Kyle Ritter

I’ve been talking to my friend Greg from Web-Fu Design in Portland about finding ways to parse Commission Junction datafeeds and insert them into a mysql database. Basically, my goal is to create a complete product catalog from these feeds yet create a search engine friendly, spiderable catalog.

I think using mysql, php, and mod_rewrite is the key, as I’ve seen how successful that tactic has been on Portland Bars.

The difference is, I’m not sure how I’ll be able to create understandable urls. I am going to be creating this for a website to buy wine from. So instead of having a page called product123456.html to deliver information about a nice 1994 Pinot Noir wine, I would have a page in a products sub directory with the actual product name as part of the url. For instance, the way I am trying to write this is as follows: products/1994-pinot-noir.html

I just think that psychologically surfers are more prone to respond to a page whose URI indicates that it has information they are looking for. After all, what does URI stand for? Do I hear Uniform Resource Indicator, keyword being INDICATOR.

In the grand scheme, I think this is a very small factor, but I think it has enough value to merit a little extra work.

I’ve decided it is time I make an honest effort towards using strict valid XHTML code. Lately, my code has often been a neapolitan mixture of HTML 4.0, a little CSS, and sometimes XHTML, often on the same page.

Much of the reason I haven’t been consistent with my coding style has been due to the need for rapid deployment, and taking on a wide array of tasks. This gave way to my pragmatic approach of “Just make it work”.

It’s long overdue for me to set more rigorous standards when it comes to the underlying code. Here are some sites I have been using for reference.

w3c XHTML page
XHTML Tutorial

I am pleased with how the Portland Bars website is turning out. Jason and I have worked our asses off getting it up and going, and we are starting to see a nice flow of traffic.

The blog is a handy asset as well, complete with a “bar detector” module that tries to match any bar name mentioned in a post then links the bar name to it’s corresponding page in the directory. This is helpful for breaking up the text and creating stop words, and helpful for getting spiders to crawl newly added bars, since we will often mention them in the blog after they are added.

In a past article I posted about Yahoo Slurp, I had mentioned Nocturnal, a nightclub here in Portland. Ironically, I have got the page listed at #10 in google, and according to my search engine logs, have recieved 13 visits based on the terms Nocturnal Portland since then. It seems that we are appearing above the actual bar itself, due to it’s artistic yet spider un-friendly design.

I programmed the site to be very friendly for visitors, hence making it friendly for search engine spiders as well. It is wonderful to see a fresh site indexed and appearing so quickly.

Mark my words, kiddies, traffic generation comes from recognizing your website as a business asset, not as a chance to satisfy your ego. This is an unpopular opinion amongst the creative but misguided designers in the community.

I don’t have time to elaborate on this, but this article brings up some solid points that I always try to make apparent to my clients in regards to optimizing your website for search engine spiders.

One point made in the article that I like to particularly drive home is this one: Be certain that geographic specific information is mentioned on each page of the site. Always have a contact page that also lists geographic specific information.

Many surfers search geographically, yet so many business websites completely disregard this fact.

For instance, on my Portland Bar website, I wanted to list the websites of all the bars in the database. Finding these websites was a nightmare, most of them only had the bar name listed in their content, with absolutely no indication of where they are located. A search for “Nocturnal Portland Oregon“, Nocturnal being a bar in Southeast Portland yielded impossible results. I was able to locate the site, but only after much effort was in visiting all of the pages that appeared for this term and search for a link.

Hopefully, my listing for Noctural – Portland Bar will help them be found in the future.

I have just finished up my most recent traffic generation campaign for Flying Rhinoceros.
Play on the Farm is a series of educational books and DVDs with a faith-based message. This website includes some great flash animation from some of the finest illustrators and animators, as well as voices by Vince Gill and Amy Grant.

Our plan for traffic generation was two fold:

Viral Marketing
This includes a Play on the Farm “send to a friend” script that allows parents and educators to inform up to 10 friends about Farmer Bob’s world

User Content Generation
All mail messages and subject lines sent through the send to a friend script are archived in a database. All messages are displayed live on the website. This creates an instant testimonial system, as well as creation of a brand new page of content every time a sender sends a message.

In order to protect the senders and recipients identities, their contact information is not stored, only the messages.

When you take on a client, especially when you are newer to the web design industry, you may feel compelled to “impress” both yourself and your client with the coolest, fanciest design. Maybe you want to add all kinds of features that you feel are fun tasks that enhance the website.

You are designing as a power user, as a web professional, you spend countless hours surfing websites, and have seen every design under the sun. You want your newest project to be fun, not boring.

After a few months of solid traffic and usage monitoring, you will begin to see otherwise.

Reality check time – Most web surfers aren’t power users. They are just looking to get to the information they need, and they want to get it as quickly as possible. Even users such as myself are prone to hit the back button if we can’t immediately find what we are looking for. Do I have to spend time hunting down a link for an article? Do I have to suffer through a tedious flash intro and gratuitious animation just to look at a product image? If so, I go elsewhere. I’ve got better things to do, namely, design “boring” web sites.

Even such things that might be obvious to power users, such as *breadcrumb trail navigation, are often overlooked by Joe Surfer. Breadcrumb trail navigation makes perfect sense to us, because it seems like such an efficient way to find your way “back”.

An interesting bit of irony: seeing as Joe Surfer generally just wants his information as quickly as possible, and with breadcrumbs being the fastest way from a specific page to other relevant pages, it would make sense that Joe Surfer would appreciate and use the breadcrumbs. Yet user tests show that Joe Surfer is more than happy to just go back to the home page and start over, rather than use unfamiliar navigation.

To use the catch-phrase lingo of our long dead DotBomb brethren, you’ll need to undergo a paradigm shift from fun to boring in your methods of design. That’s right. You are not an artist. You help facilitate information delivery.

