Search Engine Optimization

September 6, 2007

A lot has been written about Search Engine Optimization (SEO) and it can be confusing and contradictory. Without going into great depth about costly models (pay for results placement, or per click, or hiring SEO consultants) versus the free methods, I thought I’d offer some basic free steps for improving your position in search engine results. For clients wanting to improve results for free, I take the simple steps described below. These have been working well for me and my clients. Note that the very best thing that you can do is to use good page titles and to write rich content for your Web pages. (And promote your site; don’t wait to be found.)

Here’s a brief intro to:

META tags (easy to do, but of limited use with today’s search engines)
Using meaningful titles for all your Web pages
Using descriptive text with image tags
Using a file with instructions for search engine robots (robots.txt)
Sending Google info about your site’s structure and how often it should be indexed
Improving results with links
Blogs and search engine results

META tags describe the content of your site, are placed in the header (between the head tags), and are not visible when the page is displayed in a browser. Although META tags are not as useful as they were in the early days of the Web, they are quick and easy to write, so I create them for my clients’ Web sites anyway. Search Engines (SEs) put less and less emphasis on using META tags as they realized that people could load META tags with keywords that were not really relevant to the site in an effort to improve results. So if you use META tags, you should make sure that they capture what is really there in the visible content of your Web site. And never rely entirely on META tags! In fact, given that not all SEs use them for relevance, you could skip them and just try the other ideas described after the META tag discussion.

Here is an example of how to use META tags:

<meta name=”robots” content=”follow, all” />
<meta name=”description” content=”Original and unique Montana posters celebrating Livingston, Bozeman, Missoula, Montana and the Rockies; Website and graphic design.”>
<meta name=”keywords” content=”montana,posters,unique,designs,web design,web hosting,sparkling dawg,montini,missoula,colada,bozeman,bozmopolitan,scotch on the rockies,lynn weaver,made in Montana,sparkling dog,sparkling dawg,blog,cocktail culture,retro posters,livingston,yellowstone” />

Use the META tag for keywords to list things people might be searching for. Even the search engines that factor in META tags will ignore a lengthy list of keywords, so don’t get too carried away with a long list of keywords. The real value is having a site that uses a lot of genuine keywords that are truly relevant in the visible text. So if you put anything in your META tag for keywords, use words that really do appear on the site. The META tag for description is where you put the blurb that you want people to see when you do appear in search engine results. SEs do look at the description, as well as the TITLE tag described next.

Far better than the META tag is to use good titles for your pages. A TITLE tag (<title></title>) that just says “Home,” or “Untitled” because you forgot to fill it in does you no good at all. Look at the title at the top of the browser window. “Search Engine Optimization : Graphic Design. Web Design & Hosting. Unique Montana Posters.” My blogging software (WordPress) has appended the title I assigned to my blog pages to the subject of this post. All pages will include the title text that I defined in my header. (Now that I look at it, I realize that I should make sure that “Sparkling Dawg Design” gets in there somewhere. But I do at least have the words I want SEs to pick up.) Blog pages are generated from scripts and content stored automatically in a database, but folks with static Web pages can, and should, write significant titles for each page on the site.

If you have images on your site, you can add useful text in the IMG tag’s “alt” and “title” arguments. Here’s an IMG tag that adds keywords about the site:
<img src=”images/VintageYellowRobot_w.jpg” alt=”a picture of a vintage robot toy” title=”this robot helps you, spambots do not”>
This approach has three virtues: when a user passes the mouse cursor over the image they get descriptive info, visually impaired visitors using software that reads the page to them know what is there, and SEs grab the added keywords about what is on your site. Try this with the robot picture I use on my spam prevention site (ctrl-zweb.com). I’ve just communicated to SEs that the site has a vintage robot picture, and has info about spambots. And if for some reason a visitor can’t see the image, they’ll know it’s a picture or a vintage robot toy. (Believe it or not, I get visitors who are searching for “vintage robot toys.” But that is a happy accident, and while I don’t want to be misleading about content, I do want to use the image, and I believe in using the “alt” argument to meet accessibility standards. Besides, I get points for not linking to ctrl-zweb.com when it isn’t germane to our topic.)

So, good titles, and both visible and invisible (in tags) page content that contains words that really convey what your site is about, will help optimize your site. Before we leave tags, there is one META tag that I do rely on, and it requires a good robots instruction file to go with it:
<meta name=”robots” content=”follow, all” />
This tells SE indexing robots to go through the whole site. Use this in conjunction with a robots.txt file that should be placed the top-level directory of your Web site. The robots.txt file allows you to specify instructions to SEs (also known as User-agents) to allow them to index your site, and to disallow indexing of any directory that you do NOT want indexed. Here are two excerpts from a typical robots.txt file. A robots.txt file consists of a number of directives like the ones below, defining which User-agents should or should not traverse which Web directories.

