CSS cloaking - don’t do it

January 5th, 2007

Have you been thinking about or doing CSS cloaking?

There has recently been chatter of Google spiders retrieving CSS files. This shouldn’t come as a surprise since it has long been the position of major search engines that most attempts to provide them with different content than users would see (cloaking) is grounds for penalties or expulsion from their indices. The downside outweighs any potential upside — just don’t do it.

Understanding the Google cache

December 28th, 2006

Google and other search engines cache pages, storing copies of them that they use for indexing. You can think of the cache as a snapshot of a page at a moment in time.

You can examine the cached copy of a particular page by creating a query like this:

http://www.google.com/search?q=cache:http://www.web1marketing.com

You can look at any page by placing it immediately after “cache:”. You will then see the time and date at which Google stored the cached copy as well as how the page appears to Google.

The purposes of inspecting the cache include the following:

  • If a page is in the cache, then the search engine has been able to find it and has indexed it.
  • You can examine what the search engines see of a particular page to verify that content is visible (and not hidden by CSS, JavaScript, etc.).
  • You can see how recently Google has taken a snapshot of that page and thus get an idea of how likely it is that recent updates have been “noticed.”

Some important caveats about the cache:

  • Just because a page is cached doesn’t mean it’s easy to find or going to rank well.
  • It takes time for some pages to be found and cached. A page that isn’t present may be too new for Google to have found and cached. However, if you know your site is frequently visited by Google and hasn’t been cached in a reasonable period of time (several weeks), make sure your links to that page can be found and followed by search engines. More on this.
  • There is a (considerable) delay between content updates, caching, indexing, and PageRank changes. The date of the cached copy does not mean that the search rankings or PageRank values reflect that version of the content. There is always a lag.
  • Just because you see it in the cache doesn’t mean that the search engines are reading it exactly as you would expect. Google — and other search engines — purposefully are vague about how they actually look at and rank pages so that it’s hard to take advantage of their algorithms. The cache may draw images from your current web server.
  • A page may be found by Google in various ways. Someone else may have linked to it in order for Google to find it. Presence of a cache doesn’t mean that you have a great navigation scheme.

Note that almost all of the above information applies to the other major search engines.

Using Google Webmaster Tools

December 20th, 2006

Google’s Webmaster Tools shed light on how your website is seen by Google and do various indexing-related tasks.

Like the Yahoo Site Explorer we described earlier,  the best functionality is limited to those who can upload a file and thus “verify” themselves as site owners. Once you have done so, you’ll be able to do the following:

  • Tell Google about pages and updates using sitemaps.
  • See when the site was last indexed.
  • See what pages have been indexed and where Google found errors.
  • Examine the robots.txt file.
  • Adjust the crawl rate of your site by Google.
  • Specify URL canonicalization (http://web1marketing.com versus http://www.web1marketing.com).
  • Enable advanced image search.
  • View summaries of your site’s PageRank.
  • Examine top search queries that led people to your site.

The Google Webmaster Tools are quite useful for diagnosing indexing errors as well as providing insight into how people find your site.

Using Yahoo Site Explorer

December 20th, 2006

Yahoo Site Explorer is a handy tool for examining how Yahoo has indexed a website and what sites link back to it. It is similar to Google’s Webmaster Tools.

The tool can be used to show the indexed pages in a domain and all subdomains as well as “inlinks” (in-bound links) to the page or domain.

If you are authenticated for a particular website, you have access to much more information. Authentication requires placement of a file with a special code on that website in order to confirm that you do in fact have access to the site source. Authentication then provides several new features:

  • Monitor and manage the feeds you have submitted for the site.
  • See when crawlers last visited the site and updated the index.
  • Resubmit your feed to let Yahoo know when it has been updated.
  • See what subdomains are known to Yahoo!
  • View additional metadata about the URLs from your site.

This data can be used to check if content is being indexed, how often it is being visited, and who is linking to it.

