Monday, February 26, 2007

Google Webmaster Tools Query Stats

I am not sure how long this feature has been around, but I was just scrolling through my list of sites and wanted to see the recent query stats on a few of my sites. I noticed that there is an option to see query stats by search location. This is great, now for foreign sites I can actually see exactly what’s going on within that country.

This seems especially helpful if you are having a hard time getting rankings up in a specific country. In my case trying to rank sub domains from a .com site hosted in the US.

Labels: ,

Friday, February 23, 2007

Fear of Personalized Search and SEO

There seems to be a general fear of how personalized search is going to affect how SEO works. What strikes me the most is the common unwillingness of SEO’s today to advance their abilities to cope with change. I am very intrigued by the sheer volume of posts that appear daily about every little bump in Google. As an SEO community we need to move beyond this type of thinking and adapt to the future of how search will work.

One thing is for sure, Google is not going to risk the user experience of Google Web Search. It is also blatantly clear that Personalized Search will only be offered to signed in users. How else can they appropriately track search behavior to encompass a proper result? Keep in mind that cookies are slowly going away, many users are turning off accepting cookies for privacy.

I see the field of SEO becoming a strong part of user experience enhancement. If we can help to make sure sites are highly accessible and are attractive to users we will perform well in the SERP’s. It is sad that there are still folks out there creating garbage sites just to drive revenue and not fill a particular need. We are going to see a huge shakeup of these results sooner rather than later. Which will open the doors for sites who deserve to be exposed moving up and taking the place.

Labels: ,

Wednesday, February 21, 2007

VC’s Investing in Startups

A very interesting scenario has come to mind that is not and old one. If you have a great idea and are able to get VC funding, one of the major requirements is a proper business plan.

A serious question that should be at the top of your mind is how are you going to drive traffic to your newly derived startup? Putting up a billboard in Iowa and expecting to see traffic is almost suicidal. So why would you omit the line item regarding traffic generation?

Yes, SEM is a great way to drive traffic to new sites. We know that, its something that has become a standard in the industry. In some cases it may not be possible to have a profitable business with SEM alone.

So just a thought, while budgeting you may want to consider a line item for SEO. Especially since SEO can drive so much GM it should not be a small amount of money either. Take a few minutes and call a few consultants and agencies for appraisal. You may be surprised how much more is involved than you first thought. It’s one of the reasons we are in business.

Labels: ,

Tuesday, February 13, 2007

When is SEO just too expensive?

Now this may sound funny, but I think it’s interesting to look at the cost of production of some of the largest most powerful websites out there. As an SEO I tend to judge huge sites radically based upon their ability to optimize code. We may think well let’s work some magic here and do a URL re-write or let’s optimize the way the templates are generated. But the most common thing that I tend to neglect is the overall cost!

Did you know that a simple code change, not a complicated URL re-write on a very powerful site could cost as much as $300k? So now that the company has paid for their consultant to come in and tell them what to do, they still have to fork out this much in resources. It’s quite a humbling thought.

This includes, engineering time. If the engineering teams are setup on certain groups or resources there is a certain cost associated with each hour of his or her time. That cost is then multiplied by the number of QA engineers associated with the testing of the code and then the user front end testing required to make sure that the user base is not going to freak out. This is before it’s even pushed to production and taking into account the simple cost of copying the code from staging, to production and then testing to make sure everything is live and happy.

I think back to some of my past articles, like its 2007 and most ecommerce sites are not SEO optimized. Yes the big companies have a large risk, we think its small but the cost is quite substantial. There still is no excuse for the smaller sites that can make these tweaks and it may cost them several thousand dollars in engineering time.

Ok this is a result of my brain going in crazy directions today with too much caffeine.

Labels:

A few suggestions for sites that carry classifieds type content

A question that I just responded to on webmasterworld.com, brought back some general questions that I have spent a lot of time theorizing around this year. How do you handle content that is expiring quickly?

First just getting the content indexed in a reasonable amount of time is very difficult, if your ad lasts anywhere less than 5 days this may be impossible. Especially if you delete the ad! I would recommend not deleting the ad and rather spending more time figuring out how to get a user to an ad that is similar and relevant. How you ask? Use a search engine that will be the cleanest way to index and catalog your available options.

