I've worked for bad companies and good companies. Even had my own. What makes one company more successful than another? Well, here's a few tips:

• Answer the phone. Sounds silly doesn't it. Well regardless of how some might spin it or try to convince you otherwise with fancy graphs or stats or ROI statements, your customers want to speak to a human. They want to speak to a human who can help them. Your customers do not want to be put on hold, they do not want to be transferred to someone else. They want one number to call, and they want to be able to reach someone on their schedule, not yours.


• Put it online. Your customers are generally not stupid. So if you sell a product or service, put it online. Make it clear what it is, why one would want to buy it, what you need to use it, and the price. If there are support issues that can come up after purchase, put the resolutions online in the form of support faqs or troubleshooting tips or a knowledgebase. You'll empower your customers to help themselves and save money and resources by not needing as many folks manning the phones. This is no way detracts from the first item above.


• Automate everything you can. If you have online ordering, automate provisioning. Automate shipping. Automate billing. Give quotes? Automate it. Take bookings? Automate it. Wonder why ebay and amazon.com are so successful? Take a guess...


• Design your site to be as user-friendly and as easily navicable as humanly possible. Don't try to do it yourself. Hire a pro, and don't be cheap. If it's worth doing, it's worth doing right, and worth doing right the first time.


• Integrate your contact management, support ticketing and sales into a single interface. When your customer calls in you should be able to see what they bought in the past, what issues he had, who he spoke to, who makes the buying decisions - everything. It's not hard if you have the right systems in place.


• Under-promise and over-deliver. Think 'Scotty' on Star Trek.


• Be up-front and forth-right with everything in all your dealings.

More later. And oh, price is NOT an objection. Price is irrelevant.

Stuff I've Learned

Read: The hard way...

• If you have a Linux box, install it all^tm. Yes, when you do the install, just install EVERYTHING. Trust me on this. Your life will be so much easier. Otherwise you'll be trying to satisfy dependencies till the cows come home. Just spend a little time locking it down afterwards.


• Never, EVER, enter your main email address into online forms. Use a throw-away hotmail/gmail/yahoomail address. More times that not you get added onto a mailing list that then gets sold to other companies to send yet more spam. And really, when's the last time you ever read a website's terms of use or privacy policy?


• For God's sake run a virus-scanner and a firewall. Don't say I didn't warn you...


• Back it up! Then backup your backups. Backup often, and keep at least one backup off-site in case of fire or theft. If you can't remember to backup, subscribe to a service that does it for you automatically, such as any Connected.com reseller.


• Don't buy the cheap stuff, but don't buy top-of-the line either. Buy last-year's model when they blow it out cheap ;)

Sure, you know all this. Just one guy's suggestions to make your life a whole lot easier and maybe save you some sleepless nights wondering how you'll ever get your vacation pictures back when your hard drive crashes...

I turfed my robots.txt file, however thanks to the Belchfire.net member who emailed me and pointed out a formatting error. Why did I delete it? Because I was unable to confirm the precise formatting of the file. If it was written in stone, and I could be assured of 100% compliance by all search engines, I would have just made the necessary changes and left it in place. But since I couldn't, I figured, 'What's the point?'. In addition, a robots.txt file can be just an ad for what you don't what indexed. So now you've just told everyone where the stuff is you want kept secret.

A better method is to not publish what you don't want known and then use other methods to restrict/block access to those files and folders such as .htaccess files or code within your pages to restrict access by IP or user-agent. Since I've always done this anyway the robots.txt file was redundance for me in this regard, so I saw no reason to maintain it. Lack of a robots.txt file will not effect your web site's ability to be indexed by search engines.

You CAN make money on the internet.

Belchfire.net was recently featured on About.com. While Belchfire.net is not a wallpaper site specifically, they've rated us one of The Top 5 Very Best Computer Wallpaper site of 2004. Most excellent. Thanks guys!

If you're new to the net or are looking for some great pointers and resources, drop by the About.com Beginners Guide to the Internet.

