SEO from Scratch

For a change I'm getting to work on a search engine optimization project the proper way. My firm has been hired to do SEO for a start-up company. What's nice to see is this start-up is making search engine optimization part of the project plan right up front.

Before they even engaged a design firm, they had identified their SEO needs and started working with us to help define the requirements that the design firm must meet. Now SEO is part of the entire project plan.

I'm looking forward to not trying to retro fit SEO into poorly designed page templates, or trying to fix a site just after it was launched.

I'll be posting more about this project as various steps have been completed. On of the challenges faced my start-ups in the SEO world, is that of longevity. The search engines give some wait to how long a domain has been around. So instead of keeping everything under wraps until the site is launches in 2008, or just leaving a standard "Under Construction" page of a hosting company's parking page as a place holder, we launched a subject matter place holder page, which has a few keyword phrase in it. What will this do? First, it helps establish the site in the search engines eyes and secondly it starts telling the search engines what this page/site is about. So take a look at the place holder for 479 Popcorn to get an idea of what I'm talking.

More on 479 Popcorn as the next steps in SEO project are completed


Google's Market Share US

Overall the number of searches in the US in October increased 12 percent from September to 10.5 billion. Google's market share continued to increase to 58.5% or 6.1 billion searches during this period according to comScore.

Just remember that all searches are not created equal and these numbers include searches on such web properties as YouTube and other search engine web properties. What does this all mean? If the experts are right, then the US search market is destined to resemble the Canadian market where Google has well over 80% of all searches. In essence, Google is becoming a monopoly when it comes to search.

Here are the other major search engines' market shares for the US in October:

  • Yahoo - 22.9%
  • MSN - 9.7%
  • Ask - 4.7%
  • Time Warner (AOL) - 4.2%

To more information please see the Yahoo News article


WebTrends 8.1A Success at Last

After numerous attempts to get the new WebTrends version 8.1a working on my test server I finally succeeded. That's the good news, what I had to do to get here is another story and the bad news.

First for those who've been following my blog posts on WebTrends can tell that I haven't been happy with the new version of WebTrends. I've been on the phone regularly with WebTrends technical support staff and just couldn't get it working the way I wanted. As someone who installs, configures and manages WebTrends for clients across North America this has been disappointing to say the least.

Since WebTrends 8.1 release back in August I've had a few opportunities to do clean installs on new servers. They went relatively OK as long as we got all the setting right the first time. If the client had first tried to install WebTrends and then I needed to reinstall to add missing components that's when another set of nightmares began. But these nightmares got me on the road to figuring out the good clean installation.

So now back to my test server. On several occasions, I had WebTrends 8.1 working, but speed was an issue and it kept crashing. In the end, I was one of the problems. This is a test server and I was using Windows XP Pro as the OS. Because of the version of IIS isn't a full version and only allows 1 web site to work at a time. I had previously used this instance of IIS to host my test SDC server. This was problem number 1. The SDC component even when it wasn't operation seemed to cause a conflict. And when I'd switch between them WebTrends would crash. So I deleted the SDC component and WebTrends ran, but it ran slow.

My WebTrends test server has been around for a while starting with version 7.0 and has always gone through upgrades. It was the 8.1 upgrade that ultimately was causing me problems. To get WebTrends to run properly, I latterly had to uninstall the previous version of WebTrends, delete all related files (OK simply rename the directory) including the instance of MySQL and start from scratch. WebTrends is now flying on my test server and I haven't had a single problem in days. That's good news for me, but if you have an existing WebTrends installation and need to upgrade you're not in the position to blow it all away and start over. So proceed carefully, do a complete server back up and be ready to revert if you run into problems.

I'm starting to like the new interface despite less than a handful of things that annoy me. I'm still waiting for at least one more round of patches/fixes with this version before recommend to clients to proceed with the upgrade. However, if you're starting from scratch there is now no reason to stay away from version 8.1a of WebTrends.


Log Scrubber Version 2.0 Released

My company K'nechtology Inc. this past Monday released version 2.0 of it's popular log scrubber.

For those who may have not seen previous posts about the Log Scrubber here is a brief overview. The Log Scrubber allows for the optimization of both Apache and IIS access log files. What does optimization of log files mean?

Log file optimization simply means removing unnecessary hit (line entries) from the log file. This allows for the size of the log file to be significantly reduced and speeds up the processing time your favourite web analytics tool.

For example, do you really need to report on all the .js and .css file calls in your log file? What about those up-time utilities that are hitting a web page every 60 seconds that you are filtering out anyways from your reporting. By removing these entries uses have typically reduced their log file size by 25% and in some case even more.

This is a big deal especially if you process your log files multiple times. The time savings can really ad up. As well, if you're using a product like WebTrends that charge you based on page views, you can remove specific page views before WebTrends process your log file and now save money as well as time.

The Log Scrubber is still priced at $199 until the end of the year and there is a free 30 day trial version also available. For more information on the Log Scrubber please visit the Log Scrubber product page or read the Log Scrubber press release.


Verifying your PPC Efforts

I'm currently giving the product PPC Assurance a trial run. PPC Assurance is a product from Enquisite that makes it easy to verify that all the clicks you're getting and paying for from your Google Adword campaign are legitimate.

According to their web site up to 10% of clicks you get charged for you shouldn't be charged for. Now if you're running a small campaign that may be just a couple of bucks a month, but what if you're spending thousands per month. In this case it can equate to 100's if not thousands of dollars each month and over the course of a year, we're talking big bucks.

So with a new client SEM campaign launching last week I decided to give PPC Assurance a trial run. The set-up looked simple and it should have been. However, my expectations and requirements didn't fit into the standard business model. So a few emails and phone calls later I was up and running with a clearer sense of how this product works. Knowing what I now know, it should taken less then 15 minutes.

PPC Assure basically sits on top of your existing Enquisite account (more on that great product another time) and measures every click you get from Google Adwords. With an API into Adwords itself, PPC Assurance verifies that every click you're being charged for match your campaign criteria. It then reports on inappropriate clicks and even has a mechanism for claiming refunds from Google.

Is PPC Assurance for you. The answer is no if you're running a Global campaign with no day parting settings and no audience targeting (language, country, etc.). However, if you like most of my clients running specific ads targeted at specific audiences (country, city, state/province, etc.) then PPC Assure is something to look into.

So far my trial test is going great. The campaign is set to only display ads to people searching within Canada in English. According to PPC Assure, about 8% of ad clicks are from outside that definition and that Google is charging for a little less then 1/2 of those (about 2.5%).

I'm going to run PPC Assurance for a whole month and then try the claim mechanism to see how it works. Keep looking here for more updates on this trial evaluation.

For more information on PPC Assure visit their web site at http://ppcassurance.enquisite.com/

SEM & Web Analytics

Here's an interesting article on the getting the big picture of Search Engine Marketing lined up correctly with web analytics. Basically it boils down to the problem of different camps and different responsibilities and objectives.

If you have a few minutes give it a read at http://semphonic.blogs.com/semangel/2007/11/search-engine-m.html