Today, the good folks at SEOmoz published an extensive article documenting their statistical findings on web site ranking factors, as gathered from the data in their LinkScape analysis tool. It’s a good article, and demonstrates some interesting results they’ve garnered from the data available in the extensive LinkScape database.

One of their major takeaways in the article was a little disturbing to me.

“Alt attributes of images are probably pretty important places to use your keywords[.]” Explaining (Some of) Google’s Algorithm with Pretty Charts & Math Stuff, October 22nd, 2009

I’m not in anyway disputing their results; their data indicates that placing keywords in alt attributes is of benefit to search engine rankings. Whether that’s true or not is irrelevant to me; I simply want to discuss how this information should be best used.

SEOmoz, of course, is a company dedicated to the study and practice of search engine optimization and marketing. Their goal is to learn what they need to know in order to best put into practice the promotion of web sites. That’s great. My goal, however, is to make sure that users with disabilities are able to use and access web sites successfully without having to jump through unnecessary or unhelpful hoops along their way.

This is a particular case where the SEO method must be used cautiously and selectively if at all. What I want to convey in this article is the fact that while using keywords in alt attributes may help your web site rank, it can also result in a significantly less accessible web site, if applied poorly.

What’s the problem with alt attributes?

While sighted users will never even be aware of an alt attribute value in normal web browsing, screen reader users depend on them. Excess verbiage can render an image-based menu unusable, as I observed in a recent site review at Practical eCommerce. The same unnecessary use of keyword terminology in contextual images can easily confuse or distract a user; and the use of keywords with spacer or ornamental images can cause a web site to be completely unnavigable.

It’s all a question of information overload: practically speaking, if a web site uses images to convey information, a screen reader user can’t disable them without rendering the web site unusable. If the site also fills other images with extra text, the same user may be overwhelmed by an unnecessary volume of keyword phrases.

The SEOmoz report does continue to remark that “Keyword stuffing may be holding you back,” and the overuse of keywords in alt attributes can certainly qualify as keyword stuffing.

You shouldn’t take away from this article that using a keyword in an image alt attribute is totally unacceptable. That’s really not the case: just be selective. I wouldn’t condemn you for using the text “About ProductName” instead of “About” for a navigational image, or using a sensible alt attribute for a contextual image, such as “Woman using our ProductName.” Just remember that keyword stuffing is keyword stuffing, wherever you put the words.

And never place any value in the alt attribute for a purely decorational or spacing image. Please. Just an empty attribute.

Best practices: keywords in alt attributes
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

Shari Thurow, a well-respected researcher in the search engine optimization and usability realm, suggested that separating the two concepts is, in fact, ridiculous.

Well, that may be. However, I think that it’s crucial to break a task into parts if you want to gain a thorough understanding of the whole. Search engine marketing is an excellent example of a whole which is greater than the sum of it’s parts.

As I see it, building a search engine friendly site is one of the first stages of best practice search marketing. The adage “if you build it, they will come” fails to hold, however: a site which is constructed merely to be search engine friendly will gain little to no traffic.

Being part of the process

Being search engine friendly is a part of the process of search engine optimization; which is, itself, a part of the process of search engine marketing. In addition to these two aspects, search engine marketing may also include pay-per-click advertising, print advertising, link building and social media participation. Search engine marketing is a large area, and very, very few people are expert in all aspects. I’m certainly not.

From a marketing standpoint, what parts of this marketing whole are necessary for your business to succeed is going to vary radically depending on your industry and the way your business intersects with the internet. It will also depend on your definition of success. If you’re looking to maximize growth, you’ll probably want to be investing in all aspects of marketing.

So I’m arguing that search marketing, while clearly a practice in which the parts of the whole are highly interwoven and carry clear dependencies on each other, can nonetheless be separated into it’s component parts for a variety of reasons, including for the sake of discussion.

Now let me take this a step further. Not only is it possible to separate search engine marketing into separate aspects for discussion, it’s valuable.

If you want to understand the interactions between the different aspects of a task, it’s important to have some information about all parts. In this context, it’s necessary to treat the whole of search engine marketing in a given discussion. However, when you want to understand the details of a specific task, it’s important to stay focused on your part of that task.

It’s necessary for practitioners in search engine marketing to know, in general, what the impact their work will be on all aspects of the marketing campaign. It is crucial for practitioners in search marketing to know, in detail, exactly how to perform their own tasks in the best possible manner for their clients. It’s important to treat an area of expertise specifically. Talking through the nature of that area; comparing and contrasting it to other related areas; considering the specific nature of tasks within that area of expertise: these are all ways of better defining and refining knowledge on a specific subject.

