Search Engine Optimization Blog – SEM News & Opinion

I saw the other day in my Adwords account that Google now offers the ability to connect your Google Base with your Google Adwords. I thought this sure is cool then I noticed that they require you to turn on Data Sharing in your Google Analytics. Every new thing they do seems to be some way for them to get your data. What are they doing with all this data?

I have put a lot of thought in what is in the future for Google. They are so big that there is serious talk of making them a monopoly. I foresee in the future that there will be a point when bandwidth will be unlimited or near unlimited. It may not be in the next 10 years but it could be in 20-30. I have read that scientists have successfully teleported data. At some point bandwidth won’t be an issue. This means that we become a completely wireless world. Cable companies, phone companies, cell phone companies will lose their monopoly.

Google is working toward managing all data that exists. When all data is on the Internet they will be entrenched as the company that knows how to find everything. In the future you will be able to set up an Adwords account and be able to advertise anywhere you want. They already have ads for TV. Not many people use it but when all TV is on the Internet Google will have worked out the bugs and have the advertiser base to handle it.

I was watching a video of Marissa Mayer from Google. There is a lot of good stuff in there but I wanted to cover this one part.

Google does a lot of A/B testing. They asked users if more results per page would be better and they said yes. Google went and did some testing to see if this was true. They found out that traffic went down by 20%. It was not because people were finding what they wanted it was because of web page load time. Remember here we are talking Google who return results in 0.16 seconds. She said returning 30 results only doubled the time so worst case it was still less than one second load time. She did not say when this test was done so it could have been when there were more people on dial up. Even if it was Google still loads very fast on dial up even 5 years ago. She then talked about Google Maps and how when it launched it was 120k in size. They reduced the size by 30% and they got 30% more maps requests. People want content fast.

Just about every site I have had to do an SEO evaluation for is over 500k and I have seen several that were 1mb. People need to understand that websites are not a TV commercial or a magazine spread. On TV and magazines you get the information instantly. You do the same thing on a website and people will be just sitting there for a bit. If a site takes longer than a few seconds to load people will hit the back button.

I have seen websites with 100k just in text from javscript, css, and html. Then there are the images. This is where things get out of hand. I had one client that had put huge .bmp files on the front page. This is an extreme and I have only seen once. The point is scale down your images and remove some altogether. Only use css that you are going to use on that page. Don’t load the css code for every page on your site for every page load.

People also need to calm down on the javascript. Very few people know how to code javascript very well so most webpages just have a script they downloaded and they have no idea what it does. Also stay away from the rapid deployment asp.net stuff. A hello world page can be 100k or more. That stuff is designed for intranet web apps. It should never be used on an active public website.

If you really care about people spending time on your website make it fast. People want information and accomplish goals they don’t want a bunch of pretty pictures and fancy navigation. Your goal is not to impress your boss or friends it is to keep people on your website. The biggest mistake that people make is they show their website already loaded to a friend or boss and ask them what they think. That is an invalid test. First off the person is only looking at it like they do a TV commercial or magazine. That is not how they will look at it if they found the site by themselves. There is a big difference. Most people are in a big hurry very few people spend a ton of time on the Internet. Granted there is always an audience that does want all the bells and whistles and is willing to wait for the load time but they are in the minority.

One of the thing I run into the most when evaluating a website for seo is the same title on every page. This can be fixed with a little bit of code. If all the pages are static you just go to each page. If the page is using a header file then you have to make some changes. If they have a lot of static pages with an include file for the header you are still going to have to go and change every page. If you have a lot of data in a database you can just make some changes in the header file. Here is the php code to fix the problem.

Website using header file on a bunch of static pages.

Include File:

Static Page
$title = "Unique Title";
include 'header.inc';
?>

Your still going to have to go to every page and put that on it. This allows you to keep a header file that you can change in one place. If your site is completely dynamic your going to need to add a title field to your database and populate it with Titles. Then use the same if statement on the title tag but echo the title from the db.

Google has announced that you can tell them what URL is correct. This is very helpful for enforcing URL’s. The new feature is called

Google says they treat it as a “hint that we honor strongly” I still believe that you should enforce url’s yourself programaticly using 301 redirects. One of the reasons for having nice short URL’s that contain Keywords in them is to make it easier for people to link to you. It is also to make it easier to email your link and post links in chats or forums. You don’t want people using www.tinyurl.com for your links because you are now advertising tiny url and not your site.

When you build a website you need to completely understand how a search engine looks at you. Never let a search engine make any decisions. When a search engine comes to your website you need to know exactly how it is going to crawl your site.

Pages should only have one url. If you have a database driven website it is very common that the same content can be pulled up with an unlimited number URL’s. You have to enforce your url’s. Your content needs to be tied to only one url. Put logic into your site so that if it detects that content is being pulled up by the wrong url that you 301 redirect to the correct url. All URL’s should be consistent. The best way to do this is to make every url lower case. If you use a windows server you need to be extra careful because IIS will send you to the same content and produce duplicate content.

I typed HYIP into google and saw a bunch of adwords advertisers and the first result was a deffinition by Wikipedia (of course)

“”A high-yield investment program (HYIP) is a type of Ponzi scheme, which is an investment scam that promises an unsustainably high return” The ads were obvious scams and the first organic result was a big warning. The people bidding on “hyip” are fishing for really stupid people. It just amazes me that people are that stupid.

Shouldn’t google adwords watch out for scams?

The Internet marketing industry is buzzing about the possibility of Google switching to AJAX based pages. This would break many programs that scrape Google. Many people scrape Google for many reasons some harmless and some very malicious. Every time a visitor queries Google it costs them a very small am amount of money. This does not seem like a big deal until you consider how many people are doing it and how often.

This will probably only stop armatures. Scraping Google is so profitable that the pros will figure out a way and fast and some may already use tactics that are not affected by it. I’m sure that scraping Google has gotten so out of hand that this switch would save Google a lot of money.

It appears that the first step in making flash content spiderable Google has added the ability for googlebot to execute javascript. I’m not sure how robust it is but I do know that it can. I’m working with a client that had a very large amount of content that was previously not in Google because the content required javascript to be seen. Out of nowhere this site has over 600,000 pages of new content in Google that was previously not there.

I was very confused at first. I knew that those pages were not there before and that Google should not be able to see them. I put one of the url’s in a spider simulator and the text was being shown. I put it firefox and IE with javascript turned off and content would not show up. I put in in lynx and the content did not show up either. I wonder why and how the top spider simulators are running javascript.

I was doing a search today on a product name and I noticed that one of the Adwords results had an address right below it. It does not do this every time. Sometimes is just says Houston, TX.

When I did a search on google for “houston accounting software”. I saw 10 results compared to the one to three they normally show.

I was reading an article on the front page of Digg today that talked about how people spam social media sites to get more links. This is not true. The real reason is that sites like www.digg.com are search engines themselves. I discovered this one time by having an article in digg that did not get very many diggs but for some reason still to this day gets traffic from digg. What the spammers do is submit tons and tons of pages just so they show up in searches for all kinds of long tail search terms. It also helps to keep doing it so that they show up on upcoming searches. I wish Google was that easy to game. A lot of people don’t realize that digg is an awesome search engine. I use it all the time to find stuff.