The New SEO 2.0 Trends

Web 2.0 is a social entity, and obviously SEO for Web 2.0 needs to be social.

Building social networks on Facebook, LinkedIn, Bebo, YouTube, FriendFeed, Twitter and the like, is only one aspect. If you build them, the followers will not necessarily come. You have to give visitors a reason to become members of your community, and more importantly, you have to give them a reason to click on the links you submit to their attention if you want to fully benefit from the “network effect” so many web marketers are talking about today. Twitter is the perfect example of how this “network effect” can be beneficial. For example, Dell managed to make $3 million in revenue using Twitter to announce special offers and to communicate with their consumers.

Link baiting, another modern SEO technique, has a social aspect too: by publishing content people are really interested in, you basically encourage them to “call out” your site. Good content goes viral in a matter of hours. The readers will “tweet” your link, pass it along (this works pretty much like “word of mouth” ), mention your content on their sites, and etc. This is how natural links are built, and this is the only meaningful way to start a linking campaign nowadays.

Another important SEO aspect is defined as “long tail.” The term was originally coined by Chris Anderson in 2004 to describe the niche strategy of some businesses that sell a large number of unique items, each in relatively small quantities. As you see, the original meaning of the term had nothing to do with SEO. SEOs, however, like the term, and applied it to define an SEO strategy that deals with long keyword phrases (typically containing up to 5 words per phrase), that usually deliver less traffic but higher conversions.

Measuring SEO 2.0

Obsessing with page views and Alexa rankings is obviously not the way to measure SEO success anymore. Your analytics program offers a number of metrics that are more or less relevant, if you know how to read them.

For example, if the best performing keywords are the ones mentioning your brand, then your SEO is not that great. Good SEO delivers traffic for non-brand keywords more than for brand keywords. Of course, it is important to have searchers looking for your brand, but you don’t want to be dependent on it for natural traffic.

Another important metric, if you use Google Webmasters Tools for example, is the number of unique pages crawled and indexed by the search engines. The more unique pages you have indexed in the search engines, the higher your chances to drive traffic and generate sales.

The number of pages driving traffic to your site is also very important: it is in your best interest to keep this number high, to fully benefit from the “long tail” advantage mentioned above. Pages that don’t drive traffic are practically dead pages: they only clutter your site instead of bringing an SEO advantage.

Of course, there are other metrics that help in measuring how successful your SEO 2.0 is. Can you think of any? The comments are open, let’s talk!

Image courtesy SEO Refuge: SEO in the Afterlife

Bing uses different indexing algorithms and a redesigned SERP, which don’t always deliver the most relevant results, but do perform satisfactory in the fields most targeted by Microsoft: health, local, travel and shopping. These are the fields with the highest potential for revenue for both publishers and search engines (while they offer the search functionality for free for the general users, all search engines monetize through ads and the four industries above are the best paying ones).

It is important to know that Bing gives preferential treatment to these four industries: it becomes clear why some searches return irrelevant results, while others appear more relevant than Google. But this doesn’t mean that Bing changes SEO. In a whitepaper released June 1, 2009, “Bing - New Features Relevant to Webmasters”, the Bing Webmaster Center Team notes:

Ultimately, SEO is still SEO. Bing doesn’t change that. Bing’s new user interface design simply adds new opportunities to searchers to find what the information they want more quickly and easily, and that benefits webmasters who have taken the time to work on the quality of their content and website design.

SEO for Bing

However, since the algorithm is different and the search results are structured to fit Bing’s new schema, we can actually talk about “SEO for Bing.” Some of the aspects that influence indexing are described in the whitepaper mentioned above:

• Bing makes it easier to compete for broad terms, because it surfaces more categories automatically, increasing the number of results on the page and generating more content – SEO strategies will have to adapt to optimize content for these categories
• Keyword searches are presented with Quick Tabs that present branches of the parent keyword. This surfaces many websites that rank highly for those keyword combinations – content-rich sites will out-convert sites with lesser relevant text
• Multi-threaded SERP design surfaces many more pages that will be associated with the searcher’s primary keywords than would have surfaced in a single-threaded SERP list
• Bing removes duplicate results from categorized results lists, which allows other, lower ranked pages to be shown in the categorized results on its SERP - hence the poor quality of some results.

