Nowadays, we hear both SEO professionals and search engine representatives talk about how much they care about the user experience. For example, usability is now a service that some SEO firms offer, regardless of staff qualifications. Consider the credibility of a usability firm that does not conduct and analyze usability tests. Or the benefits and limitations of website heuristic evaluations, popularized by usability guru Jakob Nielsen.

Another website development service that is re-emerging in the SEO industry is information architecture (IA). Like link development, user-generated content, and long-tail keyword optimization, information architecture—as an SEO strategy—has existed for many years. However, an SEO professional’s spin on information architecture can lead website owners down the wrong path. Here are some common mistakes that SEO professionals make that can negatively affect a website’s search engine visibility and ROI:

Mistake #1: Not considering different types of searcher behaviors

Querying is not the only finding behavior that searchers use to discover and locate desired content. It seems as if usability professionals, information architects and SEO professionals focus on different aspects of findability. Usability professionals and information architects tend to focus on browsing behaviors, whereas SEO professionals tend to focus only on querying behaviors.

Content is more findable when website owners accommodate searchers’ browsing and querying behaviors.

Many information architects and usability professionals focus on finding behaviors without considering how their categorization, organization, labeling, and prioritization affects both web and site search engine results.

On the other hand, SEO professionals often create information architectures without considering how users/searchers organize content.

In reality, querying and browsing are both critical components of findability. Websites should accommodate both types of searcher behaviors, not one at the expense of the other. Therefore, for the best user experience, SEO professionals, information architects, web developers and usability professionals should all understand the various aspects of findability.

Mistake #2: Creating a website information architecture exclusively from keyword research data

Unfortunately, this mistake is rampant in the search marketing industry because, to some degree, some information architectures are more effective with keywords. However, many SEO professionals let keywords in site navigation get out of hand. There are multiple relationships among web pages, and many of these relationships are not purely topic-based.

For example, when is it appropriate to categorize or prioritize content alphabetically? Should content be organized by date and time? By industry? By user type? Information architects take all of these items into consideration when creating site navigation scheme(s) and supplemental page interlinking.

I believe that keyword research data and analysis should be a part of the discovery process for both information architects and usability professionals in addition to the tools and techniques they already use, such as formative and summative usability testing. That being said, if an SEO professional wishes to become a more effective information architect, he/she should learn how to do the key skills of an information architect, including but not limited to:

• Categorization and classification
• Organization of content/information
• Creating and maintaining effective taxonomies and ontologies
• How to effectively prioritize
• Usability testing

Information architecture is not for dummies. For now, I believe that the hybrid of information architect and SEO professional is rare. Hopefully, we will see more evolution in this area. Until the time comes when these industries merge, I think it is best for website owners to hire an information architect and an SEO professional to work together to create a website’s information architecture.

Mistake #3: Not understanding user/searcher mental models

If you ask a web developer to create a website’s information architecture, the result is often a website that matches the mental model of the web developer, not the target audience. Likewise, if you ask an SEO professional to create a website’s information architecture, the result is often a website that matches the mental model of an SEO professional, not the target audience.

Another excuse I hear is to imitate the site architecture of websites whose pages rank well, regardless of usability. Plenty of poorly architected websites rank well. That does not mean that they convert well or get long-term, cumulative link development. A well-architected website often receives higher quality link development, can more easily address duplicate content delivery, and archives content easily.

On the flip side, I believe that information architects should understand searcher goals and behaviors not only within websites (site search), but also on the commercial web search engines.

As much as I appreciate the evolution and legitimization of SEO, I know our industry has a long way to go. I once remember a reader comment that said information architecture is an extension or subset of SEO. I respectfully disagree. I believe SEO is an extension or subset of information architecture. And if our industry is going to successfully evolve, then we need to learn skills and accumulate knowledge from other industries. Because, ultimately, we are all on the same team.

Perhaps one overlooked benefit of successful SEO is customer loyalty. I observe this in long-term search behaviors all of the time. When users/searchers discover a website that contains desired information, they will often use search engines to go back to that site (refinding search behavior). If the website is optimized well, then returning to the site, or a specific page on a site, is an easy process. Advanced searchers will often use Google or Bing (or their favorite search engine) to find specific articles on a site, such as the following:

lung cancer treatment site:www.cancer.gov

When I observed these frequent navigational queries, I realized that loyalty might be a major benefit of SEO. Google gullibility might be a behavior that SEO professionals rely on to get initial traffic for websites, but findable content delivered on a user-friendly website will get searchers to stay on the site and keep them coming back.

