It is widely known fact that Google’s new Caffeine update is ready to go live. Matt Cutts announced at PubCon that they will launch it after the holidays season in US. He confirmed that caffeine can be accessed using 209.85.225.103 (50% of the requests).

From what we have gathered, it seems like it is live in India. At least the data centers our team members have checked.

Over the weekend, I noticed that emails from Google Alerts increased 5 folds for the keywords I have set alerts. I have subscribed to these alerts for a few years now and generally get 2-3 emails a per day per alert. Yesterday I received about 12 emails per alert. This leads me to believe that Google has switched the backend for alerts from current Google to Caffeine update. Caffeine has bigger index for all the websites we have looked at. This may just be a temporary bump as Caffeine is “catching up” with sending alerts for pages which current Google has not been able to find yet. Has anyone else noticed this?

{ 0 comments }

Google Executing Javascript

by Gaurav on December 9, 2009

in SEO Advice

A lot of people don’t know that Google has been executing Javascript in the HTML pages and including the output in their index. We found this a few months back. We noticed a website with Javascript drop down menus and no static links at all had most of their pages indexed by Google.

What does it change for SEO and webmasters?

  1. Links in Javascript are counted as “regular” links – Links which are displayed on your webpages after Javascript execution will be included by Google as part of your link graph. This is a good thing because you can use Javascript to add links to improve user experience. Bad thing about this is that all your outbound links will also be treated as “regular” links. You will have to make sure that you point it to Google and other search engines that they need to exclude the internal and outbound links which you prefer. As Matt Cutts explains that you can either nofollow or exclude the link or js file using robots.txt. See video for more detail.

  2. Duplicate Content Issues – We have seen multi level drop down menus and option boxes causing Google to crawl and index duplicate content following URL patterns based on the GET strings. One of the advanced optimization technique used by SEOs for work flows is to add static links on the pages to next steps, but use a drop down menu for a better customer experience. For instance the URLs for steps may look like /step1.html -> /step2.html -> /step3.html. However, its counterpart form might generate URLs like /step1.html?v=widget1 -> /step2.html?v=widget1&x=widget2 -> /step3.html?v=widget1&x=widget2&y=submit. 

    These would certainly create multiple URLs for the same content. Solutions to these may involve adding optional parameters in Google Webmaster Tools (only for Google) or perhaps using POST instead of GET.
  3. Obfuscation may not work – A lot of website owners use Javascript to obfuscate email addresses, phone numbers or even product prices in certain cases. This will get Google index and serve (display) the obfuscated content in Google results. Matt Cutts recommends not using Javascript to obfuscate emails (or other things). He mentions that using robots.txt to disallow the js file is the last thing you should try to obfuscate. Checkout his video:

Note: Even though Google and other search engines have advanced to read links in JS, your internal (or external) links can not be entirely replaced by Javascript yet.

{ 0 comments }

After launching the regional tags, Google announced yet another update to the SERPs. They changed how images are displayed in the universal search by making the first image bigger than the rest of the images and adding two rows of images instead of just 1. The change only applies to universal search and not image search. Google claims to show the bigger picture only when they are certain about the image result.

What do you think will be impact on the CTR from universal search to the page where image is hosted? Will it increase or decrease? Will other pictures in the universal search get lower CTR? We will start testing this today. Would love to hear if someone is already testing it.

Dog Google Image Universal Search

{ 0 comments }

We just noticed that Google is showing breadcrumbs in SERPs for a lot more pages now. Google was showing the “smart links” which intelligently parse the website’s hierarchy and displays it with deep links in SERPs. This is something similar to “site links” but can be controlled by SEOs and website owners to certain extent. Google started testing this sometime ago, but I haven’t seen them on a live search before. This is very interesting and opens up a lot of possibilities for SEOs.

Breadcrumbs on Google SERPs is a screencast I created for those who are not able to find the breadcrumbs.

Update 1: Some of our client website have these too. The breadcrumbs are not generated for any and every page, but the page which get considerable amount of traffic from Google and have been in index for a while.

{ 4 comments }

Last week I went to lunch with a friend to a nice Mexican restaurant in San Francisco called Mercedes Restaurant. It was easy to find on Google and had great reviews. But I noticed that the word Mercedes in the title of the homepage was regular type face in Google search result pages instead of being bold. It took me a few seconds to realize that it was actually misspelled as Mercdes. After looking at the source of their website, I found an H1 tag on the page which was misspelled the same way as the title. As a side note, the H1 tag is for some reason not visible on the page.

Both the title and the H1 tag are important elements of SEO and creating brand awareness. To make things worse, they have both of these tags used globally for each page on the website. Google and other search engines help them by assuming (based on Click Through Rate, I guess) that their website is what people are looking for when they search for mercedes restaurant San Francisco.

This was one of the several examples where website owners don’t pay attention to detail on the copy for the website and are solely dependent on the search engines to find a connection between the misspelled keywords and the actual website.

If you have a small website with 10-20 pages, you should follow these steps:

  1. Run an automated spell checker like this on your website.
  2. Tool will help you, but you will have to manually go through the results.
  3. Specifically check Title and Meta Description (they are visible on viewing the source of the page).
  4. Go to the main search engines and search for your brand name/main keyword and make sure that the results show your website as you intend to.

{ 0 comments }

Highest paid keywords

October 20, 2009

Ever wondered what are the highest paid keyword on Google and how much would it cost? Laurie Sullivan recently wrote an article on MediaPost with some interesting stats on PPC rates for the 3 major search engines.
Highest priced keywords:

On Google was “Mesothelioma” for $99.44 per click
On Yahoo was “Mesothelioma” for $60.68 per click
On Bing was “auto insurance [...]

Read the full article →

Sitemap Autodiscovery

September 27, 2009

XML Sitemaps are a great way to provide a complete list of your webpages to Google for efficient crawling and discovery. Although by submitting your sitemap to Google’s Webmaster Tools does not mean that all your pages will be indexed by Google, but it increases chances of Google discovering more of your website’s content and [...]

Read the full article →

HTML Meta Tags Optimization

September 15, 2009

Ever wondered where to begin with when optimizing for HTML meta tags for your website? Do you think Meta Keywords help in gaining rankings? Do you think that Meta Descriptions are useless because Google does not use them for ranking? Here is a quick snapshot for optimizing the head tags on a webpage.

Title [...]

Read the full article →

“Jump to” in Wikipedia snippet on Google SERPs

September 10, 2009

Just noticed that on searching “california sales tax rate” on Google there is a Jump to “California” link to Wikipedia’s Sales Tax in United States page. It seems like Google is using the internal link ‘#’ to have people “jump” to a particular section on the page. However, not all the states have “Jump to” links. [...]

Read the full article →

Facebook’s Twitter Integration

August 21, 2009

What does it mean for Fan page owners?
Facebook’s Twitter integration for fan pages is a great move by Facebook. Most of the Facebook fan page owners till now were updating both FB and Twitter for the same content. This change makes it easier for them to just log into Facebook and don’t worry about updating [...]

Read the full article →