Why does this matter?

It doesn’t, really. It’s all semantics. Search engine optimization is the commonly known term, and it frequently is understood to encapsulate search engine marketing. Or the other way around. The industries around search engines and marketing (and just about anything internet) are young, and the vocabularies aren’t really all the firmly established. As a result, some people have a very firm opinion of what a given term means which may not always coincide with others definitions.

Well, that’s why we write about it. We’re all hoping that our definitions will ultimately win.

Taking a holistic view of SEO in parts.
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

At their hearts, web accessibility and SEO are focused on optimizing different aspects of your web site: accessibility cares almost exclusively about the disabled user and their experience whereas SEO is focused firmly on your bottom line and your experience, as site owner, in the online aspects of running your business.

That isn’t to say that SEO (when done with best practices in mind) isn’t concerned with the user experience, but the bottom line for SEO is the marketing performance of the site. If it’s necessary to sacrifice some aspect of usability or accessibility for a measurable gain in the commercial performance of your site, the logical decision for search engine optimization is to make that sacrifice.

For the most part, SEO and web accessibility are very compatible services — the ways in which they differ are truly slight. From a technical aspect, all the same groundwork needs to be in place. However, when you begin to get into the finer details you can begin to see the subtle differences between optimizing for accessibility and optimizing for search engine success.

Site Maps

It’s subtle, but optimization is inherently concerned with subtlety. From an accessibility viewpoint, the best place for a link to your site map is at the top of the page, coupled with skip links. It should be visible and easy to locate, providing immediate access to your web site map.

From an SEO perspective, the site map — while very important — shouldn’t usually be given such key real estate on the page. For the vast majority of users, the site map is something they’re looking for as a purely secondary navigation method. Many disabled users, however, use it as a primary navigation method.

Alt Attributes

The whole debate concerning keyword in alt attributes has been thoroughly hashed over, so I’m not going to revisit the subject at length here. The fundamental difference is very transparent: SEO indicates that keywords in alt attributes (where relevant and meaningful) can be helpful to search engine rankings. Accessibility professionals are more likely to consider these descriptive alt attributes to be noise.

Although it may seem appropriate to describe every image in the site, it’s highly questionable whether this practice actually adds value, however well-composed and meaningful the attribute value actually is.

Headings and Titles

The broad outlines of what accessibility and SEO are looking for in headings and titles are very similar. They’re both looking for something which is immediately clear for the user and which explicitly describes the purpose of the page or section it begins. When it comes to detailed refinement, however, they’re likely to go in subtly different directions.

Accessibility practitioners want conciseness: the shortest unique label which can clearly convey the necessary information concerning meaning and context. This is simply an effort to minimize the noise factor for screen reader users and the learning disabled: reduce the time required to comprehend context and minimize the potential for confusion.

Search engine optimizers are less concerned with being concise and more interested with ensuring that the key terms and phrases are used in prominent contexts. Although it’s not necessarily true, it’s not uncommon for this to result in longer, more complex statements.

Conclusion

One of the best parts of the work I do is collaborating with search engine marketers. Hashing out these differences of opinion can be very productive. It’s a great way of hearing alternate perspectives and allowing both parties to gain a thorough understanding of the goals inherent to their processes. The differences between web accessibility and SEO (when done according to best practices, etc., etc.) are primarily a matter of intent. I don’t win every argument, and I can accept that: ultimately, the subtlety of these differences is such that I don’t really need to have things my way every time.

In the end, every site is a matter of compromises: meeting the line between what the owner wants and what each consultant or development firm involved in the project believes is best for the project.

Web Accessibility is not SEO
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

Here are 8 speedy checkups (in no particular order) which you can easily perform on your site to inspect it for problems. No methods suggested require special knowledge of HTML or web programming. Excluding acquiring and installing software, these tasks shouldn’t take more than a few minutes for most sites.

That doesn’t include fixing any problems found, of course…

1. Check Headings
2. Check for Image Alternatives
3. Behavior without Javascript
4. Unique Page Titles
5. Check for Broken Links
6. URL Canonicalization
7. Robots Control
8. Download Times

1. Check for Headings

The use of appropriately coded HTML headings is important for both SEO and Accessibility. Headings should be used to define the content structure of the page, semantically demonstrating the importance of that text and providing additional in-page navigation tools for screen reader users.