SEO and the System Usability Scale (SUS)

“Search specialists get paid to worry about getting people to a website through careful selection of words that attract users,” said John Sorflaten, Ph.D, who has designed user experiences using systematic methods that verify the quality needed for professional results. “There are additional ways of attracting users, too, one being ‘loyalty’ to a company or its website.”

One way usability professionals measure customer loyalty is through the System Usability Scale. Usability professionals use the SUS questionnaire to determine how well an interface serves its users. The System Usability Scale questionnaire is a commonly used, freely distributed, and reliable questionnaire consisting of 10 items. Scoring the questionnaire yields a usability score in the range of 0–100, its ease of administration and scoring makes it a popular choice among usability professionals (Finstad 2004).

“Substitute the word ‘website’ for ’system’ in the questionnaire, and it works well for measuring how usability test participants feel about their web experience” said Sorflaten. “Founders Thomas S. Tullis and Jacqueline N. Stetson found that the SUS questionnaire served better than several other popular questionnaires that took longer to administer.”

“Recently, Jeff Sauro, a buddy who specializes in measuring usability found an interesting correlation between loyalty to a website and the usability experience visitors had at a website,” Sorflaten continued. “He found that one third of their ‘loyalty’ comes from the degree to which people feel the site is ‘usable’ as measured with the SUS questionnaire. For SEO folks, it means that you can include ‘loyalty’ as an additional draw for getting people to your clients’ sites. So you have to know techniques of usability to help generate loyalty.”

Sauro’s research showed that to get loyalty from your site visitors, your SUS questionnaires needed to have an average score of 80 or above out of a top score of 100.

“It’s nice to see concrete data to back up our interpretations of SUS scores,” said Sorflaten. “To get data, however, you must interact with real end users with usability testing and questionnaires. That’s usability. No ivory tower assumptions are allowed when you want to create good websites that get a loyal following.”

Usability and the searcher experience

As an SEO professional, I have long believed that findability is a critical component of website usability. To me, that means querying behaviors are just as important as browsing behaviors. In fact, the two behaviors are inseparable. Web searchers do not simply type a keyword phrase into a search box, click search, and stop. They scan search results. They read individual search listings. They want their user-generated information scents validated in search results and on corresponding websites.

Additionally, customer loyalty can have an effect on a site’s long-term link development. Two sites can have identical or nearly identical content. Which site do you believe people link to? You guessed it—the site that is easier to use. I like the System Usability Scale because it allows us to interact directly with searchers and makes upper management happy with numbers they can use and relate to. Understanding how and why searchers interact with websites tends will generate the best ROI.

Readings on Customer Loyalty

• Determining What Individual SUS Scores Mean: Adding an Adjective Rating Scale
• The System Usability Scale and Non-Native English Speakers
• Does Better Usability Increase Customer Loyalty: The Net Promoter Score and the System Usability Scale (SUS)
• Wherefore Art Thou O Usability? – Cognitive lock-in to the rescue
• Tips and Tricks for Measuring the User Experience

Sometimes, a SERP addition is successful and becomes a permanent feature. Sometimes, the addition is a dud, and in the blink of an eye, the feature disappears before it gets too much publicity. And sometimes, the new addition just needs a little tweaking before it becomes a permanent fixture in SERPs.

I have watched social media creep into search results pages over the years. And I did select the word “creep” for a reason. In reality, social media existed long before Twitter and Facebook came into existence and became popular. People wrote articles/posts, and responses to articles, long before blogs came into existence. Yet links to articles and their responses, product reviews, and the like have appeared in search results for many, many years. We just didn’t use the label “social media” at the time.

Personally, I preferred using SERPs before other forms of social media became the rage. I think the commercial web search engines are giving social media items far more attention and validation than they deserve.

Blogs, forums, and tweets… a conversation

I understand the allure of real-time search. In one word? Recency. Search engine representatives assume that web searchers always want the latest and greatest information.

For example, an online shopper does not want to click on a link in a SERP to a merchant site, only to discover that the product is out of stock, especially during the holiday season. In an emergency situation, such as a natural disaster, searchers want to know where to go and what to do. News always has a recency element. In addition, response to newsworthy items can be important to users, especially if there is a difference of opinion. These conversations and debates can be a great source of information before consensus is reached.

My question is: do these conversations truly deserve prominence on the front page of search results? Which is more important: the conclusion/consensus or the content that leads to the conclusion/consensus? Or are they both important?

Before answering these questions, consider your objectivity. If your site is benefitting from these real-time, online conversations, then maybe your answer is yes. “Social media rocks because my site makes thousands of dollars from it,” is one point of view. If you are the victim of an unfair or inaccurate statement, then maybe you have a different perspective. There are equally as many people and/or sites that have been hurt by social media antics, many quite unfairly.