According to German SEO expert Tad Chef, who responded to a “Will Bing change SEO?” Twitter survey for SitePoint, Bing definitely plays an important role in the search field for US consumers:

Click on the image to enlarge.

Other respondents (like GlobalPatriot for example) take the “wait and see” approach.

Click on the image to enlarge.

“SEO for Bing” - Best Practices

Like any other search engine, Bing derives standard captions from unique titles and meta description tags. Writing relevant page titles and meta descriptions is a must, especially if you want to win visits from the searchers who use Bing’s “Document Preview” to be sure that the content they are seeking is actually available on your site.

Good content and SEO remain a very important aspect in developing a site, and this is also vividly encouraged by the Bing Webmaster Center Team.

Best of all, the type of SEO work and tasks webmasters need to perform to be successful in Bing haven’t changed—all of the skills and knowledge that webmasters have invested in previously applies fully today with Bing. Moreover, investments in solid, reputable SEO work made for Bing will bring similar improvements in your website’s page rank in Google and Yahoo! as well.

“Bing: New Features Relevant to Webmasters” does not mention other SEO practices: there is no mention of an algorithm, no mention of linking strategies that might work, nothing but “on page” SEO. However, we can safely assume that the SEO strategies that made a site rank high in Live’s SERPs still work. One thing is certain: Bing is based on Powerset, which attempted to rank sites based on semantic technologies. This could mean that “meatier” content has more chances than sites that focus more on visuals.

I will, of course, publish my take on Bing and SEO, but not today. Today I invite you to discover what a modern search engine should really look like.

Visual Is Good

Searchme is a visual search engine, beautifully executed, although they pretty much spoil it by displaying ads by Google on the results page.

Click on the image to enlarge.

The results are usually relevant — for example, searching for mystery novels will give Amazon’s Mystery and Thrillers on the forth position, and some other related and relevant results following, but…

Searching for “sitepoint blogs” comes with an unexpected surprise:

Click on the image to enlarge.

How could this happen? The content on the page listed by Searchme scrapes original content from SitePoint, this is how. But the more important question is: what types of content filters are used by Searchme (note that I set the preferences to filter out adult content) and how is the company going to address this issue?

On one hand, being able to see a screenshot of a page before clicking on a link is positive: we will no longer waste time wondering whether the content we expect is there. The way Searchme displays these screenshots is pretty relevant too. Note in the image below that Searchme generates screenshots that highlight the keywords on the site and that these screenshots are accompanied by relevant snippet of content.

Social Is Good, Too

Another thing I like about Searchme is the “share” feature, which allows sharing the search results via Twitter, Facebook or email. What is very interesting here is that when you click the share button you will not send to your network the bulk search results, but the one that appears in the foreground, with the other results stretching on the right and left, as per the standard Searchme display mode.

Preferences Are Good As Well

Searchme’s preferences allow users to choose whether they want to view adult content or not (although, as I already proved above this feature doesn’t work properly); and offers other choices like the ability to toggle between two different backgrounds for the search engine (night theme and day theme), open links in the same window or a new one, turn media autoplay on or off, and finally chose display mode between Searchme full and Searchme lite.

Stacks Are Great

I simply love this feature: Searchme allows you to save the most relevant results of your choice in a visual gallery. You can create as many stacks as you want, name them and categorize them, and of course, share these with your friends, add them to a blog, and etc:

Searchme.com — the Brand

Not the best brand name; actually, it is so bad that I almost didn’t click on their link when I first saw it. My first reaction was: oh, no, not another Google killer wannabe! I imagined a cheap attempt to create a meta search engine that returns screenshots of the given query, followed by some snippets of text — pretty much like the Searchme Lite version, that doesn’t impress at all, and it doesn’t even represent Searchme’s main strength.