Software required: Firefox Web Browser, FANGS Extension for Firefox.

What to do: After installing Fangs, navigate to the Tools menu in Firefox and select Fangs. Flip to the “Headings List” tab in Fangs. The output displayed in this list shows all of the headings on the current page and their heading levels. Sometimes you’ll see output with a number but no text — these are most likely headings which are using images within the heading element.

Which leads to the next point…

2. Check for Image Alternative Text

Providing alternative text attributes for your images is critical to accessibility. The logic behind how to provide alt attributes appropriately can be complicated, but the presence of the attribute itself is always necessary.

The quickest check for this problem is very simple: disable images in your browser.

Software required: Any browser.

What to do:

• Firefox: Tools » Options » Content » "Load Images Automatically" -- Unchecked
• Internet Explorer: Tools » Internet Options » Advanced » Scroll to "Multimedia" » "Show pictures" -- Unchecked
• Opera: Tools » Preferences » Web Pages » "Images" -- Select "No Images"
• Safari 3 for Windows: Although you can disable images, the browser does not automatically display alt attributes in their place.

The results will vary somewhat from browser to browser, but in general they’re fairly consistent. If an alt attribute is present and contains text, that text will be displayed in place of the image. Look at the text and decide whether it’s appropriate! If there is no alt attribute text, the image will be left as a blank space. If there is no alt attribute at all, however, the image will instead be replaced in some cases by the file name of the image, in others by a simple broken image icon. If the navigation button for “Purchase my Widgets” suddenly says “Nav_image_10x10_30.gif” then that’s really not useful.

3. Behavior without Javascript

The definitive key concern about Javascript is that search spiders browse the web without it. For accessibility, the issue is significantly more complex, but there can be no question that making sure that your site is usable without Javascript is a major concern for both concerns.

As with images, the simple test is to disable Javascript in your browser.

Software required: Any browser.

What to do:

• Firefox: Tools » Options » Content » "Enable Javascript" -- Unchecked
• Internet Explorer: Tools » Internet Options » Security » Scroll to "Scripting" » "Active Scripting" -- Disable
• Opera: Tools » Quick Preferences » "Enable Javascript" -- Unchecked"
• Safari 3 for Windows: Edit » Preferences » Security » "Enable Javascript" -- Unchecked

If there are problems, they should become immediately apparent. I’ve seen occasions where the entire content area vanished or the entire navigation menu suddenly disappeared. It may be more subtle, but the simple test is to look for key content which is no longer available to your browser.

4. Page Titles

A unique page title for every page of your site is a crucial part of best practices for both accessibility and SEO. It helps orient your visitors, and allows searchers to know what they’re coming to.

Test Page Titles (Simple version)

Software required: Any browser.

What to do: Look at the top of your browser window. Does it say something like “Untitled Document” or just read the URL of your site? You’ve probably got a problem. In order to really get anywhere with this, you need to check most pages of your site, so for larger sites, you may want a more thorough method.

Test Page Titles (Thorough version)

Software required: Xenu Link Sleuth or similar site spidering tool.

What to do: Install the software and run it. Navigate to the File menu and select “Check URL.” Enter the URL of your web site. Select “OK.” Wait. The tool will now spider your web site and generate a list of linked documents on your site. This will include all images, external links, etc., by default, although it can be configured differently.

One of the columns in the results is “Title.” If you sort the results by type (select the header of the column “Type”) such that text/html filters to the top, you’ll have a list of pages on your site. Simply scan down the “Title” column. Are they all the same, or are large numbers of pages the same? That’s a problem. Every page should have a unique title.

5. Broken Links

Don’t close Xenu yet — because it’ll also check for broken links. Re-sort the output by “Status.” Find the section with the status of “not found.” Any file which came up as not found is, simply stated, a broken link. Broken links can have a lot of consequences for your website — loss of trust, frustration for your visitors, or even causing an inability for visitors to get key information.

6. URL Canonicalization

URL Canonicalization is a big, fancy word which simply means that all possible variations on a page’s URL lead ultimately to the same URL.