Spend some time conducting user tests. Gather up friends and family of all experience levels to surf the website. Make notes on what they look at, how they get their, and where they are lagging. This will pay off for both you and your client: Your client will be happier with website performance, and happy clients generally help you gain new clients.

* Breadcrumb trails are a path of links generally found just above the content page that show the path and all the stop one could make on the way to a home page.

And example of breadcrumb navigation:

Home>>Website Tips>>2004 Articles

For those living in the dark, MSN’s indexing robot has been heavily crawling websites for the past few monthes. Some folks with larger sites have gone so far as to ban MSN bot due to aggressive crawling. Most of my sites have no more than a couple of hundred pages, so it hasn’t affected me much.

Then, about 2 monthes ago, they gave users and opportunity to test their ranking algorithms using their tech preview.

I was rather impressed myself, many of my sites (all of which contain useful, relevant information) enjoyed top rankings, while other webmasters were throwing a fit: They felt that the results reflected low value, irrelevant websites that had been unethically manipulated for high rankings.

When MSN goes live, I believe we will see fiercer closer to evenly split competition between Google, Yahoo, and MSN for web surfers. Kind of like the internet worlds version of the “Big 3″ automakers. As of now, Google retains the high market share of internet searches. Many companies, such as AOL and until recently, Yahoo, were using it’s search results for their own engines.

As it stands, many peoples browser default to MSN’s home page, which gives them a natural advantage. For webmasters, we will be seeing a whole new set of results in our server’s referrer logs. One thing I’m looking forward to: Watching the Big 3 constantly adjusting there algorithms to deliver better results than the other. This will be beneficial to both surfers, and webmasters who optimize their websites in an ethical, relevant fashion. Both parties stand to win more, while it will leave Black Hat optimizers continually scrambling to learn the newest way to manipulate the algorithms.

Jack Humphrey wrote a great article for SiteProNews called Active Vs. Passive Website Traffic Generation.

It really got me thinking about my own traffic generation tactics. I do actively seek to generate traffic from search engines and link directories, but also, I don’t spend too much time generating reciprocals and joining link exchanges. I don’t frequently trade links with other webmasters, nor network with them.

Yet I hungrily read the latest news and regularly post information to both my blog and my other informational based sites, while seeking out like minded sites that I could trade with. Then I looked at my Portland Web Design Company website: I have little traffic to it, and have spent no time linking it, except throughout client sites and sites within my own network. While I make sure to optimize, am I passively seeking traffic when I should be actively seeking?

I use this weblog as both a professional promotional tool and as a venting utensil. While it might seem like neither should go hand in hand, I am most satisfied with the results. I’ve had a few referrals come in due to my posts. While some of my posts are decidely un-professional, it puts a human touch to what otherwise could simply be a cold hard information.

Well back to work. I need to actively promote my website.

It’s not a good idea. They have access to your site, and if they have done their job properly and it is well promoted, should be earning you revenue through sales or referals, and at least be placed in front of your customers eyevalls. If you try to screw your web designer, bad things like this can happen.

Designers: ALWAYS take a minimum 33% upfront, long before you ever fire up Photoshop or write a single line of HTML. If you are a professional, then you are worth it. That is how the business works. Collect the remainder of the fees upon delivery.

I’ve been building a database of bars and nightclubs in my home city of Portland.

However, I didn’t like the messy URLs generated when I passed the “bar_id” parameter to the PHP script:

<a href=bar_info.php?bar_id=1> Shanghai Tunnel </a>

I wanted a nice clean way to make the URL contain both the bar’s name, extract information based on a value passed to the database, AND appear to both search engines and visitors as a static HTML page.

First I added a new column to my mysql table called “static_url”. Then I wrote a function that generated a 2nd version of the bar’s name with all odd characters stripped out, while spaces where replaced with a hyphen (“-”). Then this info was added to the static_url column of my table for the corresponding bar.

For instance, the record row for “Shanghai Tunnel” now contains a value of “shanghai-tunnel” for the “static_url” column

Unless there are two Shanghai Tunnels in Portland, this is unique a non-numeric indentifier for each bar. So instead of accessing bar information by passing a numeric ID value, I could now pass “shanghai-tunnel” as a value.
I can extract the same information about Shanghai Tunnel using “bar_info.php?static_url=shanghai-tunnel” as I could with “bar_info.php?bar_id=1″.

So my next step is to make bar_info.php?static_url=shanghai-tunnel look like a static html page. So I

1. Point it to a non-existant folder called “bar”
2. Point it to a non-existant file called “shanghai-tunnel.html” located in the non-existant folder called “bar”

This is what my URL now looks like
<a href=http://www.pdxbars.com/bar/shanghai-tunnel.html> Shanghai Tunnel </a>

Now it’s time to make these non-existant files appear real. Here comes the magic of mod_rewrite and regular expressions (which is something I’m still learning about). Basically, we want to tell the URL “bar/some-bar.html” to be processed as “bar_info.php?static_url=some-bar”

Open up your .htaccess file (make sure you have a back up in case you screw something up)

RewriteEngine on
RewriteBase /
RewriteRule ^bar/([^.]+).*$ bar_info.php?static_url=$1

Make sure your .htaccess is in your root web folder, and off you go. Modify the script at will, this should work with any sort of web programming language such as Perl, ASP, ColdFusion, JSP, or whatever, provided you are running apache with mod_rewrite enabled.

*For those uptight programmers out there who don’t like the idea of using non-numeric identifiers. My decision was based on the fact that
http://www.pdxbars.com/bar/shanghai-tunnel.html
tells more about the page to both users and search engines than
http://www.pdxbars.com/bar/2361826123.html