Click on the image to enlarge.

What were they thinking when choosing this name? That people will fall in love with it and suddenly go “searchmeing” instead of “googling”? Is it as obvious for you as it is for me why Searchme is not a good name for a search engine? The call to action, “search me” reminds me of a situation where an innocent person is accused of stealing something and encouraging his/her accuser: “Search me, you’ll see that I don’t have it.”

One main rule of branding is to choose a name that is memorable, and can “differentiate your cow from all of the other cattle on the range. Even if all the cattle on the range look pretty much alike” (Al Ries and Laura Ries; “The 22 Immutable Laws of Branding”). Some companies live under the impression that the quality they offer is enough to make a brand stand out, but quality alone is not what makes a brand. I could go on forever explaining why Searchme is a bad brand name, but right now I don’t think it is very important. The important factor is the technology used by this search engine to display search results. The brand name may change: after all, what is Bing if not MSN and Live rebranded?

Conclusion

Whenever a new search engine emerges we expect something fresh and revolutionary. Searchme is far from being perfect now, but it is a startup, and we have to give all startups the chance to learn and to grow. But, in its current form, Searchme almost convinced me to make the switch. Searchme full (and not Searchme lite) is what the search engine of the future should offer: relevant results in a visually appealing form, with sharing functionality, added interface flexibility and the option to save results in collections! Can you think of anything else?

Apparently it doesn’t matter how many times reputable SEOs like Michael Gray, Aaron Wall, Dany Sullivan and others advise not to obsess about PageRanks – there will always be people who proudly showcase their success in getting a higher PageRank, or get terrified when their sites’ PageRanks drop. Last week I had a commenter who was so convinced that scraper sites helped his site get a higher PR that he almost turned into an advocate of scrapers.

What many webmasters don’t know is that there are in fact two entirely different things that are called “PageRank”. The one everybody is obsessing about is the toolbar PageRank, which really doesn’t matter, because:

• What you see is not what you get. PageRanks are computed continuously so when you see a PageRank 5 that site could in fact have a PageRank 7 or 2 (Matt Cutts).

  PageRank is computed continuously; there are machines that take inputs to the PageRank algorithm at Google and compute the resulting PageRanks. So at any given time, a url in Google’s system has up-to-date PageRank as a result of running the computation with the inputs to the algorithm. From time-to-time, that internal PageRank value is exported so that it’s visible to Google Toolbar users.

• Toolbar PageRank updates do not influence SERPs (Matt Cutts)
  

  It’s just a toolbar PageRank update. Even if you don’t show much PageRank, Google still has 200+ other signals we use in our ranking. It’s definitely common to see lower-PageRank sites ranking above higher-PageRank sites–which tends to confuse the people who obsess too much about PageRank, and who don’t focus on other factors that search engines might use to rank pages.

• Google lowers manually PageRanks for sites that are selling links (Danny Sullivan)
  

  So I pinged Google, and they confirmed that PageRank scores are being lowered for some sites that sell links. In addition, Google said that some sites that are selling links may indeed end up being dropped from its search engine or have penalties attached to prevent them from ranking well.

• It doesn’t determine your Google footprint. Criticizing Steve Rubel’s take on PageRank importance, Vannessa Fox writes:
  

  If there’s one thing that PageRank is not, it’s the determination of your Google footprint. The internal “real” PageRank isn’t even that. Lots of things go into determining your Google footprint. His discussion in the comments goes further down this path of misunderstanding what PageRank is. He agrees with someone in the comments who says that “PageRank is the sum of all other measurements.” It’s not. It’s one measurement added in with a whole bunch of others.

The only thing that does matter is the internal PageRank assigned by Google to a site, but that’s something no one except Google knows.