The issue is simply that all of these URLs can be exactly the same page:

• http://www.yourdomain.com/index.php
• http://www.yourdomain.com/
• http://www.yourdomain.com/index.php?t=success&r=fail
• http://www.yourdomain.com/index.php?r=fail&t=success
• http://yourdomain.com/index.php
• http://yourdomain.com/
• http://yourdomain.com/index.php?t=success&r=fail
• http://yourdomain.com/index.php?r=fail&t=success

And it’s entirely probable there could be more. Now, checking for all variations can be a somewhat tedious task. However, the one big variation which is likely to apply to every page of your site is the “www” or no “www” conundrum. It really doesn’t matter which you choose as your preferred site — but you should absolutely choose just one.

Testing the behavior of your site with this problem is absurdly easy.

Software required: Any browser.

What to do: Go to your website. Place your cursor in the address bar of your browser and change the URL. If your site is currently displaying “http://www.yourdomain.com,” remove the “www.”. If it’s currently displaying “http://yourdomain.com,” add it in.

There are three basic possibilities for what will happen at this point:

1. The site will automatically redirect to the alternate version. If you change the address, hit return, and the URL immediately switches back to what it was before, then this is what has happened. This is the correct behavior, although from a technical standpoint you’ll also want to check that the right type of redirect has happened. (You want a 301 redirect.)
2. Nothing will happen. The site will reload identically with the new URL. This isn’t a total disaster, but it does mean that your site isn’t canonicalized — both versions of your URLs are available. You should investigate implementing URL canonicalization.
3. You’ll get a file not found or server error. This is bad. This may mean that your hosting service is not configured to send both versions of your URL to your account, although this isn’t the only explanation. You can’t ever assume that people will link to the correct version of your URL or check their links, so it’s very important that every possible variation will eventually arrive at the right destination!

7. Robots Control

Most of the time, robots control isn’t a critical part of your site. The various directives, either in meta elements on your web site pages or in a robots.txt file on your site are usually intended to prevent specific information from being indexed by search engine spiders. If it’s missing entirely, it’s not really the end of the world.

However, if you have a robots.txt problem, you have a serious problem. If you’ve accidentally set your robots.txt file or meta elements to block robots from crawling or indexing your pages, then you’re simply dead in the water.

This is somewhat of a multiple step process:

1. Enter your home page URL in your browser address bar like this: http://www.yourdomain.com/robots.txt. If you receive an error message, than the file simply isn’t there. You may wish to add one — to reduce the number of error messages in your server logs, if nothing else, but you can be pretty certain that you aren’t blocking search spiders with it.
2. If you do have a robots.txt file, you’ll see it displayed in your browser. It’s a very simple file format, consisting primarily of three types of statement: User-Agent, Disallow and Allow. By default, everything is allowed; so most robots.txt files only contain User-Agent and Disallow statements. This is the Disallow statement you really don’t want to see:

   User-Agent: *
   Disallow: /
   

   In simple terms, this says “No search spiders are allowed to access this site.” The single forward slash means that the statement refers to all directories on the site. The * means all user agents — which includes Google’s search spider, Yahoo’s search spider, and so on.

   A more common file will be something like this:

   User-Agent: *
   Disallow: /private/
   Disallow: /images/
   Disallow: /admin/
   

   This is more specifically disallowing three folders which the web site owner does not want available to the public.

   If you want to check specific URLs, or find that your robots.txt is quite complicated (they can be!) Google offers robots.txt checking tools in their webmaster tools. (Requires a Google account.)

That’s taken care of your robots.txt file; but what about more specific directions? Meta elements can provide instructions to robots as well, which operate on a page specific basis. These are much harder to check, since it requires a page by page check — and, frankly, there may be specific pages which you don’t want indexed. Regardless, this is what to look for:

Or variations on that theme. The code states that, for the current page, spiders should not index the information and should not follow links from the page. If you find this on a page which you’re having trouble getting in the search index: that’s the reason!

8. Check your Web site’s Download Time

The time it takes your site to download can have a huge impact on the likelihood a visitor will stick around. Although broadband is popular, it’s still just barely a majority of the internet market. Assuming that visitors will wait around to see that high-resolution image of your product is simply foolish.

It’s easy to check approximately how long it will take your site to load at a variety of download speeds. The actual download times for a given user will be hugely variable, since there are many more factors than raw bandwidth at play when accessing a web site — but this quick test will at least give you an idea.

Visit the web site optimization analyzer. This tool will give you the total anticipated download time at a variety of bandwidths, and also breaks down, file by file, the total size of the elements of your web page.

