Internet Entrepreneurs Blog

Here’s a relatively new search engine optimization myth: The simpler a page markup, the more likely the page will be indexed and the higher its rank. This myth seems to have grown out of the movement toward valid XHTML and the separation of web content from the actual page layout itself. Instead of laying out pages using complicated table structures, designers are encouraged to use separate style sheets for a cleaner code and (conceivably) better search engine placement.

We tested two page layouts, each with the same target keyword with similar keyword densities, page titles, etc. to see if there was a difference in 1. WHEN the page was indexed and 2. Which page ranked HIGHER once both pages were listed in the major search engines. The first page used a simple, valid XHTML layout while the content on the second page was nested in nasty, redundant, overly complex HTML tables.

On the first point, WHEN a page was indexed, the complex page came out on top, showing up in Google almost a week before our simple page. Over the following weeks there was some jockeying of position but ultimately the complex page has settled in the top spot over the simple page. So for our Google test, it appears the venerable search engine doesn’t have a problem with complex markup and in fact may prefer it on some level.

Yahoo! had the opposite take on our test and placed the simple page layout on top of the complex page in their search results. If you’ve been keeping up with our test results on this blog you’ll know this isn’t the first or even second time that Google and Yahoo! have disagreed on such a seemingly fundamental point. Based on our limited testing it appears Yahoo! tends to reward many of the “traditional” SEO tricks while Google often yields counter-conventional results.

We have one more search engine ranking test result to share with you, then we’re out. If you have an SEO myth you’d like us to test, leave a comment here and we’ll put one in the hopper for ya.

Here’s another Monday morning softball lob SEO question: How important is the ‘meta description’ tag in getting your pages indexed and ranked in major search engines? This time the answer rhymes with scary

Instead of leading with our test results, let me first give you some anecdotal evidence of the importance of good meta description tags.  In my rush to create new pages I (sometimes) forget to add meta description to my pages - good descriptions, anyway. Invariably the new pages are slow to get picked up and indexed (if at all) and only when I go back and add in UNIQUE meta descriptions are the pages indexed by Google and others.

After launching our blog on singletracks.com in early 2007 we found that our blog posts weren’t being listed as unique pages by the search engines - only the blog homepage was being picked up. It turns out the Word Press software we were using didn’t create unique meta descriptions for individual blog post pages so we fixed that by excerpting the first sentence or two and using that for the page description. Within a week or two we found our blog page listings had multiplied. Easy fix, big result.

Now, to our test results: The test page we created without a meta description is ranked in Google above our page with the description included. I know, I know, I said this would be a softball question - so why the crazy result? Honestly I’m not entirely sure what’s going on but I do have a theory. Our test pitted two pages, each with our target keyword written 6 times in the body of the page at similar keyword densities. One page had the keyword in the meta description two times (once per sentence) while the other page did not include a meta description tag at all. As you probably already know Google prefers to use the meta description as the document summary but when one doesn’t exist it attempts to create its own summary based on the actual page content. Since our page content contains the target keyword 6 times (3 times in the first paragraph alone) it ranks above the page whose meta description (or, document summary) contains the keyword just twice.

Strangely Yahoo! agrees and prefers the meta-descriptionless page as well. Our keyword page without the meta description was not listed by Yahoo! at all.

So what does this mean? I certainly don’t recommend skipping meta description tags on your pages based on my own experience but I do think this test proves an important point. Optimizing your pages for keywords doesn’t stop at the page content and internet marketers should carefully construct meta description tags to include target keywords as often as possible. Our test was performed in a vacuum of sorts but in the real world your competitors will be optimizing both their content and meta tags - don’t be left behind!

One thing search engine marketers might lead you to believe is that the big search engines want to make it difficult for you to get your content indexed. The common wisdom seems to be that online publishers need to do everything just right or the search engines will reject the publishers’ pages. Even worse, some marketers tell you if the search engines think your site is ‘gaming’ their listings, they’ll ban you for life (or something similarly terrible).

The thing is, I just don’t buy into this line of reasoning. The fact is, search engines like Google NEED publishers’ content to keep their search engine results relevant. Google probably won’t start rejecting content from CNN.com just because CNN decides to start stuffing their meta tags with bogus keywords - CNN still produces valuable content that answers online queries.

Even if your site isn’t as big or as well known as CNN, I’d argue that your content may be even more important to the search engines. Google and Yahoo! are useful tools for finding niche content that simply won’t be found by browsing CNN or Amazon. If search engines pick and choose which sites they list they are no longer complete resources - and hence, their value to users is diminished.