For example, let’s use the SEO industry. Every year or so, our esteemed Editor-In-Chief Danny Sullivan ends up defending the SEO industry due to widespread ignorance and prejudice. See Dear Fox News: SEO Is Not Search Engine Scamming (2009), Why the SEO Folks Were Mad at You (2007), Yes Virginia, SEO Is Rocket Science (2006), Worthless Shady Criminals: A Defense of SEO (2005). Search engine optimization, as a profession, has been around for a very long time. There is plenty of great content available on the web that explains our industry, even the different tactics used to obtain qualified search engine traffic. Yet we must suffer through this sort of ignorance year after year. Does this ignorance deserve top search engine visibility just because it is recent?

People treat top Google (and other search engines) results as the “best” results, which is unfortunate. Usability guru Jakob Nielsen calls this search behavior Google gullibility. Are real-time results the best results? Or are most of these conversations just fluff?

User-generated content… really?

User-generated content is another item that has been in existence long before the keyword phrase “user-generated content” became popular.

What many people might not realize is that some companies and SEO firms pay people to write user reviews—whether the reviewers actually purchased a product/service or not. You can actually use Google and find people who are willing to write fake reviews or blog posts for payment. People pay reviewers to write bad reviews about their competitors… and they will purchase their competitors products just so they can write the bad reviews. It is not necessary to purchase a product or service to write a blog post. I see it happen all the time.

And if the review is genuine? Well, people who have a negative experience with a website tend to tell more people about that negative experience than people who had a great experience. The end result? Some really great products, services, and content get unfairly evaluated. It happens more frequently than any of us care to admit.

I honestly believe that the commercial web search engines are giving social media items far more attention and validation than they deserve. I can remember when I really admired our industry. Instead of a marketing department telling us what we want and how we should get it, web searchers turned it around. With the search engines, we are able to tell marketing departments what content we want to see on a website, not the other way around. If I want to watch a video, then I will type in the word “video” as a keyword. Don’t shove a video in my search space when I do not want to see one.

How about you, Search Engine Land readers. What do you think about search engine results pages including social media content? Are they getting better… or worse?

In my opinion, too many search professionals associate search behavior with querying behavior only. In reality, web searching is more complex than simply typing keywords into a text-entry field. Website owners and search professionals alike often overlook finding behaviors after a searcher clicks on a link to a website from a search engine results page (SERP). One of those finding behaviors is called orientation or orienting.

Orientation, SERPs and landing pages

What exactly is orientation? On a website, orientation is a behavior whereby users determine their position in a website with reference to another point—establishing a sense of place.

Many times, the reference point is a home page or a website’s domain name. However, when people click a link from an organic search listing or a search engine ad, they don’t always go to a site’s home page. They most likely land on a page in the middle of the website, or a landing page created specifically as a destination from someone clicking on a search engine ad.

For web searchers to feel confident that a page or a site offers the product, service or information they desire, web pages should present clear “you are here” cues. Web searchers use a wide variety of “you are here” cues to determine a sense of place on a website—both textual and graphical cues. As a search usability professional, I want to understand which textual and graphical cues are important to my target audience. Where should these cues be placed on a web page? If it is a textual cue, how should it be formatted (color, font/typeface, white space)? If it is a graphical or multimedia cue, how large or small (in dimension) should it be? Where should these cues be placed on a category page or an article page?

Here are some questions we commonly ask web searchers during usability testing to determine their mental models before they click on a link on a SERP:

• Whose website are you about to view? How did you determine this?
• Which section of the website, if any, are you about to view? How did you determine this?
• What content do you believe you will see after clicking on this link?
• Do you believe that the information you desire will be available after you click on this link? Why or why not?

For example, if web searchers use a URL or domain name as an information scent, which is extremely common for navigational searches, they often look at the logo, which is a graphical “you are here” cue to establish ownership of the site. A tagline or a slogan can also be a “you are here” that establishes and reinforces site ownership. Web searchers can also quickly scan the URL, which is a textual cue. This orienting process occurs very quickly (often in less than 1 second) and is a normal process when people navigate from web page to web page.

The benefits of accommodating orienting behavior

Why should website owners accommodate orienting behavior? One reason is user confidence. Providing consistent “you are here” cues throughout a website communicates trust, reliability, and dependability because searcher mental models are being reinforced and validated on every page.

Additionally, consistent placement, usage and formatting of “you are here” cues are important because they decrease demands on users’ attention, allowing them to accomplish their desired goals more efficiently and with fewer errors. In other words, if searchers spend too much time trying to establish a “sense of place” on landing pages they are spending less time and effort trying to accomplish their desired goals—goals that are important to business owners as well as they lead directly to conversions (add to cart, subscribe, enroll, etc).