Even if you’re satisfied with the time required to download your site (8 seconds at 56k used to be the benchmark; although you need to make your own decision depending on your site’s needs,) you can use this information to streamline your site. Optimizing graphics and only loading scripts on pages which actually use them can make a huge difference.

Conclusion

Touching on these issues won’t clear up every problem you could have: but they can save you a lot of headaches. Fixing these issues will generally improve the overall condition of your web site in a variety of ways. You’ll end up in a better position to work on SEO and you’ll offer your visitors a more accessible web site.

It can’t hurt to check!

Web site Tune-up: 8 Quick Checkups
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

“Sitelinks” are additional links Google generates from the contents of a site in order to help users navigate your site — they provide these links in their search results for selected terms. Most sites don’t have site links, so I’m finding it pretty cool to notice them for myself!

If you can’t see the image, the current Sitelinks for this site are:

• Gallery
• Blog Home
• What is Web Accessibility
• SEO and Accessibility
• Articles

On the whole, I’m pretty pleased with the selection chosen here. They’ve pretty well pinned down the key areas of the site: web design, web accessibility, search marketing, and my writing on these topics. Everything is reasonably represented. Perhaps, in my forthcoming site redesign (don’t keep a look out; it’s not going to be that soon,) I’ll make a point to better promote these specific areas of the site.

Google Site Links for “Joe Dolson”
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

So, naturally, a market is born. There’s no questioning the correlation between links as a factor to judge the value of a web site and the existence of a market to sell those links. Obviously, the immediate gut reaction is something along the lines of:

“But that’s like…prostituting your site!”

Well, that’s perhaps a bit strong. After all, when it all comes right down to it, it’s just another form of advertising. However, unlike most traditional advertising, it’s a method of advertising which has a direct and measurable impact on the success of your business.

Thus you have a conflict of interest between search engines, who want to judge your web site on the basis of the quality of it’s content while using the interest of the general public as a measuring tool, and business owners who would like inflate the quality measurement by exercising the dimensions of their wallets.

It’s a complicated difference of opinion: advertisers and publishers see no reason they shouldn’t be able to purchase advertisements of whatever form they choose from whatever site is willing to sell them. However, if search engines prefer “pure” results, then they will pursue whatever means they can to negate the value of links they identify as purchased.

But how do I know when I’m purchasing a link?

To a high-level search marketer or professional web developer, what constitutes a link buy may seem obvious. Search engines have clearly targeted the underground market of reciprocal link sharing, link brokering, and text link advertising. They very specifically want to target links or advertisements which are capable of passing ranking value.

The average web user or beginner web publisher, however, may be less certain when that line is being crossed.

Purchased Link Scenarios

Scenario #1

A friend and business owner offers you a free meal at his restaurant if you’ll write up a review on your food blog. Your friend doesn’t specify any requirements for the review: positive, negative, whatever. You write it up, adding a link to the website for the restaurant, and go on with your life. You didn’t mention that the meal was free, because you didn’t want to give your readers the impression that maybe you were biased, even though you wrote a very fair review. You also didn’t mention that the restaurant owner is a friend of yours, for the same reason.

Did your friend buy that link from you?

Google’s official stance on paid links states:

Not all paid links violate our guidelines. Buying and selling links is a normal part of the economy of the web when done for advertising purposes, and not for manipulation of search results. Links purchased for advertising should be designated as such. This can be done in several ways, such as:

• Adding a rel=”nofollow” attribute to the a tag
• Redirecting the links to an intermediate page that is blocked from search engines with a robots.txt file

Why should I report paid links to Google?

You didn’t mark the link as “nofollow,” nor did you redirect the link. After all, you’re just an average blog author — you didn’t even know that Google had guidelines on this. On the other hand, you weren’t exactly paid to provide the review…but seriously, that’s just arguing semantics.

From Google’s perspective, yes, that’s probably a paid link — although it’s entirely likely that none of the parties involved knew it. On the plus side, it’s pretty much impossible for Google to detect any exchange like this. There’s no online trace for a personal interaction like this which would throw up any suspicious signals.

Scenario #2

You’re an independent consultant in the construction industry. You become aware of an organization you are eligible to join which maintains an online list of members and provides links to their web sites. It costs $50/year to join, and you’re really not very interested in the organization overall, but you figure that $50/year for a link to your site is pretty worthwhile — and perhaps the organization will come in handy at some point.

Did you buy that link?