Still aren’t convinced that Google needs your content? Consider Google image search. Google has basically cached (that is, saved a copy to their own servers) photos from virtually every website on the internet and serves these images in response to user queries. Some companies aren’t too happy about the fact that Google is making money off their content (images and otherwise) and some publishers have even sued to block Google from accessing their content. Publishers blocking search engines? I thought it was the other way around

Of course search engines do need to be on the lookout for sites that provide little value in response to search queries (i.e. spam sites that simply provide paid links to other sites with zero original content). These ’spammers’ are getting more and more sophisticated but rest assured: If your site contains original and useful content, you WILL receive fair placement in the major search engines. It may take time but don’t worry - the search engines want and need your content!

This is another one of my favorite search engine ranking myths: if you ‘overstuff’ your page with keywords the search engines will penalize you in their rankings. For example: your site sells lamps. How many times should you use the word lamp on your homepage or, more importantly, how ‘dense’ should the word ‘lamp’ be in your copy? Once per sentence too much? How about every other word? How about EVERY word? If every word on your homepage were ‘lamp,’ would you be de-listed from Google and Yahoo?

Most SEO consultants paint a complicated picture of search engines as sophisticated algorithms that mere mortals cannot possibly understand - but I don’t buy it. My theory is that search engine algorithms are lazy and a little stupid so the keyword test is the perfect way to see just how ’smart’ the search engines really are. The conventional wisdom is that keyword density for a page should be somewhere between 1% and 7% to be considered ‘optimized.’ Not too much, not too little - just right for those finicky search engines.

So here’s the set up: 3 pages on the same domain, each with different keyword densities ranging from 6% (our ‘optimized’ page), 28% for the ‘overstuffed’ page, and 100% for the search engine gamers’ paradise platter. Now it’s time for you to guess - Which page is on top in Google? If you said the optimized page, you’re dead wrong - in fact, you picked the poorest performing page. The page with 100% keyword density is on top, next comes the 28% page, followed in last place (and a supplemental result no less) is the 6% page. Yahoo! appears to have a slightly more sophisticated palate, choosing the 28% page for the top spot but it still has love for the 100% page in second place. No Yahoo! love at all for the 6% ‘optimized’ page.

I know what you’re thinking: this just can’t be, it’s too simple. Remember that our test is conducted in a bit of a vacuum and that we’ve tried hard to isolate the keyword effect from any other factors (like pagerank, title tags, meta description, etc.). I personally don’t want to believe the results because it just doesn’t seem fair that a page packed to the brim with a single keyword can rank above pages with actual useful content. I’d love to hear from anyone with a counter example…

I’ve read some articles recently that claim that including meta keywords on your web page may be a futile effort since most search engines are ignoring them these days. The argument is that meta keywords are too easy to game and therefore search engines have stopped paying attention (though most SEOs will tell you to leave them in since they can’t hurt).

According to our tests, however, the meta keyword tag is just as important as ever. We created two test web pages: one with our target keyword placed in the meta keyword tag, the other without any meta keyword tags (though both pages contained the same # of keywords within the page text). The result? Our meta keyword page is ranked above the page without in Google and the meta keyword page is the only page that was indexed by Yahoo. So it seems the major search engines do still care about your meta keywords - so don’t leave them out!

Here’s an interesting question: How can you find the best SEO (Search Engine Optimization) firm on the internet? How about using your favorite search engine, which for me, means Google.

Searching for ‘Search Engine Optimization’ on Google returns more than 40 million (!) results so you might be tempted to pick one of the first results to find the best firm. But wait - those first 3 sites in the shaded rectangle are paid results meaning those SEOs didn’t make the natural search cut at all. Ditto for the firms listed on the right hand side of the page. Nope, if you’re an SEO firm and you can’t optimize your own business page for top natural search results, how are you going to help customers do the same for their business?

It’s interesting to note that the first 2 natural results aren’t for SEO firms at all. The first is for Wikipedia which somehow always seems to have great search engine placement no matter the search terms. The next result is from Google warning you about shifty SEO experts who claim to know more than they really do. The third result is for an actual SEO firm, Submit Express, and though I haven’t used them myself their high placement gives some legitimacy to their ability to get clients top billing in major search engines (Google, at the very least).

The fifth result, BruceClay.com, is almost as impressive, especially for such a seemingly small operation. Of course Bruce’s site is super long with tons of text that search engines love, vaguely reminiscent of those scammy sounding e-book sites. Me? I think I’ll pass on Bruce Clay.

When looking for an SEO firm to help improve your natural search result placement, consider how the SEO does with his/her OWN website and be wary of any firm that has to use paid search results to get customers.

Here’s a quickie: Is placing keywords in HTML heading tags (h1, h2, h3, etc.) a good idea for search engine optimization? Specifically, do search engines give higher precedence to pages with keywords in h1 tags vs. lower tags like h2?