Finally, recognition, recall and memorability tends to increase when you have provided consistent “you are here” cues on a site. In the event that searchers wish to re-find content on sites via the commercial web search engines, retrieving that content is easier. With minimal effort, searchers encode these “you are here” cues within their memory along with the information they learn on a web page, making content easier to retrieve at a later time.

Therefore, website owners, interaction designers and search engine advertisers need to spend more time making the orienting process as quick and easy as possible. Orientation is a search behavior that no SEO professional, search engine advertiser or website owner should dismiss. Quick-and-easy orientation contributes to a positive brand experience, increases conversions and sales and makes content easier to find.

Website information architects try to determine how users categorize, organize and label information on a site. Information architects use a number of methods to determine the best site architecture, including but not limited to:

• Field interviews
• Direct, one-on-one observation of users/searchers performing their normal, daily tasks
• Usability testing
• Data from web analytics software, site search engines, and so forth

On a website, an information architect’s goal is to determine a formal site navigation and other forms of page interlinking, ones that best correspond to the mental models of the site’s users. An effective site architecture should enable users/searchers to accomplish their goals more easily and efficiently. With every click, a user’s information scent should be reinforced and validated without distracting, confusing or annoying the user. Additionally, a site’s information architecture should communicate the “aboutness” of page content to both search engines and site visitors.

After three iterations of usability testing on one particular business-to-business healthcare website, I noticed something interesting: pages that the in-house SEO professional created did not match the mental models of the primary and secondary target audience. Words such as “fluff,” “propaganda,” and my personal favorite, “what the [expletive],” were used to verbally describe these pages. Furthermore, these same words appeared in test participant comments and category/section labels.

Digging a little deeper, I also noticed that all of their competitors created web pages that did not match user/searcher mental models. Yet these pages were ranking well. So even this company’s competitors were not creating sites to accommodate user/searcher mental models.

Why would any website owner create an information architecture where a considerable number of SEOed pages belong in the category labeled, “What the [expletive]?” Why do SEO professionals continually build pages, and entire websites, that do not match searcher mental models?

What is a mental model?

A mental model, also known as a conceptual model, is an explanation of a person’s thought process about how something works in the real world, faithfully representing root motivations and matching behaviors. Everyone has a mental model about how a website or a search engine works, and no one person has the same mental model as another person. Nevertheless, some portions of mental models are consistent from person to person.

As an example, let’s use an elevator. Most of us have the same expectations and experiences with riding in an elevator. If we press the button labeled “2” inside the elevator, we expect the elevator to take us to the second floor. If we press the button labeled “5,” we expect the elevator to take us to the fifth floor.

How do we know we are moving toward the fifth floor? On most elevators we usually see a number that lights up when we arrive at or pass a floor. What happens when we arrive at our destination, the fifth floor?

• The number “5” is illuminated on the elevator panel
• The elevator stops
• We often hear a “ding” to indicate that the elevator doors are about to open
• The elevator doors open
• Usually, when we exit the elevator, we can immediately see some sort of visual cue that we have arrived on the fifth floor, such as a sign showing room numbers 501-540, and/or the number “5” somewhere within our immediate visual range.

The textual and visual cues on an elevator are similar to the textual and visual cues on a website. When searchers click on a link on a search engine results page, they expect to be delivered to a page that contains their targeted keywords. But keywords are not the only item on a web page that searchers expect to see.

Searchers have mental models of websites and web pages. They expect to know which elements on a web page are clickable and those that aren’t. On ecommerce websites, searchers expect to see product photos. Headings, categorization and navigation labels on a healthcare site that targets physicians and other healthcare professionals will be quite different from headings, categorization and navigation labels on a healthcare site that targets consumers.

Mental models of SEO professionals

How do many SEO professionals address searcher mental models and site architecture? Here is a partial list:

• PageRank (PR) sculpting (via nofollow tags and other methods)
• Siloing
• Using targeted microsites
• Link farms and other forms of search engine spam

Unfortunately, many SEO professionals are not validating searcher mental models, though they honestly and sincerely believe they are. If a page ranks and a web searcher clicks on a search listing, then the assumption is that the web page matches the searcher’s mental model. Other pages ranking well? Then the SEO assumes searcher goals are obviously being met. More clicks? Even more evidence. Low bounce rate? Web searchers must love the site, even though more clicks on a website can indicate confusion, not user satisfaction.

Never mind that information architects and usability professionals continually see “what the [expletive]” as a site architecture label.

It seems as if SEO professionals and website owners are building websites and pages based on their personal mental models, not the mental models of the target audience. No professional, qualified information architect would recommend a site architecture based on data purely from keyword research tools.