I've moved my other site, Belchfire.net, from PHP-Nuke to Invision Power Board. In addition to changing the content system on the site, I also changed URLs from www.belchfire.net to themes.belchfire.net.

Never having done this before, I have no idea how this will effect my search engine listings and rankings. Prior to the move (completed on August 4, 2004), www.belchfire.net had over 150,000 backlinks on the major search engines (Google, MSN, Yahoo, AllTheWeb, etc.). I've redirected the top pages and sections using 301 http response codes, (i.e. http://www.belchfire.net/xpthemes.html redirects to http://themes.belchfire.net/index.php?act=Downloads and sends a 301 'Moved Permanently' response to the client), but until Google and other sites update at the end of the month, I'm holding my breath that I don't lose all the links and listings I've worked so hard to maintain over the last three years.

Listings prior to sitemove:

Search Engine Saturation		Back-Links
AllTheWeb	62500	AllTheWeb	72900
AltaVista	63000	AltaVista	73500
Google	18200	Google	3630
HotBot	7219	HotBot	18634
MSN	-	MSN	17542
Yahoo	32400	Yahoo	74500

Google PageRank:  6

Yahoo Rank:  5

I kept the old site up at a new URL, oldsite.belchfire.net, and linked it off the themes.belchfire.net homepage. Good news is Google is spidering the new site well. I'll keep you posted on how my rankings fare during the next Googledance. Stay tuned!

I see it time and time again, and it's such a simple thing that many overlook. Search engines look at a number of things when deciding how to index your pages. One of which is the name of the page.

Let's say you have you have a web gallery with artwork for sale. You might be tempted to name your gallery mainpage simply 'gallery.html'. This doesn't tell search engines the nature of your gallery. Name the page something that reflects the focus of the gallery. For example, a gallery focused on supermodels might be named 'supermodel_gallery.html' or 'supermodel gallery.html'. Similarly, a gallery details page for a specific piece of art shouldn't be named 'galleryimage33.html'. Name it to reflect the subject of the image. Perhaps 'Naomi Campbell 1.html'. Get the idea?

Do this for every page on your site and I guarantee your search rankings will increase over generic filenames.

Well waddyaknow. Belchfire.net had it's biggest day ever yesterday. I've never been able to exceed 5000 unique visitors per day for any length of time, and since March 1st, my web traffic has doubled:

01-Sep-03 3728 49925 950119 7.95 GB
...
01-Mar-04 4910 49680 729579 7.44 GB
30-Mar-04 9171 113725 1584520 18.93 GB

As you can see by this crude extract, things have been very consistent over the last 6 months, and just in the last 30 days things exploded. It was not a single day spike (didn't get slashdotted or anything). It was a gradule, steady increase. I attribute the phenominal gains to the exact things I've written about in this blog. It was not a single change or modification, but rather the combination of all of them. What I changed after March 1st, 2004:

• modified belchfire.net pages to more closely match the Google and Yahoo keyword and element density report from goRank research


• entered custom meta keywords and description tags on most pages (still have around 450 news articles to do)


• entered custom page titles on all major pages and sections


• created a search engine friendly sitemap using netroglycerine.com's Site Map Generator (free)


• a css trick I'll expand on later

Previous to this, I also convered all dynamic URLs on Belchfire.net to static. That didn't effect web traffic to a great extent (as Belchfire.net was already in the top 10 search results for my keywords), but it did increase the number of pages the search engines indexed.

Next post I'll look at web stats. What data to collect, how to interperete the results and how to monitor them and tweak your site based on that data.

Belchfire.net | XP Themes Community, Windows XP Themes, XP Wallpaper, Boot Screens, Logon Screens and Icons.

Few web sites have one but a sitemap can be a valuable tool to help search engines properly index your site.

A sitemap is basically a single or multiple web page that contains links to your homepage and all the other pages on your site. While search engine spiders are generally quite efficient and finding and following links in a given web site, a sitemap gives them a map to go by and ensures that all the pages you want indexed will be indexed.