To make a long story short: there’s nothing magical about having high PageRanks, and definitely not if you obtain these by employing black hat or gray hat techniques (like using scraper sites or other bad neighbours to boost PageRanks). You will never really know what the real PageRank of a site is anyway, not even if you “test it” using such a tool that checks Google PageRanks in 700 data centers.

Sure, you can optimize your site for higher PageRanks, and for more traffic, and for better SERPs, but, as Vanessa Fox said:

If you “optimize the crap” out of your site so that you rank #1 for relevant keywords, but your site isn’t compelling to searchers, that ranking will be completely meaningless as those new visitors will click right back to the search results rather than engage with your site.

An older article, on the same site, reads Deftly dealing with duplicate content and clarifies most of the issues related to this topic. Although the article is dated 2006, things haven’t changed that much in the “duplicate content” camp. Google still wants you to optimize your sites and block all on-site duplicate content appropriately. Google still wants you to keep your internal linking consistent, they still want you to handle country-specific content with appropriate TLDs and they still want you to use the preferred domain feature in webmaster tools. Boilerplate repetition rules haven’t changed, and Google still doesn’t like publishing stubs (although many sites still get away with such practices, but not for long).

It’s still astonishing that Google advises not to worry about scrapers, since these rank often above the original content in its SERPs (image below - click to enlarge).

As a general rule, scraper sites don’t hurt as much as on-site duplicate content – and did you know that your website template counts in the duplicate content calculation too? Behind what the eye sees there is the HTML code that uses “words” to generate the visible layout. If those words in the code outnumber the actual text of an article by 70 % you might have duplicate content issues. So, if you have this possibility, don’t be too lazy to write longer texts.

Last but not least, you should know that on-site duplicate content can also influence your PageRank. You probably thought that when you publish a new site the PageRank is zero. Wrong. Every time you publish a new page it will have a PageRank greater than zero. Google assigns PageRank to sites just for existing – when they first appear the PageRank is based on internal linking structure and content, and not calculated based on external links. What influences the PageRank in such a situation is the content of the site and the number of non-duplicate pages (the more the merrier). I hope this gives at least one answer to those who saw their new sites having a PR4 for example, and then got puzzled because their PR dropped. PageRanks drops because Google’s generosity doesn’t last long. If new pages get a PR greater than usual, take advantage of the opportunity and try to get as many external links as possible to support them, or else, Google’s next PR update will make you wonder WTF!

1. Remember that Google ranks pages, not web sites

Whilst sculpting might begin at your front page, shouldn’t stop there. With a large site (where sculpting is even more critical), you’ll have 2nd and 3rd levels to consider.

2. Less than 100 links per page

Unless there are very special circumstances, keep your total number of links on a page under 100. You’ve only got so much PageRank to pass through, so dividing it up over 100 links is going to stretch the value given to each link. If your page contains more links, use rel="nofollow" for links that you explicitly don’t want to pass PageRank juice to.

3. Editorial links rock!

Different types of links will pass through varying levels of PageRank. Links within your main body of text are given the largest share of the PageRank pie.

4. Think carefully about tag clouds

The user benefits of tag clouds are borderline and in most cases serve to confuse both search engine spiders and users. If you choose to use a tag cloud, nofollow all but your most popular tags.

5. Don’t waste time with footers

Footers will pass through little or no PageRank to subsequent pages, and chances are if a link is in the footer, it’s not to a page that you’re concerned about the search engine ranking anyway –- these links make an easy nofollow target.

6. Don’t go nofollow crazy

When PageRank sculpting, it’s easy to go over the top with the use of nofollow links. Without proper planning this can lock out deeper sections of your site where search engine rank is important. Take a cautious approach initially, then spend time fine tuning your use of the rel="nofollow" attribute.

7. Use secondary navigation and tertiary navigation for deeper PageRank penetration

You may have pages on your site that are two or more clicks away from your homepage. The use of secondary navigation and effective use of tertiary navigation, such as crumb trails, will ensure these pages get their fair share of the passed-through PageRank.