Would I hire an SEO professional to architect a website? Not unless that person or company has a strong education, training and experience in library/information sciences. Many professional information architects have advanced degrees in this field. However, I would hire an SEO professional to contribute to the site architecture discussion. Querying is a search behavior that no website owner should ignore.

• Facilitate clear navigation (where can I go)
• Allow for quick orientation (where am I) and
• Communicate content value (should I click on this link)

Navigation – where can I go?

On search engine results pages (SERPs), the main textual cue that communicates, “Where can I go?” is a blue, underlined text link.

For example, in natural (algorithmic) search listings, the primary blue, underlined text link contains the (X)HTML title-tag content. In the news search listings, the primary blue, underlined text link is the story headline. Title-tag content and headings serve multiple purposes:

• Relevancy (ranking). All of the major search engines use title-tag and heading content to determine page relevancy.
• Click-through. A hypertext link is the primary call-to-action on SERPs.
• User confidence. If keywords are present in the hypertext link, it increases user confidence. Searchers believe that if they click on a link that contains their keyword phrase, they will go to a page that contains that keyword-related content.

However, search engines display more than text in their SERPs. In the past few years, search engines have been displaying thumbnail photos that lead to web pages containing graphic images and videos. These thumbnails are also part of the scent of information. What do searchers expect to see when they type in the current U.S. president’s name, Barack Obama? Do they expect to see a thumbnail photo with Barack Obama in it? Or do they expect to see a picture of the Queen of England? Or a kitty cat dressed in a cowboy outfit?

Interestingly, many SEO professionals sincerely believe that link development and social media trump all on-the-page factors for optimization. I have never believed this nor do I practice it, because I see how important the scent of information is to users. No link development campaigns will be successful if information scent is not reinforced in SERPs and corresponding landing pages. If the scent of information is strong, people click. If the scent of information weakens or disappears, searchers abandon the website.

Orientation – where am I?

On a website, orientation is a behavior where searchers determine their position with reference to another point, establishing a “sense of place.” In other words, searchers quickly establish whose website they are visiting, and what section of the site (if any) they are viewing. If searchers do not believe they have “landed” in the right place, they will leave the website. Web searchers orient very quickly, sometimes within 1 second after a page loads.

Landing pages should always validate searchers’ scent of information, both textually and graphically. For example, if an online shopper wants to purchase a pink Burberry cashmere scarf, then the product landing page should contain a product photo of a pink Burberry cashmere scarf. The product page’s title-tag content should contain those keywords as well as other on-the-page text.

Value – should I click on this link?

Why should web searchers click on your organic listing and not others? Why should web searchers click on your search engine ad? How have you encouraged them to click on your link? Is your HTML title-tag content compelling as well as your snippet and/or meta-tag description? Does your ad contain desired keyword content? Is the ad legible or difficult to understand due to keyword stuffing to accommodate all sorts of keyword combinations?

Did you make a video that pertains to the keyword phrase that your target audience types as query words? Does your target audience expect to see a video? Does your video contain a bunch of marketing hype just so your site can have search engine visibility on the first page of SERPs?

The scent of information has a lot to do with user expectations. If users want to see a video about a topic, they will probably use the word “video” or “videos” as a keyword, indicating transactional intent. If they do not expect to see a video listing, they might click on the video out of curiosity…or they might not. Usability professionals commonly perform expectancy tests to determine searcher mental models.

To make web content findable, the scent of information should be clearly established and consistently maintained throughout a website. But remember: you should try to understand the scent of information from the users’ perspective. Not the SEO perspective or the marketing department’s perspective. Not the CEO’s or the IT department’s (shudder) perspective. Keyword research tools do not give you a full and accurate picture of searcher mental models.

Talk to your users. Objectively observe their behavior. See how the scent of information exists (or does not exist) on your website. The answers might surprise you.

For those of you who are interested in reading detailed information about the scent of information, please read more about Peter Pirolli and information foraging theory.

So if the least common query type is a transactional query, how should web site owners and SEOs allocate time, resources, and budgets to other types of queries? What’s the “ideal” strategy to capture searchers attention regardless of the type of query they’re using?

Search engine reps have been talking about commercial intent for a number of years now. And commercial intent encompasses all three types of search engine queries: navigational, informational, and transactional. Because, realistically, most web searchers do not go from zero to “I have to buy that,” with a single query. Many web searchers try a variety keywords and keyword combinations before they make a purchase.