If you have a web site with only a few pages, a sitemap can be created manually. If you have hundreds or even thousands of web pages, a program or script that generates a sitemap is recommended.

For an example, see the sitemap page for Belchfire.net here. Notice that the page contains only links and a description of each page, based on the page title. It's generally a good idea to index pages hierarchically, and to breakup the sitemap into multiple pages if you have a lot of pages.

Belchfire.net has around 220 000 web pages and unfortunately I haven't found a sitemapping program that can accommodate this number of pages let alone break the sitemap up into smaller pieces (so my own sitemap doesn't adhear to this rule). You don't want more than perhaps 1000 links on a single page as some spiders may stop following links past a certain point.

Here are some free and commercial sitemapping tools to try:

• Netroglycerine Sitemap Tool - free - what I used to generate the Belchfire.net sitemap - script died after around 3700 links


• Xenu's Link Sleuth - free - more of a link checker but can optionally generate a sitemap during the final report - froze up trying to run it on Belchfire.net after around 65 000 links checked


• SiteTree - free - command line java program and classes that can generate a sitemap by traversing a directory on a hard drive (looks good, but won't be of much use if you don't have command line access to your web server - not sure about the accuracy either)


• Extreeme SiteXpert - commercial - seems to be the best commercial offering offering dozens of options, templates and formats


• Site Map Pro - commercial - another good commercial offering - demo limited to 10 pages

You can even get fancy with some of these commercial offerings by having them add the first few lines of text under each link. Search engines will love the content and your sitemap may even get higher page rankings than some of your content pages. Really, a site map may or may not be of any real benefit to your human visitors (assuming you already have a decent navigation stucture), but adding a sitemap can help improve the number of pages indexed by search engines by providing them with a starting point and a map of files to index. Don't forget to add an appropriate title and meta tag info on your sitemap page(s)!

The proof's in the pudding...

Since adding a sitemap to Belchfire.net, indexed pages on the major search engines has increased from 25 000 to over 52 000 (Alltheweb, AltaVista, Google, Yahoo, Inktomi combined) and my web site traffic has increased by over 30% (an additional 1500 unique visitors per day).

What's got those in the know abuzz in the search engine world? Geo targeting, or rather, geographic location targeting.

Here's the scenario and I'm sure you're familiar with it. You're traveling to Toronto and are looking for a restaurant or perhaps a theatre near your hotel. Search an search site for 'restaurants' and you'll the Subway homepage, restaurants in Atlanta and perhaps a few directory sites - basically everywhere but where you really are. So you then narrow your search to 'restaurants in toronto'. Great, but there are thousands of restaurants in Toronto. You want one close to your hotel that serves sushi (sushi?). Hey look, it's an example... After continually narrowing your search parameters and grepping though dozens of search results pages you might, MIGHT, find one that is indeed close by that serves your favourite dishes.

But you shouldn't have to do any of this. Imagine if you could just search for something and the results are based on your geographic location. Using GeoIP information and perhaps even GPS information, search results could be narrowed down to a few kilometers or even blocks of your actual physical location.

GeoIP information isn't all that new, but tying search results to GPS information is, and it's coming soon. One of my sources has indicated that at least one of the top three search results providers is looking at offering just this in the very near future. Initially, it will probably be tailored to mobile devices such as GPS-enabled cel phones and hand-held devices, but I expect that the adoption rate of the new technology, once introduced on a wide scale, will be high and expanded into laptop and tabletpc hardware and services. So keep your ears to the ground for updates as we get them.

What can you do now if your business caters more to a local or regional market? There are some special meta tags you can place in your pages to tell supporting search engines your physical location or general geographic region. These are called the Geo Tags, or Geographic registration of HTML documents. These are some of the proposed meta data:

<META NAME="geo.position" CONTENT="latitude; longitude">

<META NAME="geo.placename" CONTENT="Place Name">

<META NAME="geo.region" CONTENT="Country Subdivision Code">

Only the geo.position latitude; longitude is necessary, the others are optional. Note however that these have not been widely adopted. The only search engine I know of that currently supports these is Northern Light (specific to MapBlast). Note also that "The tags describe the position of the resource described on the page, for instance a beach or restaurant, not the company hosting the page, the company managing the resource, or the server hosting the page" - Geo Tags Element Descriptions page.