8. Avoid Wikipedia-style linking

Whilst editorial links provide the best pass-though value, avoid the trap of linking every second word in your copy. Not only is this a sure-fire way to look spammy it will easily tip you over commandment number 2.

That’s the summary that I took away from Jane’s presentation, but just to round out our list of tips to a nice, even number, here are two more based on our own experiences here at optimizing sitepoint.com (an ongoing process!).

9. Use the SEO4Firefox plugin

If you’re already looking at any SEO basics, then no doubt you’ve got Aaron Wall’s excellent SEO4Firefox plugin installed. If you don’t, then go grab it immediately. It is an insanely useful tool.

10. Remind Google that you are a respected online entity

Similar to commandment number 6, we’ve received advice that it can be dangerous setting pages like Contact Us and your Privacy pages to nofollow. Even though these obviously aren’t pages that deserve a huge lift in search engines, by passing PageRank juice onto them you’re indicating to Google that you are a reliable, respected online entity. These are pages that might be well suited to including as a footer link.

Hopefully these tips will help ensure that you pass on as much PageRank juice as possible from your home page to your critical secondary pages, where search engine ranking really matters.

For Web searchers though, the reasons are not so obvious. Of course, they can deduce that Google delivers results based on the geographic location of the searcher, which is determined by the IP address of the user. This is generally true for Google local domains too, like google.de or google.com.au, but, the IP address is not the only factor influencing search results. Also, many webmasters will be lost if this was the only factor determining search results in Google. Knowing some of the following, you will also know why search results vary from location to location, and how to optimize your pages to be relevant for your targeted audience.

• Top level domain name: it’s already common knowledge that TLDs are strongly weighted for local search results by Google. Other TLDs like .info, .tv, .biz are given less authority because of a large number of “spammy” sites with these extensions. TLDs with high authority seem to be .edu, .gov, .org and .mil. The most popular TLD remains .com and although it can be used by sites in any country, it still doesn’t guarantee that the rankings will not fluctuate in the Google SERPs. See below.
• Server location: as I said, .com can be used by anyone, but Google also takes into consideration the geographic location where the domain is hosted. A .com hosted in Australia is seen as an Australian site, whereas a .com hosted in Germany is seen as a German site. So, if you target an American audience for example, and you want to be given priority in the US search results, getting hosting in the US is a good move.
  
• Location of ITL (incoming text links). If your target is international try to get links from sites hosted in as many different countries as possible. The same is true if you target local audiences, get as many links from sites hosted in the country you target. For example, if most of your links come from Australian sites Google will figure that the site is Australian one, or of interest to Australian searchers.
• Page language: it’s possible to rank for pages you translate in other languages too, even if your site is not hosted in the country whose language you target. In this situation the ranking factors include, aside language, encoding characters, and meta titles and descriptions, which should be translated too.
• Last but not least, take care how you set up your Google Webmaster Tools. When you sign up for this service, Google asks you to set up the geographic location of your site. You can also set this up at a later point, or edit it, but this option is not available for a country specific TLD like .de, .fr, etc. But for .com and other generic TLD sites, this Google Webmaster Tools setting replaces the server location signal and is particularly useful to set a different geo-location for each subdomain you want.

I hope this little primer on page relevance is helpful in understanding how Google reveals results for users, and how you can optimize your site for this variable. With so many factors to consider in SEO, using relevant but crucial tidbits like in these primers is one way to incrementally improve your search ranking easily and painlessly. Obviously everyone cannot change their domain name, server location, languages and links all at once, but those things that can be optimized, should be, if ranking in the search engines is important for your site.

Do Not Optimize for Keywords that People Don’t Search For: Traffic Volume vs. Google.com Competition

Research tools are great, but they can never replace human intelligence – after all, you are targeting human visitors and not bots. Many SEOs promise high SERP results in 24 hours or less for a given keyword phrase, but is that phrase relevant? Sometimes the keyword phrase in point sounds “good”, but if no one searches for it; there is almost no use for its high rankings. Over the long term it might bring a visitor or two, but in the short term you will face a situation like: “I have top positions for many keyword phrases, but the search engine traffic is very low.” The point is: do not expect traffic for keywords that people never use.

