I’m rapidly coming up on the end of the beginning. Childhood, different grades leading up to high school graduation, and then a few more years at a university are the beginning, designed to help young people get started in life and be ready to take care of themselves and contribute in some kind of productive, and hopefully meaningful, way to society.

I graduate from BYU a week from today.

I have a rather hectic week of finals, ceremonies, and celebrations ahead of me.

I’m really excited to be graduating. I’ve been working for this moment for the last 24 years… I’m thrilled that I will get to work on real projects full-time instead of spending most of my time doing what often feels like “busywork.”

I hope that I can find a good way to put all my learning and preparation to good use when I start “real” life in about two weeks.

Hopefully I’ll have a chance to take a breather, too. It may be part of our “beginning,” but that doesn’t mean that education is easy. I have a lot of things to catch up on.

I’ve designed several web sites using XHTML and CSS lately, and I’ve done my best to make them cross-browser compatible, lean, usable, and accessible. I have also tried to test them in as wide a variety of different environments as possible to ensure that they work correctly for the largest possible audience.

The biggest question is, to what lengths do you have to go to test your site?

My PC has three different browsers on it (Firefox, Opera, and IE 6) for testing. There are a few more that I could probably load onto my machine (Netscape, etc.), but I haven’t really felt like it—that’s a lot of browsers on one machine.

As for Linux systems, I have a friend who has tested a few sites for me using different builds of Mozilla. I haven’t noticed any important differences in how my sites render in Mozilla compared to Firefox (same basic code), so I have been pestering him less frequently lately.

Mac and IE7

The biggest testing challenges I have are for Macs and IE7. I don’t own a Mac, and I don’t plan on spending any amount of my few dollars on buying one. I also don’t feel like replacing my IE6 with a beta version of IE7 or messing around with a non-standard installation. How on earth do I test these two environments?

Dan Vine to the Rescue

For those who don’t have access to a Mac or an installation of IE7, Dan Vine has come to our rescue with two great services—iCapture and ieCapture. iCapture is a server that will take a snapshot of your site using Safari running on MacOS X, and ieCapture will take a screenshot of your site using IE7 Beta 2.

Since most web site design problems manifest themselves (very!) visually, these screenshots allow you to see how your site works on a Mac and in IE7.

“Let’s Discuss Your Coverages”

Just like insurance, we’re playing a probability game here. We need to make our web sites compatible in enough browsers to cover the majority of web users, but there’s a point at which adding compatibility for more obscure users just isn’t worth the expense.

My testing covers Firefox (and thereby Mozilla), Opera 8, IE 6, IE 7, and Safari. According to the W3C Schools Browser Stats, this covers about 90% of all Internet users. I know that I could test a little more thoroughly, but the testing may not be worth the extra effort, especially since the majority of browsers I left out are pretty standards-compliant.

One Glaring Omission

There is one glaring omission that I have made—I don’t test IE 5 anything. Is there a way to do this that is somewhat easy? Some service? I may have to break down and buy an old machine to put IE 5.5 on so I can see how things look. Of course, IE 5 is disappearing, and my standards-compliant, low to no-hack code should work for most other current and future browsers.

What testing (if any!) do you do on your web sites? What tools are your favorite for ensuring your site works for the largest possible audience?

After you have decided on the goals of your site, you can begin to determine how you will reach those goals—you can plan the functionality that needs to be part of your site in order to accomplish your objectives.

Functionality Should Be Tied to Goals

This is the part of web design that can really get messed up, whether by accident or by rationalization of mistakes. It’s easy to forget to (sufficiently) implement some important features, or to add features to your site that aren’t really necessary and that don’t contribute to your core objectives.

The functions of your site must satisfy two criteria before they can be considered for implementation: 1) They must be tied to at least one of your goals and 2) they must be measured. Most website functions are very measurable, and a few, like “general appearance” and “ease of use” are more difficult to measure but are critical to your site and must at least be evaluated and monitored.

The Functionality of KindredLearning.com

To illustrate, here are some of the main functions that kindredlearning.com will have:

• Online Store

  The most important function of kindredlearning.com is to provide site visitors with a way to purchase Kindred Learning’s products online. Kindred Learning is located in a remote area, doesn’t have a strong distribution network, and serves a nation-wide audience, so an online store is critical to Kindred Learning’s success. Its current store is functional, but not stellar, at helping users comfortably browse products, add them to the cart, and check out. KPIs for the shopping cart include conversion rates and average order value.

