My 10 Main SEO Takeaways from SMX East 2009

Benj Arriola — Mon, 19 Oct 2009 22:35:44 +0000

Search Marketing Expo East

In search marketing conferences, there are so much to information that it is impossible to listen to everything so everyone chooses their tracks/sessions wisely to listen to the golden nuggets of information which may not be the same golden nuggets of everyone else. Depending also on each individual’s knowledge dictates if a nugget is really golden of not. So that is why I start my title with a “My” since this is what I found to be valuable to me. Below are my top 10 takeaways from the 2009 Search Marketing Expo East in New York City.

1. Yahoo announced they no longer use the Meta Keywords tag as a ranking factor

The moment this was announced, I had my doubts, I had to double check if it was true since in our own BusinessOnLine SEO Lab, we have tested this before and we all know it is only Yahoo that reads the Meta Keywords tag. With an erratic wifi connection at Javits Conference Center in New York, I relied on my iPhone to check our SEOLabs testing sites and true enough, our unique keyword that was used only on 1 page inside the meta keywords tag that used to show up in Yahoo does not show up anymore! Initial thoughts were, wow, totally useless tag now, we can probably just delete it out. But apparently, Danny Sullivan tested it out and it was showing that Yahoo still looks at the meta keyword tag. False announcement? Faulty system? Delayed search algorithmic processing? I don’t know.

2. A page blocked in robots.txt may still appear in Google search results.

You block it in robots.txt, but it get’s linked from elsewhere, it ends up in the SERPs but with no Title or Description, just the URL. This was best summarized in a video by Matt Cutts.

Learn more about robots.txt. – The Art of SEO (Theory in Practice) Buy the book now!

3. Images will tend to rank better if they are near the title/header of a page.

While listening to Google Engineer Maile Ohye on the session ask the search engines, it was nice to see a different face than Matt Cutts from time to time and she did mention about images ranking better in Google image search and Google onebox if they are near the title/header of a page which I believe totally makes sense. There are so many images a blog or news site can have and multiple articles can appear. Just knowing which image goes with which title can be challenging, so choosing the image closest to the header tag would make perfect sense.

4. Images with captions will help rank better in Google Image search.

Again for the same reasons above, it seems alternative text (the alt attribute of the tag) is not sufficient to help these images rank well in universal search results, image search and onebox results. Maile summarizes it also well on her video on news search under the FAQs part of this presentation, starting at 11:03.

Image Database: Image and Search Retrieval – Buy this book now!

5. JPEGs will do better than PNGs and GIFs

If you listened to the video of Maile well, you should have heard this already as she summarizes it pretty well into clear cut bullet points. Maybe some of you think image search is not that important and may not give enough quality converting traffic, but it really depends on your product or service and what user intent you can spin around your images. In the presentation below by Peter Lindsey, he explains different places where images can appear, onebox, maps, shopping, image search, and various user intent like inspiration, shopping, research, health, and very interesting, as a visual dictionary. He explains how in image search it is not all about being number 1. Image searchers go through many pages before they find the perfect image to click on.

Image Databases: Search and Retrieval of Digital Imagery

6. Good Web Server and Scripts Security helps in SEO.

In the session Ask the Search Engines, there were representatives from Google, Yahoo and Bing. And the Bing representative, Sasi Parthasarathy talked more about network security, the importance of server hardening, and making your scripts secure. And he showed many examples of queries he just did the day before of various websites that were compromised where either new pages are coming our selling unrelated products like pharmaceutical products, or similar pages on spammed forums or blogs on .gov or .edu websites and more that can be harming the SEO efforts or simply hurting the reputation of the site that has been attacked. This was not really like new information to me, but it was a big reminder how many blackhats out there and that they are really many of them that could be harming your own ranking.

7. Do not use English keywords and simply translate using tools for international SEO.

For international SEO for other languages, and the lack of robust keyword research tools in other languages, some SEO practitioners use English keyword research and simply translates the words using online tools and thinks these keywords are already the best keywords for another language. Andy Atkins-Krueger who is one of the popular authorities in international search has explained this pretty clearly where a direct translation may not really mean the same thing in the translated language. It is always best to get a local speaker of the language and do the appropriate translation for you.

8. SEO factor to possibly consider: Readability

This was not from one of the SEO sessions I attended, but this idea was mainly from one of the exhibit booths, specifically from Bruce Clay who we all now is one of the pioneers in the industry and has been mapping out the search engine relationships from the very beginning and keeps on updating them and is now getting less and less confusing as the main players in the search industries keep decreasing. Going back to the topic, as I was spying out checking out their tools on how it can benefit me, everything seemed like I can find a tool or some equivalent tool I already know that I can use instead of using Bruce Clay’s SEO Tool Set, although one thing that I noticed that I have not seen in other tools is they take note of Readability.