This is where the keyword research tools I mentioned last week come in handy. Before choosing to optimize for a certain term, test it with various keyword popularity tools to see how many visitors might be interested in it; then test the phrase with Google AdWords suggestion tool to get some indication of performance within Google’s advertising network. This helps you gather some performance data, but please be aware the research is not complete without a few “refinements.”

You can see how many sites compete for the same keyword phrase by simply typing it into Google’s search query box. Use Google.com if you target international rankings, and use the local version of Google if your target is in your geographic.

For example, for the term online PR Google.com delivers over 65,000,000 results. This means over 65 million possible competitors, including some of the most powerful public relations sites on the Web.

To optimize a site for a term with so much competition might be tempting, and seeing your site on the first page in Google.com for this term is certainly an “ego booster”, but does this term convert well? The WordTracker tool I mentioned in my last article estimates about 5 daily searches for this term.

Assuming that your site ranks among the top 3 results in Google.com’s SERPs, out of 5 users who might look for “online pr” on a given day, how many will actually land on your site?

Popular Doesn’t Mean Appropriate! – Business Value vs. Traffic Volume

Starting from the example above, let’s see which is the most popular keyword phrase related to “PR.”

WordTracker also says that “PR Newswire” is the most popular term, with over 400 searches daily. Obviously, if you run a PR site you would like to capture some of the potential traffic for this term.

Google.com says that there are only 29,300,000 sites competing for the term – obviously much better than over 65,000,000 for “online PR.”

But, is it appropriate to optimize for PR Newswire? No. PR Newswire is a well known press release distribution vendor, or in other words a well known brand. Unless you are affiliated with PR Newswire, optimizing for this term is pointless: you might get the traffic, but it will not convert. When you choose a keyword phrase keep in mind what your users want and whether that phrase is relevant for your own site.

It’s hard to “guess” what a user wants, but the keyword research tools do provide a lot about what users are looking for. A search for PR Newswire indicates clearly that the users wants to find www.prnewswire.com or reviews about the company. A search for “online PR” is too nebulous: the searcher might be looking for general information, definitions, or even a company specialized in this field. In contrast, a search for “PR Prices” might indicate a users looking to buy PR services, or a PR apprentice researching the PR industry.

From this short analysis we can conclude that one of the most important factors when choosing a keyword is the “business value” – meaning how the term will convert for your business.

Be Specific and Comprehensive: Relevancy

When users search for specific products online, they try to refine their searches by typing in longer keyword phrases, getting closer to the “semantic search” ideal from a behavioral stand point. These “longer” key phrases hold higher relevancy than generic terms. Let’s see a “refined” version of a search containing “PR” – our “root” keyword for the day.

A customer running a law company will most likely type in “pr for law firms” – which is a “long tail” keyword phrase, assuming that you offer pr for law firms.

This phrase is very specific to what you are selling, and it will probably convert a visitor into a customer. Conversions depend on a large number of factors, not only on keyword choice – but we will discuss this another time.

Relevant and highly specific keyword phrases rank faster and easier than generic keyword phrases, and the competition is usually less daunting. Generic keywords do bring more traffic, but the conversion rates are very low. Unless you depend on traffic stats to monetize your site, it might be a better idea to go for relevancy rather than the market value.

Nothing Lasts Forever: Search Trends

It’s hard to predict what search users will be looking for in the next season. Sometimes search traffic fluctuates based on trends that are influenced by economical changes, season, fads and etc. It’s always a good idea to check out a keyword trend tool to determine what’s hot and what’s not. Recommended: Google Insights for Search.

