Setting goals in your Analytics will help you, but what you might find is that your high-value pages (your goals) also are typically attractive and helpful pages to your users. It’s also reasonable to assume you’ll spend some time and effort improving the SEO on those very same pages.

Take for example a News site, or a Food site, or a TV Station. They all have key pages that are consider high value (news articles, recipes, or videos, to be specific). Well, those very same pages are also interesting and coveted by the users. It’s also reasonable that those content types are well represented in Search Engines; in fact those three types of content get special treatment from Google (Google News, Recipes, and Videos).

Taking all this into account, reality sets in and these kind of sites notice large bounce rates on those single pages. Is that good? Is that bad? I’m very Gestalt about the whole thing; it is what it is, and it needs to be evaluated in context. There’s no normal, there are useful comparisons but what’s normal for one site, is weird for another.

So take for example one part of the organization that is measured on bounce rates. When it goes up, they cry; when it goes down, they party.

Take then, another group in the company that targets a SEM campaign against those key pages and the bounce rate naturally goes up. Or, take as another example, a SEO effort triggers more inbound traffic directly to those key goal-pages, also resulting in a upward bounce rate. It may seem like the upward bounce rate is a bad thing, but if the overall goal successes are up, is it really a big deal? For some, yes; for others no.

The point: Having a focus on one metric, at the expense of others can cause you to miss the overall picture. Step away from the weeds, and look wholeistically every once in a while and re-callibrate your expectainos against your overall goals. One year you might really care about bounce rates, the next it’s all about conversions, then next it’s all about somthing else. You’re allowed; and should; change.

Related Posts:

• A New slice of the pie
• Chasing your (SEM’s) long tail
• Realty 2.0
• Zillow Enters Online Mortgage Marketplace
• US Newspaper Print Ad Sales Fall 9.4%, Most on Record

There’s a site that’s oddly missing:

http://trends.google.com/websites?q=google.com

I sure hope this is temporary, or an oversight.

Related Posts:

• Google Website Trends … drool
• I Love a Luger case study
• Blog Updates
• Setting up ad ROI tracking with Google Analytics

To explain a bit further, we have data (the xml). Granted it’s not a brand-name database like Oracle, or MS-SQL or MySQL, but it’s structured data nonetheless.

The Application layer is done using XSL’s. An XSL is an xml-stylesheet. Sometimes they’re called XSLT’s (same thing). It’s really the file that controls the layout and structure and business rules between the XML-data and the Presentation.

The Presentation is typically done with CSS. CSS is also capable of position and layout, but at the time XML/XSL/CSS were invented, CSS was best left to fonts, colours, and sizes. Basic stuff.

So, often people will cram style into the XSL for simplicity. Here’s an example. We’re familiar with the structure of XML-Sitemaps. It’s a very simple XML document with , , , and

(RSS files are also very similar, just different labels). An example could be:

<loc>http://www.whatever.com/</loc>
<lastmod>2008-06-18T17:31:47+00:00</lastmod>
<changefreq>weekly</changefreq>
<priority>0.3</priority>

Now structure the basic XML with a XSL file and transform each element into HTML.

So, part of the XSL might look like this:

Here, that simple business logic to transform the
0.2 XML value into a more visual “20%”

To join it all up, there’s a number of options. The easiest is to simply write a reference to your XSL at the top of your XML. Here’s my sitemap as an example:

http://www.whatthefarq.com/sitemap.xml

Normally, sitemaps look like this … not very human-readable:

But by adding to the top:

I can transform the RAW XML into something pretty.  Given more time and energy, I could incorporate brand elements like site structure and navigation into the XSL and it could mirror the look of my site perfectly!  Does anyone have time and energy I can have?

Related Posts:

• A New slice of the pie

For the sake of my discussion, I’ll refer to the three vertical industries with which I’m most familiar; employment, real estate, and automotive.

Wikipedia has a lengthy write-up about the long tail.  Long Tail, generally, refers to a concept that we used to call the “80/20″ rule (ok, they’re not quite the same, but who’s counting?). The long tail is the name for a long-known feature of some statistical distributions. The feature is also known as heavy tails, power-law tails, or Pareto tails. In “long-tailed” distributions a high-frequency population is followed by a low-frequency population which gradually “tails off”. The events at the far end of the tail have a very low probability of occurrence.  Because, in marketing or business, the tail is often ignored due to the fact that it’s made up of small markets, and has historically been hard to reach.  Time’s a changing!

Let’s look to Search Engine Marketing (SEM) for examples.  For the sake this explanation, I’ll use Google, however the same ideas would also apply with Yahoo, MSN and the others.

Search Marketing lets us target users to a very granular degree.  There are lots of stories and border-line urban myth about how expensive search advertising has become. Lots of words out there (pharmaceutical words especially) can cost $10 or more (some even up to $30).  Well, there’s just no need to pay that much.  This is where the long tail starts to wag!