Gibablast recently announced that they now recognize the following custom meta data:

<meta name="zipcode" content="87112,87113,87114">

<meta name="city" content="albuquerque, abq, rio rancho">

<meta name="state" content="new mexico">

<meta name="country" content="usa, united states of america">

<meta name="author" content="matt wells">

<meta name="language" content="english">

<meta name="classification" content="products,product">

However it appears to be beta and I could not get any results matching my any search other than the two examples listed on their tagsdemo page.

So based on this, even if you did use these Gigablast-specific tags or the proposed Geo Tags, it likely wouldn't be of any benefit, at least for now. The best method of targeting your audience more regionally is to:

1. Choose a domain specific to your country of residence or business (.ca, .uk, .fr, etc. - full list can be found at IANA).


2. Choose a domain name with your city or country in the name (i.e. torontosushi.ca).
   
3. Enter your city/country in your page titles


4. Enter your city/country in your meta keywords and description


5. Enter your full postal address on a Contact Us page linked off your homepage, on your homepage itself, and on any other relavent pages where region-specific information appears (at minimum your city and/or country).


6. Enter region-specific information in image ALT tags where appropriate (driving instruction maps, city skylines, etc.).


7. Ensure that the text and/or ATL tags of external links to your site reflect your regionality.

But all this said, the one to watch will definately be Yahoo's new search. While we have yet to confirm, geographic positioning could be on their roadmap for future enhancements to the service. If Yahoo or another top search provider does it you can bet others will and the specs will be worked out and (hopefully) standardized.

On February 18th, Yahoo replaced Google as it search engine provider in favour of Inktomi. Yahoo owns Inktomi and and it's new search engine seems to be a hybrid of technologies from both Inktomi and Overture (also owned by Yahoo). Note that this is separate from the legacy Yahoo Directory.

According to comScore Media Metrix, Google and Yahoo's market share in January 2004 where:

Google: 79%
Yahoo: 14%

But now that Yahoo has dropped Google these numbers are expected to change significantly:

Google: 51%
Yahoo: 44%

What does this mean to webmasters? Plenty. Web traffic generated from these search engines and their partner and affiliate sites could drastically change as fewer searches come from Google. Sites previously optimized for Google may not fare as well on Yahoo due to the differences in the Google and Yahoo (Inktomi/Overture hybrid) system.

Inktomi seems to weigh more heavily on page titles, meta description and meta keywords, and this is reflected in the Yahoo search rankings page (snip):

• Users are more likely to click a link if the title matches their search. Choose terms for the title that match the concept of your document.


• Use a "description" meta-tag and write your description accurately and carefully. After the title, the description is the most important draw for users. Make sure the document title and description attract the interest of the user but also fit the content on your site.


• Use a "keyword" meta-tag to list key words for the document. Use a distinct list of keywords that relate to the specific page on your site instead of using one broad set of keywords for every page.

So where before a web site designed with Google in mind may not have had unique page titles, meta keywords or meta descriptions, they should certainly have these now, and unique on a page by page basis. Or it may be necessary to modify these on sites already using uniques.

The Yahoo search spider seems to identify itself as:

Mozilla/5.0 (compatible; Yahoo! Slurp)

So keep an eye on your web logs to see if you're being spidered. If so, check your rankings in the Yahoo search results for any noticable jump or drop.