Commercial intent via informational and navigational queries

As many SEO professionals already know, many web searchers perform informational queries before making a purchase, especially if the desired product/service is a high-priced item. Here are some examples:

Suppose a web searcher wants to get health insurance for his family. He is not going to blindly take Google’s or Bing’s interpretation of “best” (i.e. the listing in the #1 position), but he might click on that first link to establish a frame of reference. Let’s be frank here—every health insurance company and affiliate will market their healthcare plan as being the “best.” The interpretation of “best” is entirely from the searcher’s point of view. Therefore, the searcher will probably do a variety of keyword searches to determine: (a) what types of health insurance plans are available and (b) the costs associated with each plan.

The keywords associated with these queries clearly show informational intent. When a web searcher wishes to see a list of items and associated features, search engines interpret user intent as informational.

Here is another example. Many online shoppers research their desired products/services online before making a purchase offline. I commonly observe this among users who wish to buy high-ticket items such as large piece of furniture or a car. These shoppers will not buy the sofa or the car without trying it out first, in person. But they research online to see: (a) what is available, (b) if their specific car/sofa/etc. is available, and (c) where they can buy it nearby.

Again, many of the keyword searches associated with these goals communicate informational intent. A question word such as “where” or “how” is a strong indication of informational intent.

Finally, in my last column, Optimizing for Re-Finding Search Behavior, I described a web searcher who looks up the phone number to her local pharmacy every 6 to 8 weeks in order to refill her diabetes-treatment supplies. Her intent (refill diabetes supplies) is clearly transactional. But her keywords are not. They are navigational keywords.

So, as you can see, commercial intent is not so easy to interpret from a list of the keyword phrases, even the most popular ones.

Search engines and commercial intent

Search engines have a tough job. Not only must they correctly identify searcher intent and deliver relevant search results, they must also associate specific keywords and keyword phrases with searcher intent.

Can search engines truly identify commercial searcher intent? I am not so sure they can. Commercial intent certainly encompasses the 3 known query types: navigational, informational, transactional. And many keywords can be reasonably assigned to navigational intent (such as a company name or known brand) or informational intent (question words).

However, some search results are accurate. Some search results are not. What do you think? Are search engines reasonably interpreting your commercial goals? Me, personally? I believe the search engines have a long way to go. Though log file analysis can and does yield useful data, I do not think it is enough to interpret commercial intent.

Here’s an example. One usability test participant likes to view funny videos. She wanted to show me a really funny cat video after she learned that I have a tabby cat. She did not remember the name of the video or the title of the video. But she did remember the search engine she used to discover the video (Yahoo! Video) and the keywords she used to discover the video. So she went to Yahoo! Video and performed a search. I watched her scroll through multiple pages of search results. She wasn’t using keywords to identify the correct search result. She was scanning the video thumbnails, trying to recognize the particular tabby cat in the video.

Clearly, this participant was exhibiting re-finding behavior. Re-finding involves content that has been located and/or viewed previously, and it is a more direct process than an initial discovery. Re-finding involves both recognition (is this the content I am trying to relocate?) and recall (where did I see this content before?)

Re-finding behavior is more common than many search engine optimization (SEO) professionals might imagine. Log files, keyword research tools and web analytics data rarely reveal specific re-finding keywords that direct observation, usability tests, and field interviews provide. All too often, I observe keywords that SEO professionals might believe show informational or transactional intent. Upon further observation, I see many of these keywords being used to re-find desired content. And re-finding, as a querying behavior, clearly indicates navigational intent.

Re-finding as navigational intent

As I mentioned in Don’t Forget SEO For Navigational Searches, people who perform navigational queries want to go to a specific web site or a specific web page. Re-finding queries are inherently navigational queries, because web searchers want to return to content that they have found previously.

Web searchers might have a difficult time remembering the exact keyword phrase used to find desired content due to work interruptions and/or the passage of time. Determining re-finding keywords is tricky because you cannot use web analytics data and keyword research tools to determine specific re-finding keywords. You have to talk to your users. Sometimes, refinding happens very quickly, within the same day. Sometimes, refinding happens months after the initial search query.

Here is an example:

One test participant gets prescriptions filled for a family member. She must buy diabetes syringes as well as insulin. But she doesn’t need to get insulin all of the time. She refills the insulin prescription only a few times a year, and diabetes syringes need to be replenished every 6 weeks or so. She looks up the pharmacy’s phone number on the web every time she needs a prescription refilled (Google is her current search engine of choice). Her keywords include the name of the pharmacy, and the city and state where the pharmacy is located. “I used to type in the name of the street in Google because it is so easy to remember: Main Street,” she said, “but Google keeps messing it up. I get the phone number right away if I don’t type in the street name.”