Providing users with an alternative method to finding additional content is always helpful. But it’s critical that the options you are providing them are relevant! The simplest method would be to use a search appliance and ask it to provide the top 5 matches for other products or classifieds on the site that could be a good match and to display them. Thus when the search engine gets around to indexing this content it will have the ability to find related products to index and give the user a place to go.

Nothing is more annoying that searching for something and landing on a useless page. Google knows this and does whatever it can not to display such results.

From an indexing point of view a rapidly changing list of pages can be a nightmare as well. For example if you have a category for used tv’s on your classifieds site that has a next prompt at the bottom of it. The search engine assumes that the user will have to click “x” amount of times deep to get to the product at that given time. Make sure you display all pages in a numbered fashion above and below the list to make it easy to find. Then every result is only one click away from the primary category page.

These types of sites are also great candidates from coming up with a Google Sitemaps solution. You can sure that Google will pick up these pages at an expedited rate IF they are refreshed constantly. You cant generate a sitemap once a month and expect this to work.

Labels:

Thursday, February 08, 2007

Google’s -950 Penalty

Tedster from Webmasterworld.com has posted a very interesting theory about the -950 penalty and a direct correlation with a seemingly coincidental patent that was recently released. I agree with his logic about this particular patent having a unique effect of the rankings.

However, if you read through this patent it seems to me that it was designed to detect and penalize content generated from basically a content generation system. The sole purpose of these content generators is to mix-up content and make it seem themed and relevant for search engines.

The most interesting example of this is the following “An example of the cluster bit vectors are as follows, using the above phrases: TABLE-US-00001 Monica purse Cluster Bill Clinton President Lewinsky designer ID Bill Clinton 1 1 1 0 14 President 1 1 0 0 12 Monica 1 0 1 1 11 Lewinsky purse 0 0 1 1 3 designer

The first thing I noticed reading this line is that the data looks like it may belong together, but the order written obviously would be generated and not logical English. Google could take this sample and run it against other known examples of content that are themed. If this particular sample is so far beyond the normal threshold it could easily trigger such a filter. The most interesting part of spam is usually it is targeted to mostly competitive subjects, thus the sample size is very large and easy to target.

We know Google has the ability to globally pattern match snippets of content and deem them duplicative, thus a filter of this nature could be pretty easy to build on top of such technology. It’s rather ingenious!

Labels: ,

Effects that behavioral skipping has on SEO

Behavioral skipping has become a pain in my side this past year and thought I would spend a few minutes sharing my pain. The logic is, if you have site that produces a page based upon a search you know what I am dealing with.

To provide a frame of reference every single page that you view is actually search result, even though the data seems pretty simple and theoretically easy to produce. But to scale according to a large database, you cannot create 10’s of millions of pages and make sure that they are all doing what you want manually. No one can afford to maintain that type of process. So instead you try and learn the behaviors of users by what they are looking for and try and assume what they want either based upon a click or a search. In this case it’s really the same thing, but the URL’s are different.

It’s amazing to see how wide spread this type of problem is across the net. You can simply go to any major retail site and type in a popular model number of a product. It can be very easy to get 100 results back for a specific ipod, especially when the retailer has tons of sub retailers that ship on their behalf.

It became obviously clear why these sites also do not perform at all in MSN, there was a post about how you can remove your competitors from the MSN index by getting duplicate pages indexed using ?value after each page. With large retail sites you do not have to work hard at this. It becomes very easy to do without anyone else’s help.

Labels: ,

My thoughts on Jason Calacanis’ post

First I would like to clearly say that I respect Jason Calacanis as a business leader and what he has accomplished in his career. No, don’t worry I am not going to start a bashing session about this. But I think it’s important to read between what Jason and Danny wrote about this subject.

I think Jason’s general synopsis that 90% of SEO’s are snake oil salesmen may be a little rough. Yes there are quite a few shady individuals out there. But more importantly we should clarify that large mistakes come from firms who have way too many employees and high turn over. The sales staff’ focus is to sell the customer and with sales in general a certain amount of BS is involved. As someone who really knows this process inside and out, this is not where the failure happens, the failure happens within the high turn over tech staff that is doing most of the work. When you provide the keys to your site to some kid who is fresh out of school and has optimize maybe 2 sites in total, the risk moves up.