To see what’s “hot” with our “root” term, simply type in the Google Insights query box your term, then refine the search by selecting the appropriate time range and a relevant category. For “PR” in 2009, business, we note that the top searches are marketing pr, jobs pr, pr firms, media pr, etc, while a very important “rising” search is social media pr. The rise of the “social media pr” is no accident, considering the actual Web trends. Although WordTracker and other keyword suggestion tools show low daily traffic estimates for most of the “rising” search terms, remember that there are many situations where software cannot replace human intelligence. Always watch the market trends and use the rising keywords phrases to your benefit.

Conclusion

So, tools are essential for optimizing the selection of keywords for SEO value, but they are only effective if used properly. It is also important to understand that search engine relevance is a two way street. Google is trying to supply the most relevant results, search users want these relevant results, and viable businesses want to “be” relevant for potential customers. Ranking without relevance can only produce “visitors”, not long term customers. Keyword selection based on human logic and experience, refined and narrowed with the right tools, is the correct methodology for selecting the most suitable keywords.

In the first session of SMX Sydney, Rand Fishkin, CEO of SEOmoz told the story of a recent startup that woke up one morning to find their organic search engine traffic from Google had vanished for no obvious reason. Traffic from Google for this particular startup represented the lion’s share of their daily visits, so it had a dramatic, and instant, effect on their bottom line. I’m sure this is an all-too-common story of web sites becoming over dependent upon a single source of traffic.

With Search Engine Marketing, particularly SEO, you can never forget that the engines owe you nothing—they act in the interests of the customer experience (usually the relevance of the results that they deliver), which drives demand, and subsequently their advertising revenue. Granted, all of the big players play well with Search Engine Marketing crowd, but at the end of the day, they only really care about what’s in it for them—not for you …

The key piece of advice here then is to not let your healthy search engine performance today lull you into a false sense of security for tomorrow. If a switch within Google gets flicked, or a lever gets adjusted that influences the search engine algorithm, your search engine performance could change drastically. You need to be able to cope with it.

Rand highlighted the need to stay in touch with best-practice SEO techniques, so you can ensure you’re at the forefront of change. From an SEO perspective this is certainly solid advice, but the advice should not just be applied to search. Relying on any one specific channel for traffic is a big risk to take. Each and every source (direct, referral, advertising, social media etc) has some potential for change, just as search engines do.

A well-planned site will consist of an even mix of traffic sources. The site managers will work at building traffic across all channels, rather than focusing solely on one. This way, if the well was to dry up in one area, plenty of water (traffic) would still flow from other channels.

Take a look at your traffic sources in your current web analytics package and ask yourself: if one of these channels stopped delivering traffic tomorrow, what would it mean to my business? If there’s one channel that dominates all the others, it might be time to look at bringing balance to your incoming traffic sources.

Forget about keyword density – this is not what determines your site’s position in the SERPs anymore. Keyword density is too easy to game. Instead of focusing on numbers try focusing on:

• quality and value for the readers – if what you write is not what people want to read, keywords will make no difference
• content harmony - many writers tend to repeat keywords too often, trying to convince the search engines to “rank” their sites for those specific terms. The technique is called keyword stuffing, and the effects are negative: exasperated readers, low rankings in the SERPs. The solution: less is more. Keep keyword density under 3%
• relevance for what you are trying to present in front of your readers – don’t write about shoes when you sell boots. Both are footwear right, but the customers who want to buy shoes will feel mislead and frustrated to have entered the wrong store. Do not mislead your readers for the sake of SE rankings! Put it this way: traffic is meaningless if it doesn’t convert into buyers, returning visitors and/or community.

So, if keyword density is no longer a factor in search engine rankings, then what is? Are keywords no longer important for SEO? As a matter of fact, they are. Now, more than ever, you need a good keyword strategy to write copy that places your site higher in the SERPs. It’s not so much a matter of density as it is a matter of keyword placement within the copy: page title, meta description, article title, first paragraph, image title, outgoing links, etc. We’ll talk about how to select the best keywords and how to write compelling SEO copy in a future article. Today we start with the basics: some of the best free keyword research tools.

