For the next few weeks, my blog posts will primarily consist of re-authoring and re-building the Beginner's Guide to Search Engine Optimization, section by section. You can read more about this project here.
How Search Marketers Study & Learn How to Succeed in the Engines
The complicated algorithms of search engines may appear at first glance to be impenetrable, and the engines themselves provide little insight into how to achieve better results or garner more traffic. What little information on optimization and best practices that the engines themselves do provide is listed below:
- Google - How Can I Improve My Site's Ranking?
- Yahoo! - How do I get listed in Yahoo! Search "Web Results," and how to increase the ranking of my site?
- Microsoft/Live - Guidelines for Successful Indexing
- Ask - Ask.com for Webmasters
The guidelines above aren't useless - some valuable tidbits are certainly present, including this from Yahoo!:
Many factors influence whether a particular web site appears in Web Search results and where it falls in the ranking. These factors can include:
- The number of other sites linking to it
- The content on the pages
- The updates we make to our database
- The testing of new product versions
- The discovery of additional sites
- Changes to the search algorithm—and other factors
Web Search lists results ranked by relevance and offers a combination of sites from the entire Web obtained by Yahoo! Search and from the Yahoo! Directory.
And this from Microsoft's Live:
In the visible page text, include words users might choose as search query terms to find the information on your site. Limit all pages to a reasonable size. We recommend one topic per page. An HTML page with no pictures should be under 150 KB. Make sure that each page is accessible by at least one static text link. Don't put the text that you want indexed inside images. For example, if you want your company name or address to be indexed, make sure it is not displayed inside a company logo.
These tiny snippets are, thankfully, just the tip of the iceberg. Over the 12+ years that web search has existed online, search marketers have found methodologies to extract information about how the search engines rank pages and use that data to help their sites and their clients achieve better positioning. Surprisingly, the engines do support many of these efforts, though the public visibility is frequently low. Conferences on search marketing, such as the Search Marketing Expo, WebMasterWorld & Search Engine Strategies series attract engineers and representatives from all of the major engines. Search representatives also assist webmasters by ocassionally participating online in blogs, forums & groups (these are listed below in Appendix B).
However, there is perhaps no greater tool available to webmasters researching the activities of the engines than the freedom to use the search engines to perform experiments, test theories and form opinions. It is through this iterative, sometimes painstaking process, that a considerable amount of knowledge about the functions of the engines has been gleaned. A common process for testing might look something like this:
- Register a new website with nonsense keywords (e.g. ishkabibbell.com)
- Create multiple pages on that website, all targeting a similarly ludicrous term (e.g. yoogewgally)
- Test the use of different placement of text, formatting, use of keywords, link structures, etc. by making the pages as uniform as possible with only a singular difference
- Point links at the domain from indexed, well-spidered pages on other domains
- Record the search engines' activities and the rankings of the pages
- Make small alterations to the identically targeting pages to determine what factors might push a result up or down against its peers
- Record any results that appear to be effective and re-test on other domains or with other terms - if several tests consistently return the same results, chances are you've discovered a pattern that is used by the search engines.
Here's an example of a test that we at SEOmoz have performed in the past:
In this test, we started with the hypothesis that a link higher up in a page's code would carry more weight than a page lower down in the code. We tested this by creating a nonsense domain linking out to three pages, all carrying the same nonsense word exactly once. After the engines spidered the pages, we found that the page linked to from the highest link on the home page ranked first and continued our iterations of testing.
We had some concerns that the text inside the link might be the source of the rankings, and so changed the link text on the homepage to nonsense characters. Still, the pages ranked in order of highest to lowest rank. Over several more iterations (changing up the linked-to pages, changing the terms, mixing up term usage, etc.) and repeat testing, we found the pattern emerged again and again, and decided to call it "confirmed." It appears that the three major engines (Ask never indexed the domain) all place some amount of higher weight on links higher in code than lower down. Obviously, this is a time consuming and tedious process, but it can help search marketers to understand the basic patterns of how search engines determine rankings.