Yahoo will also be moving to a paid submission system on around April 15th. Can't say if they'll charge the same $39.00 US that Inktomi charged, but some are recommending that if you haven't submitted your web site for paid submission on Inktomi that you do it now. Submitting to Inktomi on their paid submission program also ensures your site is indexed fast (48 hours), plus Inktomi feeds other search sites like MSN, HotBot and Overture (and a few others). The following Inktomi resellers accept paid placements for inclusion on Inktomi, or rather, the new Yahoo submit service currently called Overture Site Match:

Network Solutions
positiontech
MARKETLEAP - my recommendation
ineedhits.com
Trellian
web gravity

While we're on the topic of search submissions, be sure to also submit your site on the Open Directory Project. Over 336 other search sites and directories include the ODP data in their results, including Netscape Search, AOL Search, Google, Lycos, HotBot and DirectHit. Basically, you submit once and your site automatically shows up on 336 other search sites. Can't beat that, and it's free.

Finally, getting your site into the Yahoo Directory will most certainly effect your page rankings on Google and likely on the new Yahoo search as well. Suggesting a site is still free, but there's no guarrantees it will be added. The paid service called Y! Express for $299.00 US is the way to go if you have the budget. Guaranteed listing in 7 days.

What's the bottom line here? Well, any webmaster serious about their search rank needs to monitor their rankings on all the major search engines and be prepaired to make changes and tweaks when the search engine landscape changes. Many sites such as those listed under our SEO Tools links offer free and paid placement monitoring tools. Your web logs are your best friend. Use a weblog stats program to monitor where people are coming from, what search terms they used and use that information to update your pages on an ongoing basis. Two of the best commercial offerings are WebTrends and Urchin. Check out Awstats for an Open Source alternative.

My own move to dynamic page titles, meta keywords and meta description info on Belchfire.net seems rather timely.

Read more links:

SEOChat

Found a very old script for PHP-Nuke that generates dynamic Meta tags and another for dynamic page titles. I've modified and updated both and put them live on Belchfire.net (these downloads are of the original files). It will be interesting to see if/how my page rankings are effected. Inktomi seems to weight more heavily on these than other search engines. Will keep you posted on the results.

Previously all pages on Belchfire.net had the very same Meta keyworkds and description, which wasn't necessarily relevant to specific pages.

Just added a new SEO links section. Check out these invaluable search engine optimization tools. Some new and unknown to many, such as goRank. They've got some useful tracking tools but their recent research comparing Google and Yahoo stats is a great read.

All the new technologies available in web site design (meaning, anything other than plain HTML) are great and really give web developers a lot of options for designing a web site. But, when designing your own web site, consider how the technologies you use will effect search engine placement and friendliness.

STICK WITH ONE AND USE OTHERS SPARINGLY

There's nothing worse than loading a web page that takes forever to load and either bogs down your system or crashes your browser. You know what I mean. You've likely come across web sites like this only to curse the creator and end-task or reboot. And Pop-ups and banners - does it ever end?

Don't mix Java, Javascript, DHTML and Flash all on the same page. No browser supports all 100%, and it requires expertise in all where many have a hard time with just one.

Search engines like text. So choose a technology that uses text primarily. For example, don't use Flash exclusively, as search engines don't know Flash.

CSS is the best textual formating technology used today. There is almost nothing that cannot be accomplished regarding layout and style using CSS. Most all current browsers support at least a sub-set of the full CSS 2.0, and the benefit is that the content of your pages don't need to contain specific formating at all. All formatting can be plugged into a single .css file that would be referenced in all of your pages. Don't like the look and feel of your site? Want to change link colours or font sizes? No problem. Edit one file and all pages update automatically.

Well you came here for something. Here's my number 1 recommendation. Everything else is secondary, short of actually submitting your URL to some search engines and getting a few people to link to you.

USE STATIC URLS

Sounds simple, but it isn't. With all the bloggers, content management systems and content frameworks, this is the most overlooked aspect of search engine optimization. Almost all of these are dynamic in nature, meaning, server-parsed web pages such as ASP (Active Server Pages), CFM (Cold Fusion Markup Language) and PHP (PHP: Hypertext Preprocessor) generate web page content dynamically from a backend database . Makes it easy to create and manage a website, but not so easy, if even possible, for search engines to figure out what is a web page and what isn't.