This searcher’s re-finding keywords are:

• Pharmacy name (which, by itself, would only lead to the main corporate site)
• City name
• State name

Now, if an SEO professional saw these keywords in web analytics data, he might likely interpret these keywords as showing informational intent. Additionally, since this web searcher only wanted the phone number and got her desired information on the first try, the “time spent on site” metric might be interpreted as poor, since she looked up the phone number and left the site immediately. In reality, this searcher is obviously happy with the search experience—she keeps using the same keywords every 6 weeks or so.

Repeat queries can be an indication of re-finding intent. But keep in mind that search engines regularly update search listings to include new and updated content; to remove spam and removed content; and to accommodate personalization and relevancy feedback, among other things. Ranking changes slow the re-finding process.

Re-finding via the commercial web search engines usually does not occur with daily tasks. It is more common with medium-frequency tasks, such as the one I mentioned above.

Tips for optimizing for refinding queries

As I mentioned previously, I commonly observe re-finding behavior in web search, image search, and video search. Here are some tips to remember to help web searchers re-find your site’s content:

• Since re-finding shows navigational intent, make sure your titles and URLs (web addresses) contain keywords. With navigational queries, the URL structure is very important. In fact, searchers who wish to re-find content often remember part of the URL structure.
• For image search, remember to use keywords in alternative text. For those of you who do not know what alternative text is, in (X)HTML, it is the text placed inside the image tag. If a graphic image does not appear on a browser screen, the alternative text appears in place of the graphic image. Although alternative text does not help a page rank higher for web search, it does have an impact on rankings for image search. Captions can also help.
• With video search, meta-tag content actually affects rankings in the video search engines. Ideally, use important keywords in the video title, meta-tags (keywords and description), and on the page where the video appears, whenever possible. If the website allows, make sure the video thumbnail also contains appropriate, keyword-focused alternative text.
• Finally, talk to your users. Ask them to show you how they re-find specific information on your website, and observe them objectively. And if you cannot be objective? (Hint: the people who develop the site are not objective.) Then hire a search usability professional who is. Believe it or not, many of your users will be more than happy to help.

If you would like more detailed information on this topic, please read the paperRefinding Is No tFinding by Robert Capra, Mary Pinney and Manuel A. Pérez-Quiñones from Virginia Tech University.

The primary persona is the group of people whose needs and expectations must be met at all times. Normally, for a website, there is only one primary persona. However, I believe that for search engine conferences, there are three primary personas:

• Decision makers
• Hands-on
• How-tos

The decision-maker persona

I did learn something new and interesting this year about search engine conference personas. For the most part, conference attendees do not want to hear what appears to be a pure sales pitch…or so I thought.

Attendees who have management positions often like to hear sales pitches and case studies. They do not always want to know the hows and the whys, the nuts-and-bolts of search engine optimization or search engine advertising. They want to know about companies or consultants who have successfully performed SEO and/or search engine advertising on sites in their industry — and the same optimization techniques are applicable to their sites.

Additionally, some attendees are in a position where they must convince decision makers that search engine marketing is a viable marketing strategy. And part of the marketing budget should be allocated to search. They need third-party case studies to support their cases.

This persona is very much one whose needs should be addressed.

The hands-on persona

This persona encompasses a wide variety of jobs:

• Website designers and developers
• Programmers and other IT staff
• Copywriters and content providers
• Advertisers

For example, many people who fit this persona want to know how to optimize a website. They will return to their company and do it themselves. They want to know the latest and the greatest coding, scripting, and programming strategies that are applicable to their types of websites. They want these search engine strategies to be scalable and as easy to implement as possible.

Copywriters, advertisers, and marketers also fit the “how-to” persona. They want to know how to write effective ads and website copy. They want to write effective calls to action. They want to monitor the effectiveness of their work via web analytics and other reporting software. I love addressing this type of persona because members not only want to know how to optimize websites and ads, they also want to know why to optimize websites and ads.

Personally, I find this persona to be the most detail oriented.

The how-to persona

On the surface, might seem that the hands-on persona and the how-to persona are the same persona. There is one crucial difference — people who fit “hands-on” persona want to actually apply the tip or strategy themselves. People who fit the “how-to” persona want to know how to implement a strategy but will not necessarily implement it themselves.

Here is an example. Many people in the marketing department must communicate with people in an IT department to do something, such as ensure 301 redirects are implemented appropriately. Duplicate content delivery is a common problem. Many IT professionals unwittingly deliver duplicate content to the search engines because, in their eyes, the content isn’t redundant. And it is often left up to marketing staff to explain the concept and give IT staff direction to address the issue. Therefore, people who fit the “how-to” persona need assistance in how to communicate some very technical concepts and strategies.

In addition, many companies do not host their own sites. These website owners want to know how to communicate effectively to the companies that do host their sites.