This process is certainly not alone in helping to educate search marketers. Competitive intelligence about signals the engines might use and how they might order results is also available through patent applications made by the major engines to the United States Patent Office. Perhaps the most famous among these is the system that spawned Google's genesis in the Stanford dormitories during the late 1990's - PageRank - documented as Patent #6285999 - Method for node ranking in a linked database. The original paper on the subject - Anatomy of a Large-Scale Hypertextual Web Search Engine - has also been the subject of considerable study and edification. To those whose comfort level with complex mathematics falls short, never fear. Although the actual equations can be academically interesting, complete understanding evades many of the most talented and successful search marketers - remedial calculus isn't required to practice search engine optimization.
Through methods like patent analysis, experiments, and live testing and tweaking, search marketers as a community have come to understand many of the basic operations of search engines and the critical components of creating websites and pages that garner high rankings and significant traffic. The rest of this guide is devoted to explaining these practices clearly and concisely.
... ooph. This is tough stuff - as always, please provide your input and edits. We'll be making changes to all of this content before it goes into the finished guide (and Mystery Guest even offered to edit - yay!)
Today's web roundup:
- Lobo links tells us how to use CSS image replacement properly
- Steve Rubel says Search is Broken - I think he means "not as advanced as a very, very advanced user like him might want it," but it's still a good point
- Dr. Peter Meyers has a terrific (though lengthy) piece on the Basics of Measuring Usability
- Gary Price exposes websites on the cutting edge of data mining - love it
- And more Gary - Send your children to public school and use the money to buy... I don't know... KrustyO's - private schools aren't worth squat
- I got stumped in Q+A - if anyone reads Arabic and can help, I'd really appreciate it. Oh wow, the brilliant Pierre Far is already on it!
I'm on David Brown (aka NEOSEO)'s podcast show tomorrow morning at 10am and it's already 2am... Man, I never get to sleep...
Rand - I would generalize the testing concept a little bit and I'd not recommend newbies to register websites with nonsense keywords. I know why you did that on the past and I am sure it works, but imagine what would happen to the search results if everybody started doing that.
I suggest the testing be modeled after the scientific method. See https://www.sciencebuddies.org/mentoring/project_scientific_method.shtml
For example:
· Ask a Question Does Google use the meta keywords tag?
· Do Background Research Check research papers, patents, etc.
· Construct a Hypothesis Google does not use the meta keyword tag
· Test Your Hypothesis by Doing an Experiment Include nonsense keywords in the meta keyword tag of the home page of your site and wait for Googlebot to crawl it and index it.
· Analyze Your Data and Draw a Conclusion Is my home page coming up for the nonsense keywords? If yes, then the hypothesis is incorrect. Otherwise is correct.
· Communicate Your Results Write a Youmoz entry to share your findings!
It is also important that the experiments be repeated multiple times to make sure the results are reproducible, and there is a direct relationship between cause and effect.
I like Hamlet's way of writing up the testing process, but overall I have to agree that this section would be better further along. After reading section IV (at least based on the proposed outline), a beginner will actually know what it is you're testing with the various search engines.
If they don't know the importance of anchor text, for instance, they'll have no idea why using random fake words helps you find results.
When I read this guide, I really was a beginner - I mean, I knew almost nothing. If that's who the guide is for, it might be a good idea to rearrange some of the info.
Lorisa - thanks, it's great to get your input as someone who's been there :)
Rand - It think lorisa is giving a really good piece of advice.
To be honest, I think that maybe a lot of this post, whilst informative is not really something I would expect to see in a "beginners" guide. Someone new to SEO or wanting to get a few tips on increasing the ranking of a page will see this and skip right on. Testing things for SEO benefit is more something for an intermediate + level.
I beg to differ,
the beginners guide is usually very less informative, something like this keeps the interest going and gives a small peek into detials.
It's kind of funny, actually, that we who read about SEO all the time see a title like, "Anatomy of a Large-Scale Hypertextual Web Search Engine" and think, "Oh, yeah, that was a great read. When is someone going to turn that into a musical?" To the average internet user, it's like a foreign language.
I found the "history of search engines" part of the SEOmoz conference to be very interesting. It explained how we got to where we are in a very logical, sequential way. By the time you reached the end, I understood exactly why the search engines use the criteria that they do to return relevant rankings.