Well, on the one hand, you just joined an organization pretty much specifically to gain a link to your site. Yes, you purchased that link. On the other hand, what are these organizations for? Many organizations exist partly to promote their areas of industry and their members — so you paid a fee to the organization, and they are promoting your interests. That’s a legitimate business relationship. But is promotion of members via a link equivalent to having members purchase a link? Hardly. It seems ludicrous to expect every professional organization to add “nofollow” to their membership listings, after all!

Scenario #3

You’re looking to promote your site, and you get in touch with somebody who tells you that, for $500, they can get you 15 really great links to your website. You don’t ask any questions, pay the money, and later find a handful of new links to your site.

Did you buy those links? Oh, hell yeah. You so bought those links. Are they trackable? Maybe, maybe not. Depends on the link broker (since that’s basically who you’ve just gone through.) The big question you should be asking right now is “Why didn’t I ask any questions?”

If you just pay some money and end up with some links, you should assume that you could be crossing the line. Basically, if you don’t know the process which led to your having these new links, you probably shouldn’t have purchased them. You’re simply running risks that you can’t readily quantify.

Scenario #4

You’ve just paid $299 for a review by the Yahoo! directory. The deal is pretty simple: you pay them $299, they’ll review your site within 7 days — and if they like it, they’ll add it to the directory. Oh — only for a year. You’ll have to pay them every year if you want it to stick around.

You’re not actually buying a link, in this case: you’re buying an opportunity. They may or may not include your site, so really you’re taking a risk, gambling that your site is what they want. You’re safe, from a search engine perspective.

But…

You may not be safe if you take this same logic and apply it to just any directory on the web. The Yahoo! directory is big, old, very public, and (of course) run by company that also happens to own a search engine.

Even if a directory actually states that you’re paying for a review and that inclusion is not guaranteed, you should probably take a gander at the actual contents of the directory. If you see a generally low quality directory, filled with sites out of their appropriate categories, with misleading descriptions, or with very poor quality content, it’s entirely possible that what they actually mean is “we’re trying to make sure that people keep submitting to our totally useless directory by claiming that we use some kind of quality control.”

And, finally, if a directory does NOT state that you’re only paying for a review or guarantees inclusion, just run away. You don’t want to be there.

In Summary

Is there really an answer to the question “what is link buying”? It depends on context, intent, and detectability. It depends on who’s looking: a search engine or a human. Link purchases can encompass a broad variety of real and legitimate business arrangements. What should you be particularly nervous about? Any service which:

1. Doesn’t allow you to control what links you purchase/sell.
2. Will sell/buy links on a bulk basis (“10,000 links IMMEDIATELY!”)
3. Requires you to disclose your business relationship with them.1
4. Asks you not to disclose your business relationship with them.1
5. Will not disclose any information about their methods.

Differentiating between purchased and legitimate links, in most cases, actually comes down to intent. At some level or another, almost every link to any commercial site on the web is likely to have some kind of business exchange behind it. The question is more on the nature of that exchange. But this is, of course, something which is 100% untraceable by any search engine. They’ll never know why you chose a particularly link.

Unless, that is, you’ve used some kind of system which leaves a trace. This, of course, is a reason to be careful when picking who you’ll work with to help build links to your site: be sure that you trust them not to waste your money or damage your site’s success.

Notes:
1. This is a complicated point: it’s not the disclosure itself which is negative — it’s the fact that it’s required. You need to think that if disclosure is required, there may be something questionable about the activity itself. However, if disclosure is expressly forbidden, then you’re really in trouble! Return to text

Buying Links: What is it?
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

Now, I don’t normally go out of my way to advertise for companies on this blog. However, there’s an extra little piece of incentive in this case which makes me a bit more inclined to push this. Specifically, for every subscriber using coupon code HOLIDAY67, Search Marketing Standard will be donating $1 to the charity Toys for Tots.

The promotion is good for two weeks – it’ll end on Monday, December 10th. Just enter the coupon code HOLIDAY67 during registration to get your subscription.

And a note to those who keep track of these things: Search Marketing Standard is a good publication. They focus on analytics-based, meaningful issues in Search Marketing. This isn’t a publication where you’re going to see anybody recommending cloaking, keyword-stuffing, or any of those many highly questionable practices in the SEO world. It’s not about accessibility, but this publication discusses methods which somebody with a commitment to standards-based, accessible web development can faithfully use to make serious progress in the search marketing arena — without breaking their other commitments.

