Sense amongst Madness - Wit amongst Folly

Here are a few of the essential newsletter for SEO and Marketing that I enjoy every issue and learn something every time. These are the cream of the crop.

Grokdotcom by Future Now. Bryan and Jeffrey Eisenberg are the Gurus of Conversion. Their bi-monthly newsletter is like sitting down and picking their brains for an hour or so. This latest issue they have a blow-by-blow analysis of a non-profit website, with a running commentary on what is working, what isn’t, and what should be done at every step of the purchasing process.

Larry Chase’s Web Digest for Marketers. This weeks issue is edited by Mike Grehan, on 11 Insider tips to improve Search Results. I find this newsletter has a little too much advertising for my taste, but still, always a good read.

IR Watch – the Newsletter. . This give newsletter by Information Retrieval expert Dr. E Garcia (Orion at searchenginewatch.com) gives all the gory technical details of SEO. From his site, “Tired of SEO speculations? Our email newsletter can change that. Not fancy, but to the point. Get it one month in advanced. Stay ahead of the curve and at the pace of computer scientists, IR students and advanced search engine marketers.

Read about search research that normally does not reach mainstream. Includes optimization tips, programming advice, software/books we recommend you to buy, and monthly highlights from our blog.”

Marketing Experiments Journal. These guys are great! Practical, down to earth, advice from the experts. Free web clinics that are jam packed with information, podcasts, blogs and more. Marketing experiments recently purchased Marketing Sherpa.

Working with affiliate sites, I am continually confronted with the problem of “What should I do next?” As a small business person, the answer is you do whatever will make the most money the fastest. Pick the low hanging fruit first.

Assuming that you have a content site already, and you are adding more content regularly, submitting content for syndication, adding quality links regularly and consistently over time. So you have say, 250 products, ranging from very competitive to almost no competition. How do you decide where to focus your SEO and prioritize your efforts?

The low hanging fruit here are the product related keywords that have a relatively high search volume, and a low level of competition.

One measure of the competitiveness is the Keyword Effectiveness Index (KEI) which measures the search volume and the number of sites competing for that phrase.

The formula for KEI is p^2/C

Where p= Popularity – Wordtracker searches
C = Competitiveness, measured by the search volume (number of results returned in a search on Google with the search phrase in quotes)

Open an Excel spreadsheet and put your keyword phrases down column A, column B will be the Wordtracker searches, and insert this formula in column C

=((A2*2)/A2) *1000

and use FILL DOWN to populate column C.

Start filling in your data down the list. Quickly you will see some keywords have loads of competition, while others are open season.

Sort your data by column C, the KEI, and you have a prioritized list of keywords to target and can develop your plan of attack. Higher KEI is better.

The high KEI phrases can be targeted immediately for fast results, the medium level KEI keywords, can be started now for a longer campaign, and the low KEI phrases, either not targeted, or a long range plan developed for them.

Problems and issues

Determining the competitiveness of keywords is a complex issue and in this post I am looking at just one method — a more technical discussion with links to other methods is at Search Engine Watch Forum.

A few problems are that KEI is a mathematical ranking system that says nothing about the quality of competition.

A modification to the formula, proposed by Dan Theis has taken the original KEI formula and tweaked it:

s = matches for “allintitle”
p = popularity (search count)
KEI = p^2/s = p / s * p

One of the factors Google uses to ranks sites, from among the reputed hundreds of factors, is the breadth of a sites’ content.

The breadth of a site is a factor in Google’s algorithm for several reasons. Google uses some form of Latent Semantic Indexing, LSI (or so I contend). LSI is a complex and exceedingly clever way for computers to understand what text is about using mathematics. This is a gross over-simplification and other people have discussed this issue and all the technical details for a different audience. Nevertheless, it serves our purpose.

One way to think about how the algorithm functions is first the keywords with the highest density are removed from the page. Hence talk of an ‘over-optimization’ penalty. With the highest density words removed, the algorithm attempts to make sense of the page with the remaining text. If the webpage text contains a complete and broad set of related keywords, the meaning of the text can be extracted by a machine.

Another factor in Google’s algorithm that broad content triggers is click tracking. Google is known to track clicks on search results occasionally. When a browsers does a search and clicks on one of the listings, a timer starts. If the browser visits a site, and doesn’t find what they are looking for, they hit the back button and choose another listing to click on. Google tracks this short visit, co-related with the keywords and the site and determines the site is not relevant. If a browser clicks on a search result and don’t come back to the search results, Google reasons that the browser must have found what they were looking for and the site is relevant to the keywords.

By creating a site that covers every aspect of a topic increases the chances that you will have something of interest to browsers using your targeted keywords.