In general my opinion is that a good SEO is one who has been optimizing sites for a period of time, hopefully a long period of time. This individual has made mistakes and has clearly learned from his or her mistakes. We can only grow in this industry by error itself. I have personally optimized 100’s upon 100’s of sites, some small and some large. But actually being in the trenches for the past 7 years has given me a leg up. Many of the top guys in the SEO industry have built names from themselves in a short period of time, and they do a great job. Thus the need of a lot of experience is not necessary, but it prepares you for sudden changes in search algorithms that can really shake up your life.

Danny was very right in his logic and reasoning around the majority of the industry. Almost everyone has good intentions and very few that are bad are still in business. A lot of us remember the trafficpower.com days, far and few between.

Now for my controversial view, when you take away the rest of the small sites out there and look at the big bad boys. The landscape of knowledgeable people changes dramatically. To optimize a large high trafficked website requires a dramatically different thinking process. Yes all the same practices apply, but how do you get the search engine to like all 100 million pages? This is where I feel there are a lot of snake oil firms selling. Most large SEO firms CANNOT handle ultra sized websites.

Most commonly the first things a person will experience working with a large company is all of the red tape and road blocks. It takes a lot of one on one time with the engineering staff and the capability of drawing ideas out of their staff. If you cannot get them on your side, you are sunk and the project will fail!

Of the consultants that I know in the industry, the only ones that I would throw at a large company are Joseph Morin, Rand Fishkin and Andy Beal. I have seen their reactions and how they present themselves in such a manner that can be conducive of a productive meeting. I am sure there are many others out there that I do not know, and forgive me for not mentioning you. People, who know me, also know I will never recommend anyone that I wouldn’t use myself.

Yes, I tend to write in a very scattered way, so I will say in short. I think Jason’s comments where a bit too generalized and Danny made it clear. But there is a difference between a standard SEO and a fortune 500 SEO.

Labels:

Wednesday, February 07, 2007

How the web has evolved

I was sent this video by a co-worker in Berlin today, this is a great video showing the migration from the early web back in the mid 90’s to what’s available now. I got a total kick out of watching it!

My Super Proposal

I am so thrilled for Rand Fishkin who proposed to his Girl last night, in what would have been the greatest media stunt ever heard of. For months I have been watching the details of this proposal get put together by Joseph Morin founder of Story Bids, and had been so excited that I had hundreds of people looking for the commercial at the Super Bowl. Well the commercial didn’t happen but the dream itself made it!

The worst part about this is I have known who JP was for a while now. The suspense has been killing me!

The amount of coverage this commercial got was amazing and I have never seen a story get so much traction. This is the ultimate link bait idea of 2006/2007!

Again lets congratulate Rand on his engagement!

Labels: , ,

Tuesday, February 06, 2007

Website accessibility for SEO

I think its time to bring up website accessibility and SEO again. One of the things that I constantly rant about is that pages load too slowly and are too big, causing rankings to move down. There are several factors that I have been looking at lately that seem to add additional credibility to this logic. The main thing I keep looking at is latency.

One of which is locations of Google crawlers coming into my global sites. A simple analysis that I have done over several sites that I manage concludes that if you have a TLD or an IP address from a country Google will cross the pond from the local country to crawl the foreign sites.

To add further credit to this I have examined other sites in the same industry who actually host their sites within country. These sites out rank 10 to 1 any sites that are hosted back in the US. Now I know what you are thinking, what does this have anything to do with accessibility? EVERYTHING! Imagine a dialup user trying to pull down your fat slow pages over the pond, what a terrible experience! It’s imperative that these basic things come into play with your strategy.

Here is a small list of things that I look at that seem to make a huge difference.

1). Page Load Speed, if it’s over 500MS on average its way too SLOW! Look at your Google Webmaster Tools under the Crawl Rate to see what Google see’s.

2). Page size, how big is your page? Does it really need to be that big? Do you really need 100 images on it? Are they properly tagged? Do you really need all that graphical crap?

3). Text vs Graphics, how much text is a part of your page? No not the same boilerplate that appears on every page. The unique core part of your document.

4). Section 508 elements? Do you have them? ALT tags, are a great start, this is easy stuff to add and can really help you out. Check out Target.com, they got sued for not having them.