WordTracker free keyword suggestion tool – it can generate up to 100 related keyword phrases and estimate their daily search volume. WordTracker uses it own formula to estimate search volumes: number of searches in the WT database divided by the total number of searches in the WT database and multiplied with the estimated total number of daily searches on all search engines. To estimate the searches on all search engines WT collects terms from Dogpile.com and Metacrawler.com (0.63% of searches across all search engines). This is the reason why you might see major differences between keywords suggested by WT and other keyword research tools (Google for example.)

Google search based keyword tool – free for all users, but made to help Google AdWords customers, it delivers keywords based on actual Google searches and it matches these to specific pages of a site. I find this tool more useful than WordTracker’s free version because it offers more data including monthly searches, competition, suggested bid (let’s you know the financial value of a certain keyword), suggested categories for the targeted keywords, keyword in URL, etc.

SEO Book Keyword suggestion tool– free and powered by Wordtracker’s API. The tool offers rough suggested daily search volumes by market for Google, Yahoo!, and MSN and links to vertical databases like Topix, Google blog search and Del.icio.us. It also links to Google Trends, Google Suggest and others to give you a more in depth analysis.

To help you generate a list of keywords, Aaron Wall offers a Free Keyword Phrases List Generator, which is a web based software that allows generating a large number of keyword phrases based on permutations of keywords entered.

Good Keywords v3 – downloadable Windows software (freeware) meant to help you get more out of your Google Keyword research. The developer Softnik Technologies claims that the software does not abuse Google’s resources in anyway.

Free Search Term Suggestion Tool from Trellian KeywordDiscovery can generate maximum 100 keywords and it was developed as an alternative to Overture. Data is collected from over 200 search engines worldwide. KeywordDiscovery also offers language specific databases.

SEMRush is a more powerful tool, that can help you discover competitor sites with common Google keywords, get a list of Google keywords for any site, investigate long-tail keywords etc. It its free version the tool offers only 10 listings for each query. Another drawback is that it only uses Google.com – not ideal for local keyword research.

KGen is a FireFox addon that allows you to see what keywords are strong on visited web pages. This is a great tool to use to analyze competitor sites. It allows you to select some words and transfer them to your clipboard and paste them on tag fields on your favorite social network or fill meta-keywords of your web pages.

Demographics Prediction is a tool designed by Microsoft to help you gather users’ age, gender, and other demographic information based on their online behavior (search queries and visited sites).

At the beginning of the article I told you to forget about keyword density for a good reason: Google is now analyzing keywords within the context they are place in, looking at semantically related terms like synonyms and other words related to the key-phrase used in search. The process is called latent semantic indexing, and it’s been around since 2005. Pages over optimized for one single keyword phrase are often being filtered out by Google for being “over optimized”, whereas pages that use a wide array of related terms have more stable rankings for the core keyword, and might rank for the related terms too.

Below are some tools to help you select semantically related terms:

Keyword Map helps you find both synonyms and some popular keyword phrases.

Urban dictionary allows you to find slang terms, and definitions written by volunteers.

A good synonyms finder is synonym.com.

Keyword Questions – people type questions in the search engines and this is a tool that helps you find them. Answering to these questions will most likely deliver more traffic to your site. The tool is great for long-tail optimization purposes.

Social media can help you see the keywords of the moment: use Twitter Search and Twitscoop to see real time keywords, Deli.cio.us will show you an array of related terms. Even Facebook can help you trace keyword popularity within the network by showing you how often a certain term is mentioned across FB walls.

Don’t forget that dictionaries, encyclopedias and thesaurus can be of great help in your keyword research efforts as well. For example, MetaGlossary harvests definitions from the entire web.

Good research is the result of using the right methods and tools in preparing for any SEO task. This is especially true for keyword insertion. With the suggestions I have provided, and a little time spent developing the right strategies and combinations of tools, anyone can not only write successful SEO copy, but write compelling content for their readers as well.

If you know more free keyword research tools, please share them in the comments below.