Take these two web addresses:

http://www.domain.dom/foo.php?opt=today&start=noon&search&terms=bar+foo+bbq%20racks

http://www.domain.dom/todays_bbq_racks-noon.html

Lets assume this is a directory listing on someone's web site for bar foo BBQ racks. Perhaps an ecommerce site. And both pages are exactly the same.

What the heck is that first one? That's basically what a search engine is going to ask, and will likely just ignore it altogether and move on. The second one? Why, that's a web page of course.

Most search engines will recognize everything in the first URL up to the question mark as a valid URL. So the page they will try to index will be http://www.domain.dom/foo.php. But if that page is simply a search box, that's all the search engine is going to index. Very few search engines take into account variables after the question mark, and those that do will only handle two or three variables. So even if a search engine does know a little about dynamic content, it might read the URL as http://www.domain.dom/foo.php?opt=today&start=noon&search, again, landing on a page that may not have any content on it at all.

Insulating search engines from these types of dynamic URLs will drastically improve both the number of pages they index on your site, and the relevance of the indexed pages.

So how exactly do you build static URLs on your web site? Well if considering a blogger or CMS, look for one that has this functionality built in already. One CMS that does is Mambo, but it's off by default. WordPress includes this functionality in the CVS release if looking for a blogger.

If not using a CMS or blogger that support search engine friendly URLs, you can use mod_rewrite (Apache) or ISAPI_Rewrite (IIS) to rewrite your dynamic URLs to static URLs. More on that later. But here's a real-world example of both a dynamic and a static URL for the same web page from Belchfire.net:

dynamic: http://www.belchfire.net/modules.php?op=modload&name=XP_Themes&file=index&do=top

static: http://www.belchfire.net/topthemes.html

Both are exactly the same web page. Most search engines skipped the dynamic URL altogether (Google pagerank 0/10). Only the static URL has been indexed by Google (pagerank 4/10).

Want proof? Of course you do. I love stats. Here are a few that can't be ignored.

Dynamic URLs, Belchfire.net:
Robot | Hits | Bandwidth | Last Visit
Googlebot (Google) | 2883 | 444.49 MB | 31 Aug 2003 - 22:01

After converting to static:
Robot | Hits | Bandwidth | Last Visit
Googlebot (Google) | 57161 | 3.91 GB | 30 Sep 2003 - 23:57

The very next time Google indexed Belchfire.net after converting to static URLs, it indexed over 55000 pages that it did not index just 30 days before. These numbers are consistent both before and after moving to static URLs. Similar results from other search engines such as Inktomi, Alexa, and MSN.

Why is this so imporant?

Bottom line is simply this... The more pages and content on your web site that get indexed, the more likely it will be that one of your pages will come up in the search results when someone searches for keywords or phrased that appear on your pages, and ultimately the more visitors your site will receive.

Google encourages it's employees to work on personal projects one day a week on company time. I found this quite interesting. While the company I work for doesn't encourage this, I figured what the hell. Time to jump into this Blog thing and say something... Anything... But mostly to talk about my search engine optimization adventures.

I've quite accidentally developed one of the most popular and top-ranked Windows XP Themes sites on the Internet. In less than three months, between January and March 2001, Belchfire.net went from obscurity (just a personal website with some bookmarks and CGIs for troubleshooting DNS issues) to the top 10 on Google for the search terms 'xp themes'. Since then, the site has ranked consistently in the top 10 on every major seach engine, and I've made a number of changes to the site to further increase placement and ranking.

I've become completely fascinated with search engine optimization and what it really takes to get high rankings on the major search engines such as Google, Yahoo, MSN, Inktomi and the like. So this web site will be my outlet to publish what I have done, how I did it, what worked, what didn't work, and anything new or interesting I find in this whole SEO topic. Plus some of the tips and tricks I've picked up along the way regarding managing a webserver. Newbie webmasters take note!

I have to admit that I would never have started this Blog had it not been for a number or articles I'd read here and there that discuss how Blogging can increase one's pagerank on Google. So this will partly be a test to confirm or dispel the effectiveness of blogging and pagerank. So, here we go. Stay tuned.