Readability tests, readability formulas, or readability metrics are formulae for evaluating the readability of text, usually by counting syllables, words, and sentences. Readability tests are often used as an alternative to conducting an actual statistical survey of human readers of the subject text (a readability survey).

- Wikipedia definition of readability

There were a few scoring systems used which I can’t remember which ones were on the SEO ToolSet, but upon search on wikipedia, there are a lot of scoring systems:

Accelerated Reader ATOS
Automated Readability Index (ARI)
Coleman-Liau Index
Dale-Chall Readability Formula
Flesch-Kincaid Readability Test
Fry Readability Formula
Gunning-Fog Index
Lexile Framework for Reading
Linsear Write
Raygor Estimate Graph
SMOG (Simple Measure Of Gobbledygook)
Spache Readability Formula

After a simple search of online tools, I was able to find two tools that can be used: SMOG online tool and using Read Able. Now would this be a critical search engine ranking factor? I am not sure, my gut feeling says it is not that critical but could still be a good factor for other purposes. Readability is a metric for showing how easy or how hard some text could be read by some age group, so it does make sense to optimize this for your readers, and not really for search engines, which is what search engines want you to do SEO anyway.

9. Google is getting better and better in reading JavaScript.

We have been reading this is various places being said by various representatives from Google in various ways, so I am convinced Google is pretty smart already in understanding JavaScript. In this specific situation where I was reminded again at SMX was during one of the sessions where Maile Ohye gave an example of using the Link Rel=Canonical tag showing various international sites in folders and using the link rel canonical to point all to the English version. I then asked if we can use geolocation detection by IP and run a JavaScript redirect so search engine would still look at the final English version as dictated by the Link rel canonical but users will experience something different depending on their geolocation and JavaScript will kick in to serve the appropriate page. I was given the answer that just because it is a JavaScript redirect does not mean Google will not follow the redirect. Conversation ended without a good answer since they asked me to ask this in the International SEO session instead. But with an answer like that, again confirms again that Google is really reading JavaScript.

10. Google is proposing the #! in URLs for crawling AJAX sites using a headless browser.

In the session where I was speaking more on CSS and code positioning and duplicate links order, there were two other Google representatives, Bruce Johnson and Kathrin Probst talking about crawling AJAX. They are proposing a new way to make AJAX sites. AJAX developers all know how convenient it would be to use the # symbol in URLs as a parameter since you can pass this around in the URL and the page does not even reload. You use it like a GET variable, but ideal for AJAX since AJAX sites do not require page reloading. And search engines normally will not see this since URLs with different text strings after the # symbol will all be treated like the same URL without the #.

Normally websites would treat these URLs as the same URL:

http://www.example.com/path/file.html

http://www.example.com/path/file.html#parameter1

http://www.example.com/path/file.html#parameter2

http://www.example.com/path/file.html#parameter1-valuex

And since the # parameters cannot be read by server-side programming languages (PHP, ASP, JSP, etc), and are read on the client-side by your web browser and JavaScript, search engines would not even see the changes on the page. So Google proposes something new, the use of the #!.

Now Google was proposing the use of a headless browser. This is a browser with no UI, but mainly like some script interpreting JavaScript code just like how a browser does. That way it will first fetch a page with the # parameters and render it first in the headless browser before Googlebot crawls the content. As this is only a proposal and is in experimental stages, to trigger the headless browser, you simply use #! instead of # in your URLs.

Being a developer myself and was sitting in that same session, I was curious to know how will we determine if Googlebot crawled the content correctly, and I asked if Google cache could be used. Kathrin replied no. And said they might make a separate tool for this. Maybe they would include it in Google Webmaster Tools. Vanessa Fox also added Google does crawl Flash links and Flash content but Google cache is also not a good way to see what Googlebot sees.

Interesting that AJAX for SEO is starting to get easier and easier and no need to do some complicated tricks once this headless browser crawling becomes a standard. This is definitely something the web analytics team would like to look at deeper for tracking AJAX.

Again this is “my” top 10 takeaways. Of course someone more interested in PPC, or SMO might have something else as their main takeaways.

After being on vacation in the Philippines during Typhoon Ondoy, speaking at SEMCon 2009 and SMX East 2009, I come back to work after 2 weeks and suddenly plays catch up with all the work that has been waiting for me and that kept me busy for the whole week that this SMX post is way overdue, but better late than later.

Bookmark to:

BusinessOnLine SEO Blog - Search Engine Marketing techniques for 3rd generation SEO» Readability