When buying traffic from a search engine, most marketing budgets simply can’t afford to buy every word all the time, so targeting is just a must.  So, think of the two scenarios:

• you pay $1 for the word “jobs”
• you pay $0.30 for the phrase “teaching jobs”

Both give you a click to your site (remember, you only pay when clicked).  So, you can either pay $1 for a visitor or you can pay $0.30 for a visitor.  You ultimately get the same thing, so why pay more?  And, in fact, if you have teaching jobs on your site, arguably the more targeted “teaching jobs” offer you a better quality visitor. The irony is that you pay less for higher quality (often).  And, in fact, the better statement to make is that you pay less when you plan and execute your campaigns in a more targeted and strategic way.

For the really advanced; add geographic targeting to your campaigns and add better landing pages to your campaigns.   In other words, buy “teaching jobs”, and target a campaign to Vancouver. Have as your landing page your Vancouver Teaching Jobs.  All of a sudden, you start to build a perfect storm.

Related Posts:

• Realty 2.0
• Brush aside the trees to see the forest
• A New slice of the pie
• PPE (Pay Per Election)
• SEO’s (and SEM’s) Long Tail

The PHP code to generate this looks like this:
<?php bloginfo('name'); ?> <?php if ( is_single() ) { ?> » Blog Archive <?php } ?> <?php wp_title(); ?>

This looks fine, but I have some issues with this for a few reasons. When all your titles are listed one after the other, they become hard to read, since the beginning of every link is identical (site title, blog archive, page title). This is very common to those that read your blog via RSS readers, or anyone that simply bookmarks more than one post. So, generally, when the Page Title comes after the Site Name and Section Name, it’s hard to read when listed. I generally recommend (not just for blogs) that we put those key elements in our Titles — Page Title, Section Title, Site Title — in that order. This way, when humans read your site’s list of pages (via an rss reader), or bookmark more than one page, your page titles become easier to read and consume.

Above is an example of bookmarks when the Site Name is used first (I love globeandmail.com, but this is a small issue). When looking at this list, the most important item you want to choose (the individual title) is not the left-most element, so it’s harder to choose and digest.

Also, for windows users, the page title shows in the taskbar:

This image shows the “good” version, where the individual page title is at the left-most part of the page title. So, if we used the wordpress default, “Site Title » Section » Title”, it’s very hard to see the page name in this short, and truncated, page title.

SEO people agree with me. The general consensus is that ordering page titles from most detailed to more general is better for SEO too, (but refer to my prior post on SEO. This is yet another example of why we should make these “SEO changes” because they’re good for users, not because they’re good for Search Engines. We should be HO experts, remember — that’s Human Optimization — LOL)

Here is an example of the PHP code I currently use to generate my wordpress titles:

<title><?php if (is_home () ) { bloginfo('name'); echo ' - Blog Home' ;}
elseif ( is_category() ) { single_cat_title(); echo ' - ' ; bloginfo('name'); }
elseif (is_tag() ) { single_tag_title(); echo ' - ' ; bloginfo('name');}
elseif (is_single() ) { single_post_title(); echo ' - ' ; bloginfo('name');}
elseif (is_page() ) { single_post_title(); echo ' - ' ; bloginfo('name');}
else { echo trim(wp_title('»',false), '» '); } ?></title>

Basically, this says:

• If it’s the home page, I use my Blog Name (as found in the setup)
• If it’s a category page, use the category name plus ” – ” then the blog name
• If it’s a tag page, use the tag name plus ” – ” then the blog name
• If it’s a single page or post, use the post title plus ” – ” then the blog name
• Finally, I strip out the silly French Right Quote (the » character) that is added by default to separate the elements

Related Posts:

• SEO is bullshit
• CAIS Study
• A New slice of the pie
• Blog Updates
• Setting up ad ROI tracking with Google Analytics
• SEO’s (and SEM’s) Long Tail
• Google News Crawler

There are some commonly recognized “Search engine optimization (SEO) Techniques” that will generally improve your site’s ranking in search engines. They include such things as

• good choices in domain names,
• using keyword phrases that your users will be searching on,
• using keyword synonyms in your content,
• using keyword phrases in the <title> tag (<title>keyword phrase</title>)
• using keyword phrases in an <h1> tag (<h1>keyword phrase</h1>)
• using keyword phrases in your links (<a>link text</a>)
• Keyword Density
• Meta Description tags
• lots of fussing over “proper” URLs
• and more…

I know I’ll get lots of criticism over this, but to optimize your site for search engines is wrong.

We should be optimizing our sites for humans. In fact, even Google doesn’t recommend the use of the many Search engine optimization (SEO) firms out there. Too often firms employ tactics that do more harm than good… OR, they offer up sound advice and tell you it’s for the search engine’s benefit, but really it’s for the benefit of humans. Sometimes, they just don’t know the difference.