Subscribe to Search Marketing Standard

Get your subscription to Search Marketing Standard!
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

The meta description doesn’t just have to be in sentence format; it’s also a great place to include structured data about the page. For example, news or blog postings can list the author, date of publication, or byline information. This can give potential visitors very relevant information that might not be displayed in the snippet otherwise. Similarly, product pages might have the key bits of information — price, age, manufacturer — scattered throughout a page, making it unlikely that a snippet will capture all of this information. Raj Krishnan

While this technique may not work at all well for purely information based content (this article, for example,) it seems like a fantastic way of demarcating your data for product-oriented pages. Not only does it create a very clear search snippet, you need to consider the possibility that Google may be able to better clarify or identify search results on the basis of tagged data. I don’t mean that your rankings will improve; Google has long been very clear on the point that meta descriptions are not used in establishing rankings. They are, however, used to establish relevance.

The article does not say that Google parses out this structured data and uses it. However, it is entirely likely and possible that they either use or intend to use structured data in such a manner.

Regardless, for products which are well represented using short blocks of structured data, such as books, movies or sound recordings, this can be an excellent way of quickly conveying all of the most critical data about the product. From a programming perspective, as well, this can provide an extremely easy and logical way of building unique meta descriptions for a large database of products without needing to write each one individually. If it’s effective, it could mean a significant savings of time and expense at this level of e-commerce development.

Of course, for meta descriptions, the only truly meaningful metric is whether they encourage qualified visitors to clickthrough to your site. Since descriptions aren’t used for ranking, only actual testing can confirm one way or the other how well a given snippet helps your site perform. I couldn’t find any examples of tagged meta descriptions in use on the various product searches I tried; I’d be interested to know if anybody is actually using this idea.

Does it help? No idea. Still, I think I’ll probably need to give it a trial to see what it can accomplish.

Tagging your meta descriptions
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

I really loathe this.

If it was an amazing, perfect, incredible search optimization technique which would bring absolutely fantastic traffic I still wouldn’t recommend the technique. Appropriate alt attributes are one of the most critical areas for the user experience of screen reader users — using them inappropriately is a great way to give this section of your market a horrible experience on your site.

To be perfectly clear — most advocates of use of the alt attribute for image optimization do not advocate keyword stuffing or any other serious abuse of the alt attribute. However, to me, any recommended use of the attribute which does not place the user first is problematic.

Bluntly, if it’s really important that an image is described, then you should describe it in plain text, written out in a presentable and understandable manner. In the site content — not buried in the code.

I’m a firm believer that most images are purely decorative. They provide little to no information to a screen reader user, and the best choice is to leave the alt attribute empty. This means that the user agent can completely ignore the image — they can identify that there is an image present, and will also be aware that it does not need to be dealt with. (WebAIM: Designing for screen reader compatibility.) An absent alt attribute is dealt with differently — the user agent will attempt to identify the image via the src attribute:

How the JAWS screen reader announces image information

The behaviours detailed apply to JAWS version 6.2, 7.0, 7.1 and 8.0.

default behaviour for image with an alt attribute containing text

Announces presence of the image and the alt text

default behaviour for null alt text alt=""

Does not announce the presence of the image

default behaviour for an img without an alt attribute

Does not announce the presence of the image

default behaviour for image with an alt attribute containing text in a link

Announces presence of the image and the alt text

default behaviour for null alt text in a link alt=""

Does not announce the presence of the image

default behaviour for an img without an alt attribute in a link

Announces the presence of the image and the value of the src attribute

Steve Faulkner (Investigating the proposed alt attribute recommendations in HTML 5)

Now, the src attribute can be useful in certain situations – a well-named image file can be quite clear from this perspective. However, leaving the alt attribute blank will provide the best experience for insignificant images.

Significant images, on the other hand, are a completely different story. They do say that an image is worth 1,000 words…is it possible to do realistic justice to any significant image in an alt attribute?

My general take is that there are three types of images on the web:

• Content-bearing text images,
• Content-bearing non-text images,
• and decorative images

The first is the only category which should have content in the alt attribute — and that text should be an exact duplicate of the text portrayed in the image. The second category should have additional descriptive information provided in the text, or a clear and obvious link provided to a more thorough description. The third should be left with an empty alt attribute. This is not significant information, and it should remain the user’s prerogative to enable the screen reader option which will provide them with additional image information.

