than normal amount of PR boost for any sites listed. Tips on getting listed in Yahoo can be read in Yahoo submitting tips.
If you’re a non-commercial site or have a site that’s almost completely non-commercial you can get into the Looksmart directory through http://www.zeal.com . I really love this site, just like Google obtains its directory results from ODP, Looksmart obtains its non-commercial listings from the Zeal web directory. Without Zeal I would have to fork out hard earned cash and all my site does is provide free information in the form of articles and tutorials.
To continue, I submitted AKA Marketing.com on a Tuesday and was listed in Zeal by Thursday morning. On Monday I checked my logs and found lots of referrals from Looksmart, I was in Looksmart already. I looked at my logs later only to find MSN had updated its database from the Looksmart database and was sending me loads of visitors because of the good listing I got. My site was listed in Zeal, Looksmart and MSN within six days. Needless to say I highly recommend you get over to Zeal.com and submit your site.
Before you can submit a site however you must pass a member quiz, the quiz is a fairly simple straightforward one, you might however learn a thing or two while doing it.
If you happen to be a webmaster that has a listing in all three of Dmoz.org, Yahoo and Looksmart then I’m guessing your site has good to very good PR and rankings.
Google ranking tips – Conclusion
In closing I would like to recap on some of the main points you have just read. Be sure to have your main keywords and keyphrases in your title tag and well spread throughout your page, use header tags wherever possible. Get as many links from as many high PR low number of outbound links pages as you can. Get listed with Dmoz.org, Yahoo and Looksmart.
Well that’s it. I hope you have enjoyed reading this Google Ranking tips article as much as I have enjoyed researching and writing it. I also hope that you can put this advice into use and help get yourself some real traffic.
Article by David Callan. David is an Internet marketing professional and webmaster of AKA Marketing.com webmaster forums. Visit his webmaster forums for the latest discussions on search engines, website authoring and Internet marketing related issues and topics.Â
Google Ranking Tips Article Source :
http://www.onlineauctiontrader.com
If you are asked if you have experience with one particular content management platform, run – unless you are certain their CMS platform allows for manual editing of Title tags, metadata, and embedded links in body text – and that system allows for CSS attributes that can be altered to support SEO concerns. Few companies will abandon legacy CMS systems because you tell them it won’t work for SEO or that it will require complex workarounds to hack the proprietary in-house CMS database.