Labels: , ,

Google back link tool, easy tool for diagnosing URL issues

As any SEO would do, as soon as a new toy is available I drop everything and start playing with it. The first thing I noticed is that this is a great way for large site owners to find URL problems. Wow what an amazing insight to huge issues! I was able to find spaces in URL’s strange tracking id’s that I had no idea still existed and trailing characters. What a great tool for my large site arsenal!

This tool is available within your Google Webmaster Tools, simply click the Link tab after you select your verified site.

Labels: ,

Monday, February 05, 2007

SEO learning's from expanding internationally

After this last year I have walked away with a lot of information about how Google reacts to expanding your brand into new countries.

My first hassle was acquiring local domains within each country:

Requirements within the UK, a Royal Postal Address, even a PO Box will do. Not a tough one.

Requirements in Germany are quite the same, a local address within Germany to acquire the domain.

France was a huge hassle, not only a local address was required but as soon as you attempted to acquire an SSL certificate the process became slow and aggravating. You can only maintain your registration for up to 1 year in France, thus if you forget to renew you could be in trouble.

The most surprising for such a low hassle culture was Australia, you need to provide a mountain of documentation and prove that you are a resident of Australia to obtain a local domain.

During this acquisition process of domains we decided to launch under sub-domains from a US hosted .com. As you may know already if you have a .com and it's hosted in the US, it can be a huge hassle to get listed locally in country. With some nifty routing and clever negotiations I was able to get IP's from other countries completely re-assigned to me locally. When you broadcast a foreign IP, it's still mapped as being from the local country thus you can be identified as being from in country. This was a very successful approach rather than the extreme alternative which was to open an expensive data center.

During this process I was also told that RIPE addresses are EU addresses and not specific per country. This is not true, and will cause your other foreign sites to be thrown in the wrong country.

While launching these new properties I also discovered that if you are in Germany and an English based site within Germany links to you, it still helps. As long as the content on the site is themed and relevant, it seems that Google simply examines theme of the links pointing into the English site, which many where German and had extremely related content.

I walked away with the basic premise that if you host and have a local TLD in country your life will be so much easier. The search engines have a long way to go to provide a simple way to handle multi country businesses.

Labels: ,

Observations on Trends on In-House SEO's

I have observed over the past year there seems to be a very strange split that is occurring within the in-house SEO group. Personally working in-house for a large company with a wide variety of SEO's, I noticed that continual research and development is a dying art. Innovation is coming form old information and logic and little time is spent reading the boards/blogs and staying up to date.

Personally I spend half my day researching and staying up to date on any advancement as it could be catastrophic if you let it go. Additionally I noticed that the return to the forums only occurs when something goes wrong, the in-house folks who spend time reading and staying up to date are commonly not asking questions when changes are occurring since they are a step ahead.

Labels: ,

Super Bowl Commercials Where Awful

Almost every year we have had some amazing Super Bowl commercials, you could remember them for weeks and they had a great buzz factor. This year to my surprise they all sucked, except the Doritos one where the guy smashes his face into the car and the chick does the same on the floor. That part was funny and cute, but that’s about it.

I cannot believe with the power of social marketing online, not one advertiser harnessed the ability to get massive link baiting or drive any sort of wild residual traffic. By the time traditional media gets a clue, we will be on to a whole new concept.

Labels:

My Photo
Name: Aaron Shear
Location: San Francisco, California, United States

I have been in the search industry since the late 90’s, no not 10-20 years. My career started early in the search Day’s at Inktomi, where I supported large search portals. For example, MSN, AOL, iWon, Hotbot, CNet too name a few. After Inktomi I became a freelance consultant. I consulted for a few of the Top SEO’s around 2002 time frame; obviously the market has changed since then. After consulting I joined a small SEO firm called SEO Inc as the CTO. At SEO Inc. I successfully optimized some of the largest clients including IGN, Sony, VEGAS.com, Beaches and Sandals Resorts to name a few. Even though SEO Inc was a ton of fun, I still wanted the ultimate SEO challenge. I moved on as the global head of SEO for Shopping.com an eBay company. This challenge was an interesting one, how do I optimize a site with 50 million products? Every month I helped the business grow by leaps and bounds. I am now consulting for mostly enterprise e-commerce clients. Yes there is more too me than this profile shows, but you will just have to ask.

My Blog Juice My status

Powered by Blogger

Add to Google Subscribe in Bloglines