New link-finding tool

December 13th, 2006

We have launched a new link-finding tool that helps you quickly find websites that accept link requests. Simply enter a keyword or topic, then click “Go”.  You’ll be provided with a long list of possible queries for various search engines that will return lists of targeted websites that you can submit your site to for back-links.  Because each search includes your keywords, the order of results provided by each search engine roughly reflects the value of those websites for that keyword.  In other words, the list of websites in the search results is already prioritized for you.

Start finding backlink opportunities.

Google-bombing for a cause

December 4th, 2006

A hate-mongering site is Google-bombing the keyword “Martin Luther King,” so we are defending it by pointing it to where it should: the official King Center website.  Martin Luther King.

Major search engines adopt same website indexing protocol

November 16th, 2006

As recently reported by C|Net News.com, Yahoo! and Microsoft have adopted Google’s sitemap protocol. This will eventually prove to be a boon to website owners / managers since it removes some of the obstacles to using the indexing tools.

The website indexing protocols allow you to tell search engines about the pages on the website that you would like indexed. Instead of waiting for them to find all pages by spidering your site, they can download the list of pages. This makes it easier on the search engines since they don’t have to follow convoluted links to deep content, and it helps webmasters by enabling faster and more complete indexing. This is particularly advantageous to large sites and those with substantial ecommerce catalogs that have a difficult time getting all pages adequately indexed.

Rumors abound about the difficult of using current tools to create and maintain the index files, but the combined forces of the big three search engines will help drive improvements in the technology and supporting tools.

First look at Amazon Clickriver beta advertising features

November 10th, 2006

We have just started out first campaign within Amazon’s Clickriver service. The program is still in beta, so you’ll need to apply for access.

Amazon’s Clickriver looks and feels a lot like AdWords. Campaigns contain ads (an unfortunate name for what is really an ad group) which contain keywords and creatives (ads). Clickriver creatives have the same general size constraints as ads in AdWords. Keyword matching options are the same as AdWords. You can run ads / creatives against each other as in AdWords. With the exception already noted, the Clickriver terminology for everything is virtually identical that of AdWords.

What about Clickriver is different? Most importantly, your ads will only show on Amazon.com. Also, you won’t find the incredibly rich feature set that makes AdWords so versatile and powerful: no dayparting, ad rotation options, budget allocation options, and so on. One annoyance is that negative keywords must be applied at a keyword level, not ads (think ad groups) or campaigns. This may sound insignificant if you don’t recognize the power of negative keywords and consider how this design decision can be a real maintenance headache.

Overall we are left with a good impression of the Clickriver beta, and we look forward to seeing how it performs in the weeks and months to come.

Updated Business.com PPC Program

November 1st, 2006

Business.com recently undertook a significant redesign of their PPC program. They have abandoned the old points-based system for a straight bid-for-position model. Other significant improvements include:

  • Users now have complete control of their campaigns and can make changes quickly and easily.
  • Better organization of keywords into groups.
  • Support for negative keywords.

The biggest negative of this program is the same one that plagues the soon-to-be-replaced Yahoo system: bid for position encourages bid inflation and the volatility that often comes with uninformed participants. We hope that Business.com will eventually combine the bid with the CTR or other success metrics to create more stability and to reward more effective advertisers.

Simple SEO Tricks

November 1st, 2006

There are some simple SEO “tricks” that can garner higher ranking with minimal effort, and that you can employ yourself rather than paying an SEO firm. The success of these SEO techniques is — however — predicated upon a number of other factors such as proper site structure (see our SEO site analyses) and use of the most appropriate keywords (we also provide keyword research services).

  • Write more content.
  • Use keywords in your titles tags.
  • Use keyword-rich headings.
  • Make sure users don’t need JavaScript to view all of your content.
  • Put keywords in image ALT text.

Content is critical for SEO, so if you don’t have any, the other techniques aren’t going to help much. It is also the most time-consuming of the techniques listed here.

Speaking of copywriting, good SEO copy does not sound like a parrot repeating your keywords. It should be written for real humans, not just search engines. Likewise good page titles and headings use your keywords, but in ways that are not displeasing to people.