Will that info be in this guide? That might be good to put near the beginning.
That is an excellent point.
I agree, once you understand how something works and operates (generally speaking). You can then begin to better understand the methods used to achieve a desired goal.
I have to disagree with those who think the testing example is too advanced for a newbie. I read the old Beginner's Guide a few months ago, and I thought it could have been a little more informative. I think by the time a SEO newbie comes to SEOMoz they already understand what SEO and SEM is, they just want to learn how to do it.
I think that teaching newbies to test things for themselves is a good way to get them thinking for themselves. I believe that it will also teach newbies to rely on experience and testing rather then relying on other people's blogs. In the long run a newbie that learns to test their own theories will be able to contribute more to the community. They will have fresh content and new ideas to share instead of just a summary of other people's blogs and white papers.
Don't get me wrong; I'm a big fan of testing and think Rand's example is great. It just seems to be very advanced for that early in a Beginner's Guide. It might be as simple as saving it for later in the document, after more core concepts have been covered.
I agree with Protocol96. This is much more informative than the traditional start-out or beginners guides.
Rand when your are finished giving this an update, will you update the downloadable versions?
I like the experimental approach to describing SEO tactics--and given it is such a dynamic medium, understanding the framework in clear language and not advanced mathematics is good for non-expert SEO guys like myself.
Well, first off, thank you kindly. That article is actually a combination and major re-edit of three blog posts, so it is admittedly a bit on the long side. Rand was kind enough not to mention that I badgered him into reading it :)
Now, on to the comments:
I do feel like I should say that I really got a lot out of the original version the first time and am looking forward to the full rewrite. Having re-dedicated myself to writing/blogging lately, I know how difficult putting together a piece like this can be.
Hey, I think it's great to have some testing stuff in a beginners guide.
"Not sure how something works? Then set up a test and see for yourself!"
By showing this example you're are putting a bit of power and control into the hands of your readers. If people think this is a bit strong then maybe pop a disclaimer at the beginning of that section along the lines of:
"It's ok if you don't get this right now, you can come back later when you have a bit more experience."
As a beginner/newbie I’m still finding it difficult to filter the guidance I need so my observations maybe incorrect, however, it appears to me that the value of this type of information will depend on the reader. If it is a beginner using the guide as a stepping stone to further develop their knowledge then content like this is invaluable, however, if the reader is a manager looking to gain a broad understanding for management purposes then they are likely to skim across it.
I think it might be easier to fit various elements together in the right order for beginners once most of the primary topics have been introduced. It's obviously quite difficult to write a book in a public blog, post-by-post, as all the editing that would usually be done at a later stage (when reviewing) is siezed upon immediately - but this is good use of the moz readership too.
I think that beginners prefer to have concrete examples - I know I do - so that the ''what' and why' is complemented with a definite 'how'.
btw, thanks for the link to the css image replacement page - although i knew the 'what', this is a perfect example of providing the 'how' as well...
Hi Rand, can you let us download the whole guide once you've finished with the lessons?
Thanks for sharing, You made a very nice job!!
The measuring usability document is great everyone should check it out.
good blog, and good re-write..
- I would say every SEO company should own a nonsensical domain name and use the testing to build a case study, of which they could paste first on their portfolio.
-.rb
I articulate has clarified many things to me, thanks
I'm in the "For" category on keeping the testing information because it also addresses the varied interests of beginners. Maybe someone reading this guide wants to get into the testy portion heavily, this will give them that inrotduction, while those that aren't interested will skim it and sum it up in their own minds (although Dr. Pete's suggestiong would be a great add at the end of the piece, sort of, "So from that specific example, here's the basic process"). And then reitterate the critical thinking aspect covered in this section.
Rand, I bet the gears got turning reading Rubel's article. It actually reminded me a little of your Popular Searches page, although not addressing the walled garden. Perhaps an entity something like Meebo, but for search, but as pointed out in the comments, you're really going to have to trust that all-in-one ap that allows you to search all your personal accounts.
Well, as one of those newbies, I'm finding this very helplful. I might not be ready to do the sorts of testing you mention--but it's useful to starts thinking in those terms.
Looking forward to the rest of the material.