Finding the critical balance

Remember when you were a college or university freshman, and during freshman orientation, an orientation speaker said, “Look at the person sitting to your left. Look at the person sitting to your right. Only one of you will graduate,” ?

Well, I like to think of conference personas in a similar way. Look at the person sitting to your left. It might be a “hands-on” web developer, who wants to know an AJAX workaround and could care less about a case study. Look at the person to your right. That person might be a marketing manager who is having problems deciding whether or not to use Flash on a the site, and must report back to the Chief Marketing Officer.

All types of people with a wide variety of job responsibilities attend search engine conferences. And these appear to be the 3 personas, from my perspective. Personally, I wish all presenters and attendees would remember that. Do any of you see other types of personas?

Now that the commercial web search engines have data about searcher behavior for the past 10-15 years, I have modified my approach to SEO to include a 4th building block: 

• Text component: Using words and phrases your target audience types into search queries.
• Link component: Giving search engines user-friendly, easy access to those keyword phrases.
• Popularity component: The number and quality of links pointing to a website.
• Behavior component: Accommodating searcher characteristics, intents, and behaviors. 

In reality, the human element of SEO has always been an important part of the optimization process. Sometimes, searcher behavior is simple and sometimes it can be quite complex.

Sometimes searcher behavior is simple

Many search engine marketers and website owners base their entire search marketing strategy on the simplicity of search behavior. To them, the querying-to-purchase process is straightforward:

1. Searcher types keywords in a search box and clicks “Search.”
2. Clicks on first listing in search results page.
3. Lands on web page.
4. Adds to shopping cart.
5. Buys product or service.

See? Simple and straightforward. Believe it or not, sometimes searcher behavior is this straightforward. During field studies and usability tests, we often observe searchers click on the first or second search listing simply because it is there. Usability guru Jakob Nielsen calls this search behavior Google Gullibility. Unfortunately, some SEO professionals, brand marketers, and website owners alike rely on Google Gullibility as a major search marketing strategy.

Searchers do not necessarily click because they believe it is the best web page or website according to Google’s algorithm. Nor do they click because they want to make a purchase from the first website they encounter. Searchers often click on the first few search listings as a frame of reference for further research—to see what products and services are available, and to see what characteristics they should consider before purchasing a product or service. And sometimes, searchers do click on the first or second search listing because they performed a navigational query, and the search engine delivered the most appropriate listings at the top.

The simplicity of searcher behavior—searchers click because it is there. All of the fancy-schmancy eyetracking data, keyword analysis, click analysis, and statistics seem overdone and even useless when the answer is, “Duh! Searchers look and click because it is there.”

Nevertheless, seemingly simple behaviors are more complex than we might imagine. When we examine physical characteristics, usage patterns and the psychology of choice, all searcher behavior cannot be written off as simple Google Gullibility.

Sometimes searcher behavior is complex

When observing searcher behavior, you often see patterns emerge. For example, men scan search results pages differently than women do. The behaviors of advanced searchers are quite different from the behaviors of novice and intermediate searchers. Interestingly, many novice searchers honestly believe they are advanced searchers, and there is nothing a search usability professional can say to convince them otherwise. Mobile search behavior is different from laptop/desktop search behavior. The list goes on and on.

One reason searcher behavior is complex is that there is no single behavior associated with searching. In fact, the word “search” has come to mean querying behavior only. In reality, as I mentioned in Understanding Search Usability, “search” consists of a wide variety of behaviors. I like to think of search as the following equation:

Search = Browsing + Retrieval

As I also mentioned in the previous article, to understand search usability, I believe one must fully comprehend the concept of berrypicking (See Marcia Bates’ The Design of Browsing and Berrypicking Techniques for the Online Search Interface). In berrypicking, searcher behaviors are not static. Searchers use a wide variety of behaviors to look for desired content, and these behaviors evolve with each perceived failure and success.

Gender differences, language and age differences, usage behavior, searcher goals and how they express these goals with keywords—all of these items are a part of evaluating how people search and why people search. SEO professionals do not want pages simply to rank well. They want the right web page(s) to appear at the right time for the right searcher goal. Is the searcher ready to buy? Then the SEO professionals should optimize a product or category page to appear at the top of search results or in a search ad. Does the searcher want to go to a specific page on a website? Then the SEO professional optimizes a web page to appear for that specific navigational query.

Sound complex? It is complex. Many advanced SEO professionals study the various aspects of searcher behavior.

I have always found it odd when some of the simplest human behaviors are quite complex, and when seemingly complex behaviors are quite simple. Search behavior is no exception. The human element of SEO has always been an important part of the optimization process. Now, I have added it to my building blocks of successful SEO. Have you?