Many SEO firms are well intentioned, and many offer good advice. I just wish they would call themselves Usability Experts or Human Readable Optimization firms.

Let’s break down some of the most common and effective tactics listed above:

• Domain names: This is really simple. If you have a site about the environment, using the word “environment” in your domain name will be helpful to the people visiting your site. It helps them understand who you are and what you’re about. Now, if you’d like to use your brand name instead, fine. It won’t hurt you because your brand name is also an important keyword to you and your visitors. Simply use the best domain name you can get that makes the most sense to your audience and stick with it. Promote it, use it, share it, and link to it. There’s no need to fuss about the search engines here.
• using keyword phrases that your users will be searching on: The SEO guys have it backwards here. What they saying, basically, is if you want to rank higher with the word “environment”, use the word environment in your content. Well don’t be daft! The bigger point is; write useful content about the environment and those interested will find you.
• using keyword synonyms: Well, this is true, but not because it satisfies the search engines. Do this because it satisfies humans, and makes for good reading.
• the <title> tag: Lots is written about this. The <title> tag is basically your pages title (as obvious as that sounds). The title shows in the very top of your browser, and in bookmarks, and the clickable title in search engines, and when in the taskbar (for windows users). The title is pretty important. You’d be pretty surprised how many sites out there have one title for every single page. This is bad. When bookmarking more than one page, it’s impossible to differentiate them unless you have different titles on each page. It’s best to make individual titles for each and every page and ensure your title describes the page (it’s a title, after all). As for the title itself, I’m of the opinion that the more detailed parts of your title should come first, followed by the section and site name. For example, on a fictional shoe retail site (called ShoeStore), a good page title could be “flip flops – Men’s shoes – ShoeStore”. It should be short and it often mirrors or echoes the pages breadcrumb trail.
• the <h1> tag: The <h1> tag is designed to be the pages Headline, so in my previous fictional ShoeStore example, the page should have an <h1>Flip flops</h1> tag. The H1 is useful because many screen readers look for it and it’s just generally a best practice to use it and designers can make it look any way they like using Cascading Style Sheets (CSS). Search engines treat the content inside H1′s with higher priority since they tend to declare the contents subject or persona. Over use of the H1 tag can be detrimental. When everything is important, nothing is.
• use keywords in links: Do this because it is just good web-writing. Here’s an example:
  • Bad: I found an interesting story about the New Brunswick flood. Click here to see it.
  • Good: I found an interesting story about the New Brunswick flood.
• The first example must interrupt the flow of the sentence to offer up the “click here” part, where the second example simply makes the text clickable at the most appropriate time. This is good writing for the web. We shouldn’t interrupt the reader just to offer a link, when it’s not necessary. The SEO experts agree with me, but for different reasons. They say you should hyperlink your text with keywords. Often you’ll end up doing exactly the same thing. See? In my second example “the New Brunswick flood” would be considered important keywords.
• Keyword Density: Write lots of original content and you’ll never need to worry about keyword density! This is why Blogs are so powerful. If you have a blog, and you’re updating it regularity, then you’ll have lots of words on your site in no time. This is great. Search engines like this, because humans like this! We like to read, we like to read interesting and original content. If you have that on your site, the search engines will recognize that and make your content searchable.
• Meta Description tags: The description of your page is important because it gives a clue to the search engines what’s on your page. Some use them, some do not.
• URLs: Whole books could be written on URLs. But it all boils down to making them short and human readable. Why? Because we’re human; we like to share. Short and readable URLs are easy to email or instant message or twitter or dictate over the phone! We’ve all seen URLs that are long and ugly with lots of parameters and values. They break when you email them, and they’re just hard to work with. Yuk. Short and human readable are key!

Sadly, there are lots of examples of “black hat” Search engine optimization (SEO) companies and tactics out there, and even sadder still, lots of companies are paying good money to these companies.

So, you may have noticed a general theme here…

SEO = Usability

SEO tactics are really Human Readable and Usability tactics. Period. That’s why I fundamentally believe that SEO is really bullshit, and we should focus on Human Optimization (but what company would want to call themselves HO experts?).

In fact, I hope this page is a good example. When google “sees” this page, it extracts the following taxonomy (the numbers behind each term indicate a weighting score, of sorts)

search engine optimization (60)
search engine marketing (14)
search engine ranking (12)
search engine position (11)
search engine rank (8)
search engine submission (7)
search engine (41)
internet marketing (5)
optimization (8)
seo (19)
website (7)

So, these are the terms that Google extracts from this page. Pretty accurate, I’d say. All that without special attention to “SEO tactics”. I wrote this page for humans.

Related Posts:

• Better, more user-focused, web page titles
• A New slice of the pie
• Blog Updates
• Setting up ad ROI tracking with Google Analytics
• SEO’s (and SEM’s) Long Tail
• Google News Crawler