• Newsletter Opt-in Form, Archives, and Syndication Feed

  Kindred Learning’s monthly home school newsletter is a powerful way to communicate with its target market. Offers, news, and articles are distributed to subscribers through the newsletter. Archives and a syndication feed (RSS) would be very powerful for driving traffic and improving search rankings. Opt-ins, opt-outs, open-rates, and the number of subscribers are all measurable indicators of how this function of the site is doing.

• Home School Articles

  Home school articles would be great “spider food” for attracting traffic to the site, and useful articles would provide a helpful service to site visitors. Traffic, conversion rates, and keyword rankings are all measurable indicators of how well home school articles are doing.

• Company News

  Information about new products, product updates, sales, and visits to home school conventions is very important to Kindred Learning’s target market and provides a strong incentive for visitors to return to the site. Company News could also be published using an RSS feed and measured accordingly.

• Information About the Company

  Because most of the traffic to the site comes from search engines, kindredlearning.com needs a very well-done information section to acquaint visitors with the company and its unique selling points. This is one of the functions of a site that is difficult to measure. One way to measure it could be to compare the conversion rates of paired pages—one with and one without corporate information, testimonials, or other credibility elements.

Some functionality that may be added to the site later includes a message board or other community features, a home school cooperative locator tool, product reviews, and a blog run by a member(s) of the Kindred Learning staff.

How the Functionality is Tied to Goals

Each of the above functions is tied to one of Kindred Learning’s goals. The online store is the primary method of driving sales for Kindred Learning. The newsletter and its archives and RSS feed primarily drive sales and nurture relationships with customers. Home school articles attract new leads and serve current customers. The company news section improves communication in one direction, from the company to the customers. Information about Kindred Learning helps reassure prospects that it is a good idea to become a customer.

Conspicuously missing from Kindred Learning’s initial strategy is a method for customers and prospects to communicate their desires to Kindred Learning. The information section will obviously list an e-mail address, but this may not always be the preferred method to elicite customer responses. A message board, product reviews, and/or a blog that allows comments would be great ways to interact with customers and get their feedback. Kindred Learning, however, already knows of many things it can do to improve and has other, non-web-site feedback channels established, so feedback functionality on the web site can wait for a little while.

Hopefully this example will help you to see how the functionality of a web site is determined by its primary goals. Remember, if it doesn’t contribute to your site goals or it is very difficult measure, it may be smarter to leave it off your web site.

An important part of Internet marketing or business on the web is knowing what your competition is up to. Visiting your competitors’ web sites, reading their marketing copy, etc., is a great start when it comes to getting to know them.

What do you do, however, when you need hard numbers?

Where Can You Get Web Site Traffic Stats?

I spent my entire last post skewering Alexa for its inconsistencies and problems. Where can you get web site traffic rankings and information?

Fortunately, if you have data on your own web site, you aren’t completely out of luck. You can compare data from your site against other sites using Alexa, as long as you are in the same market—if site visitors to your site and to your competitors’ sites are similar kinds of people, then relative rankings, page views, reach, etc., in Alexa won’t be skewed so badly.

For example, if you compare two sites’ page views, you can get an idea of their daily traffic. If your site has 2 pages views per million on Alexa, and your competitor has 4, then your competitor is getting approximately double the page views you do. If your daily traffic is 150 unique visitors, then they are getting about 300.

My favorite new tool for using Alexa is Alexaholic. It allows you to compare up to 5 sites, over many different time periods, on much larger graphs. Check it out.

Page Rank, Inbound Links, and Rankings Oh My!

Another neat tool that aggregates information about your competitors is URL Trends site tracker. The free version tracks sites you want to follow on a monthly basis, giving you information about how a site stacks up in Google, Yahoo, MSN, Alexa, Del.icio.us, and a few others. You usually don’t need information on your competitors more than monthly, so this tool works well for most purposes.

If You’re Obsessed…

If you are really, really curious, you could research a site keyword by keyword. Use inventory.overture.com to estimate how many times a particular keyword is searched (multiply by about 3 to account for Google and MSN traffic, too), then search using those keywords. If your competitor is found on the first page of results, give it about 10% of the overall searches. If your competitor is 1st, 2nd, or 3rd, give it about 30%, 20%, and 15%. No, I don’t remember where I heard those percentages or if they are accurate. Let me know if you have heard of a better way to estimate traffic from search engines using this or a similar method.

Conclusion

Competitive intelligence is important, and as SEO gets more competitive, competitive intelligence will become more important to your success.

Are there any other reliable methods for researching competitors that you know of?