This tells Google to index all directories because “Disallow” is empty.
User-agent: Googlebot
Disallow:

Whereas this tells all search engines to not index a directory named cgi-bin. That directory contains scripts that run on your server and should be kept out of search engine indexes:
User-agent: *
Disallow: /cgi-bin/

If you had a personal directory on your site where you keep your vacation photos for friends to view, but did not want SEs to index content there, you might do this (using the real directory name, of course):
User-agent: *
Disallow: /myvacationpictures/

For more info on how to safely and effectively use a robots.txt file, see http://www.robotstxt.org/wc/robots.html. You should know what you’re doing unless you want the wrong content showing up in search engine results, or if you don’t want to risk shutting out search engines from things they should index. So if this is a do-it-yourself task, read up a bit before jumping in to robots.txt files. Once you have the robots.txt file that you want, upload it to the top-level or root directory of your site. That is the directory where your home page (index.html, or index.htm, or index.php) resides.

Google provides Webmaster tools that allow you to actually see how the powerful Google engine is using your robots.txt file. If you sign up for a free Google account, you’ll get access to Webmaster tools that analyze your robots.txt file, and that also allow you to upload a file that tells Google about the pages on your site and how often you want it indexed. The latter file that you will need to upload to Google via the Webmaster tools interface is called sitemap.xml. You can automatically generate one at http://www.xml-sitemaps.com by simply entering your site’s URL, then you download the resulting file to your computer, and upload it to the top-level directory of your site. (Bloggers can get a blog-based Google site map generator to add to their blogging software, and that makes this process much easier.) To take advantage of the ability to have Google use your sitemap.xml file, you have to sign up for a free Google account. On the Google home page, click the Sign In link at the top right and follow the instructions or go directly to Google account creation. Google will ask you to verify that you are the site owner, so you’ll need to be able to put a test file on your server or create a special META tag for Google to test or you won’t be able to use these free Google features.

Another way to get SE attention is to have a lot of good links. Get your friends to put links to your site on their sites (and return the favor). Or keep a list of useful links on your Website, pointers to sites that will interest your visitors and that are related to yours. I got hits on sparklingdawg.com even when it was a static Web site because I had a page of Livingston Links. I also got my sparklingdawg.com link listed with free or inexpensive ($10 a year) sites for Montana products, and listed artsmontana.com with online Montana arts directories. (Note that you can allow a user to follow a link while keeping your site open in the browser by using the target argument in your hyperlink: <a href=”http://www.artsmontana.com” target=”_blank”>. If you are using a Web publishing tool instead of hand-coding your links, there is probably an option to specify the target in the hyperlink dialog box. Be aware that setting the “target” to “_blank” opens a new window, or with newer browser versions, a tab.

Also, keep your Web content fresh. If you have a Web site that just sits there without your taking some steps to get it linked to, or trying some of the steps described above, or leaving the content untouched for months at a time, you’ll stay at the bottom of search results. Be aware that changes you make today might take a day or two to show up in search engine results. SEs index frequently, but not that frequently.

Consider converting your static site to, or building your new site as a blog. You can get a blog hosted for free (wordpress.com, typepad.com, blogger.com), or you can host it on your own server. Blogging software is free and there are a lot of blog themes (templates for the structure and the look of your blog) and blog plugins (add-ons or extensions, like the Google sitemap generator for blogs mentioned above) that are also available for free download. SEs love blogs because blogs have two qualities: lots of links, and regularly updated content. (To see how one of my blogs fares in SE results: enter “livingston arts events” in Google; artsmontana.com tops the list. Note that every time an SE does indexing, results can change because new sites are added to the Web and the content on millions of sites is always changing.)

Cross-linking with other blogs is easy and fun (assuming you like the content and tone of the blogs you might link to). Every time that you leave a comment on another blog, you’re asked for your Web site name, so that automatically gets you linked to. But, in this as in all of the other techniques mentioned, it is a bad idea to cheat. It’s bad blogging courtesy to just comment for the sake of propagating your site link, and a smart blogger will shut you out or a spam filter will get you if your comment isn’t really relevant. Bogus linking schemes, irrelevant keyword seeding, and other no-no’s are counterproductive in the end. Besides, they’ll adversely affect your Web karma.

If you decide to create a blog, my personal recommendation is to use WordPress (WP) blogging software hosted on your own server. I’ve used WP extensively and find it quite good, and it seems to have more free themes and plugins than any other tool. Note, too, that when you host your own blog, you can do many things that you cannot when you use a free hosting service. Blogs don’t have to be bloggy; you can use one for a more traditional site (without writing posts) and your users won’t always know the difference.

Web sites are definitely not a good example of “if you build it, they will come.” Unless you are Kevin Costner with a corn field to spare, you will have to go out and find your potential visitors, and let them know you are there and have something worthwhile to offer. One simple thing that I am amazed people forget to do: put your Web site on every bit of material you send out into the world, whether it is an ad, a business, card, a portfolio, or an email. And for heaven’s sake don’t weigh the message down with http://www (just put veryfinewebsite.com in your materials), and never say “log on to veryfinewebsite.com.” This is one of my pet peeves; unless visitors really do have to log in (not on!) to your site, this expression is not only inaccurate but very old school (unhip!).

Livingston folks who have questions can leave them here or give me a call. I do this for a living, but I am happy to answer questions from community members if the answers are brief. My goal is to make those things about the web that are actually useful and worthy (alas, much is not) available to people in my community who are not Web or tech-savvy. If you want me to do the tasks described above for your site, or want any other Web help (hosting, Web site design, blog setup), ask me about rates.

And tomorrow I am going to see if using the words “Kevin Costner” on my Web site has improved my search results in any way. Probably not.

Posted by Lynn | Filed Under Web Tips |