Examples of Broad Coverage

A search for “art” on Google shows the top site is art.com and the next is art.net
A search for “baby” — the top site is babycenter.com and the #2 is baby.com

www.art.com is a great site and a great example of a site that covers their topic broadly. A browser searching for the keyword “art” may have many things in mind, but if they go to art.com, chances are they will find what they are looking for.

Consider some of the things that someone searching for “art” may be looking for:

1. Different types of art: posters, tapestries or canvases, etc.
2. Different styles of art: modern, still life, advertising, photography, etc.
3. Different artists: Picasso, Monet, etc.
4. Different Schools of art: Impressionist, Realism, Romanticism etc.

Consider how this same principal applies to smaller sites targeting 2 and 3 word keyword phrases.

Tips for Making Sites with Broad Coverage

1. Keyword Research. Find variations of your targeted keyword phrases and create content and add affiliate products to increase the breadth of your coverage.

2. Check your competitors. Visit the top 20 sites for your targeted keywords and make a note of the categories on their sites.

For more tips on creating a content site with broad coverage HERE

Duplicate content is a serious issue for most SEO, and while I can understand the tough line that Google takes, it seems to me they are getting a little carried away — but then they probably have little choice. If they didn’t take such a hard line the amount of dup content out would grow exponentially.

Not only is duplicate content an issue between sites, but also internally, similar content. Pages that are too similar to each other are penalized. This is a vexing problem with hotel sites, real estate sites, and a whole host of other sites where the essential product is the same, only the specifications are slightly different.

Using a duplicate content checker like CopyScape will not detect similar content, only duplicate content. Copyscape checks for plagiarism, which is an exact copy. Presumably they extract a snippet and search for it in quotes to find another indexed copy or something like that.

According to Google SpamCop Matt Cutts, the pages have to be quote VERY DIFFERENT unquote. See the full collection of his SEO talks here. According to Cutts, Google’s algorithm doesn’t just check the pages once, but has a whole series of checks where pages are tested right up until milliseconds before the results are shown.

The details of duplicate content, and similar content, are complex and interesting. A full discussion is here at studtdubl.com.

Without getting into all the technical details, which others do very well for a different audience than here, I would like to present a few tips and pointers for fixing similar content.

Here are some tips for making content VERY different:

Use a duplicate content checker HERE and try getting the pages below 20% similar.

In fact, the duplicate content algorithm doesn’t work on a percentage basis — nevertheless, if you can get it down to a low level, (i.e. below 10 or 20%) you can skate through. (Oh right — 40% is not a low level!)

Test all your pages with the dup content checker and make a list. Probably the pages say the same thing or use the same way of speaking about the product. Things like, “all of our materials are prepared by… ” Or, do an “Extended find” for the text using an HTML editor. Likely the phrase that is repeated is part of a paragraph that you used over and over when building the site. Count the number of pages.

Depending on how long the list is, there are 2 ways to go — adding new unique content or rewriting the existing content, or both.

First, look at the paragraph and write 20 versions of the paragraph that all say the same thing, but are totally different. If you really set your mind to it, it isn’t that difficult.

Next, research and write new unique content for each page or small group of pages. Depending on how long the text is, you may only have to add 100 or 200 words to dilute the existing text enough to squeak by.

Next mix and match the 20 unique descriptions with the new text.

Sage Luce, President of SageRock.com shares his thoughts on what to expect from a brand new search engine optimization campaign.

1. A brand new domain will not come up on Google for 6 months – 1 year.
Absolutely — it is a long hard slog, and it is all about quality and content.

2. Getting good positions on Google is harder than ever. You must have good quality links and you have to have a good site to get good links.
Quality is everything now. Who you link to and how you link to them, and who long you link to them — plus who links to them, and how long.

3. You have to have lots of content — you have to be a resource and offer something that is unique. Absolutely. What separates you from the pack? Why should anyone visit your site when there are a million others to choose from?

4. You have to use phrases that you want to rank for in your content. If the keyword phrase you are targeting for doesn’t appear on the page, you won’t rank for it.

5. It takes as long as it takes. Every industry is different, and every website is different. The older your site, the easier it is, the newer your site, the longer it will take. I tell client expect 2 – 3 months for MSN, 2 or 3 months after that for Yahoo, and a year for Google. Maybe we will get a pleasant surprise and it will be earlier, but coun’t on a year.

6. Stop the multiple domains — this makes your project exponentially different. Have one site and focus on it. As each domain gets tougher and tougher to place on the first page, it is less and less attractive to operate multiple sites.

7. You have to have good content and lots of it. Some people even say SEO is dead and it is all about content optimization now. Maybe that is a little strong but they do have a point. Content is very very important.

8. Trust your SEO for a least a year. If they have done their keyword research, they are building links, they are building content you are on the right path. Give them 365 days.

After all that, he does say, the ROI of a good position on Google is better than any other advertising medium.

See the Full list of videos from Sage Rock

Comments? Suggestions?