The question with images and accessibility isn’t precisely a debate between whether images are content or not — instead, it’s the question whether images are useful content to those users with visual disabilities. The question you need to ask yourself is whether an alt attribute on the image will provide useful information to a user. Is it helpful to mark an image as “my cat” or “landscape of mountains with a flower in the foreground,” or does it simply add noise to the content?

This isn’t to say, however, that there aren’t perfectly reasonable ways to optimize for image search which won’t also interfere with the user experience with your website. Liana Evans recently published an article discussing image optimization at Search Engine Guide.

She does suggest using the alt attribute, but does at least emphasize moderation and specifically mentions accessibility as an issue to consider. The other four suggestions are all very sensible and simple ways to ensure that your images are available and easy to find. Li doesn’t mention the content surrounding the images in this article, although she does mention it in a Cre8asiteForums thread on image optimization. In this thread, Li specifically mentions that alt attribute use for image optimization is only a very slight influence.

On the whole, accessibility should always trump marketing. However, there really isn’t a huge divide between the two. In any circumstances I can imagine, there’s no search marketing benefit to choosing a less-accessible solution. Perhaps you’ll be less able to take advantage of some search engine loophole. This isn’t a serious loss for most businesses — there’s no guarantee that the same loophole will still be around next month, anyhow.

Further resources:

It’s important to note that these resources are intended to provide you with a wide variety of viewpoints on the subject. I think these articles are all worth reading; I don’t necessarily agree with them on every point.

Alt Attribute Use

• Mike Cherim: Alt and Accessibility
• Dive Into Accessibility: Providing Text Equivalents for Images
• 456 Berea Street: The Alt and Title Attributes
• Simon Willison, Guild of Accessible Web Designers: Writing Good Alt Text

Image Optimization

• Chris Silver Smith: Need more traffic? Try Image search optimization
• Grant Crowell: Optimizing Images for Search Engines
• Chris Silver Smith: Using Images for Local SEO
• Brian Vuyk: Using Google Image Search to Drive Traffic to Your Site

Search Optimization, Accessibility, and Images: Best Practices
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design

A key element to SEO is the concept of the “perfect search result.” For my purposes, this is defined as the search result which:

1. Provides access to the most relevant information pertaining to the search.
2. Appears as the first or only result of its media type to a search.
3. Satisfies the searcher’s needs for the search they performed.

There is hardly such a thing as a perfect search result. As the explicitness or uniqueness of the search increases, the perfection of the results increases — but the challenge of constructing a search which will return perfect results is pretty significant. Once you accept that, the whole world of search optimization becomes much more clear. The goal of search optimization is to make your site the most perfect result for the broadest range of relevant searches. It’s a fuzzy definition, I know — but it works for me.

The perfect result is entirely in the eye of the searcher – for the vast majority of searches, there is no guaranteed correct answer. It is the task of anybody attempting search engine optimization to attempt to tune a site until it will appear effectively for some set of desirable and probable search phrases.

Most levels of search engine optimization revolve around topical clarification. Optimization is the act of studying the website and taking appropriate steps to make it more clearly apparent what the site is about. Clarifying the subject and topical areas of a website is of benefit to both site visitors and to search engines.

What do relevant links provide? Subject clarification. What do effective page titles provide? Page-specific clarity. What do headings do? Specify the topic of a given segment of a page. These are the basic, basic steps which need to be taken in order to help a search engine provide better results.

Search engine optimization, when performed correctly, is not at all about trying to trick a search engine into inflating the value of your site. It’s about trying to inform the search engine about your site so that the search engine can provide better results.

Now, of course, in the eye of the searcher, your site may not actually be the best possible result. This is inevitable. No site can provide the perfect information to every potential visitor. It comes back to the search terms: searchers do not always provide explicit enough search phrases to actually find their own best answer.

This is why the range of search terms should be as broad as reasonable: because searchers don’t always know what they’re searching for. As an example, if your site makes use of your own custom industry terminology for “widgets”, describing them throughout your site as “gidgetized pseudo-elemental devices,” you’ll potentially have the only results for your custom terminology. You’ll always be first, and anybody who searches for your term will get to your site.

Of course, most users will search for “widgets,” and visit your competitors. Those who do wind up at your site are possibly just those people who ran across your phrase and wanted to know what it meant!

So, you can see that there’s a paradox between having a perfect search result and actually receiving traffic: it’s easy to accomplish a unique search result or top rankings only if you accept that the specific term for which you rank is very uncommon.

Helping search engines provide better results
Copyright 2004 – 2010 Joseph C Dolson, Accessible Web Design