If you’re optimizing for Google rankings, you’ll definitely want to put your most important keywords between h1 tags since our h1 keyword page was indexed sooner and listed higher than our h2 keyword page. But just like our previous tests have shown, Yahoo! has a different idea about search engine ranking placing the h2 page above the h1 page in search results. Go figure.

On most of my own pages I tend to place the same text in my title tags as in my first h1 on the page, mainly for convenience sake but it really makes sense from an optimization standpoint as well. Your page title should contain the most important, most descriptive keywords and repeating them in an important location on the page itself is good business. If you’re optimizing for Google, reserve your h1 tags for the most important keywords on the page and watch your listings fly to the top!

I wrote a couple years ago about a neat little trick I figured out for getting my dynamically generated web pages listed in the search engines by converting id references from numerical to alphabetical characters. Eventually the search engines caught up and it was big news when Google and MSN started including dynamically generated web pages in their search results, making this hack interesting but not terribly useful.

I wanted to revisit this concept in light of our current testing to find out not just if a page with a numerical id query string would be indexed but whether search engines had any kind of preference with regards to query strings. The answer, while not exactly shocking, is interesting nonetheless. Here’s the set-up:

Two pages, each with similar keyword density, title tags, etc. but each with a unique query string: one is page.php?id=123, the other is page.php?id=abc. Based on our tests, both pages were indexed by Google at the same time but the alphabetical query string (abc) is listed as the first result - meaning, Google prefers query strings that look like words. Of course once again, Yahoo! takes the opposite track and prefers the ‘123′ page so you’ll have to decide which search engine is more important to your site

If you already have pages with numerical query strings I don’t recommend converting to an alphabetical scheme because it will take time for your pages to be re-indexed (and you may run into duplicate content issues in the interim). Rather, consider what types of query strings your pages will use when creating your next website to maximize your search engine ranking success!

Okay, since it’s Monday I thought I’d toss a softball lob SEO question: How important is the page title in improving your search engine rankings? I’ll give you a hint - the answer rhymes with “berry”.

Yup, our tests confirm what you probably already know to be true: A page with your target keyword in the title will be listed above a page without the keyword. Search engines love page titles for a number of reasons - they are descriptive, they’re near the top of your markup, and their capacity for gaming is limited.

Obviously title tags are meant to be descriptive of your page content but clearly they don’t have to be. Of course your human readers will be confused when the title bar in their browser says something like “German Shepherds make great pets” but your page is about semiconductors.

Search engines also really like title tags because they’re at the top of your page, meaning their crawlers don’t have to read the whole document just to understand what your page is about. I’ll write about this later but it’s usually best to assume search engines work in the simplest way possible - that their spiders are written to be as fast and uncomplicated as they can be. Title tags are near the top so they have priority over the stuff that comes later (your content).

Finally, title tags are technically limited in length by the W3C (the web standards org) to 64 characters. Of course this is just a guideline and some browsers may display much more in their title bars. The point is that search engines are free to truncate your title tags wherever they like meaning you need to place only the most relevant words into your page title. Search engines are also more comfortable ignoring titles over a certain length than they are, say, ignoring your content if it goes too long. The title tag is certainly important but its power is limited by nature.

Take a look at your site and make sure all your pages have unique, descriptive titles and watch your indexed pages soar!

This is one of my favorite myths about search engine rankings because it cuts to the heart of early attempts at “search engine gaming.” The story goes that in the early days folks would place words like “sex” in a white font on pages with white backgrounds so their readers couldn’t read the text but search engines would. Common wisdom said at the time that you would be penalized for such a stunt because it just wasn’t fair or in the spirit of legitimate search engine marketing.

I’ve always been suspicious of this myth myself, especially today when things like font and background colors can be hidden deep within pages of CSS or locked up in complicated background images. Even in the old days an industrious designer could simply set a page background to #000000 (black), place a white background image (.gif) over the top and then write all the “hidden” white text he or she so desired. Were we to believe that the search engines were employing some type of screen capture tool to see how pages actually rendered, images and all? I think not.

Anyway, to our test results. For this test we built two pages: one with black text on a white background, the other with white text on a white background (no css or images involved), both with the same keyword. Both of our pages were indexed by Google around the same time and guess which one came out on top? The black text on a white background - no big surprise. What is surprising, however, is that the white on white page was fully indexed by Google just as if it were a normal page. No penalty for hiding text.

In our Yahoo! test, the white on white page was actually indexed at the expense of the black on white page which was not listed at all. Apparently Yahoo! prefers search engine gaming.

Seriously though, while it is possible to get hidden text indexed by the search engines it is not a strategy we recommend, except in moderation. If you have a sentence or a few words you’d like to have indexed but don’t necessarily want your readers to see, you probably won’t be penalized - just don’t overdo it. Unless of course you’re trying to get listed in Yahoo!