Here’s an interesting article on Google’s recent experiments with indexing the “invisible web”:

Google: A Clear & Present Danger to Corporate Data Privacy

My take: yes it may cost companies some money to correct exposure of their supposedly private data - but if Google can index it by the approach they are taking, it was never private in any meaningful sense. Chances are, those companies’ competitors already have copies of it.

If your data is meant to be private it is your responsibility to make it so, not Google’s.

Google Webmaster Tools is the primary source of information about how Google is indexing your site. It’s also very useful for finding problems with your site that might otherwise go unnoticed.

Here’s a list of tips covering how to use Webmaster Tools to keep your optimisation efforts on track. I hope they’ll be useful.

(more…)

I’ve just given this entire site a makeover. I got fed up with the hybrid site management I’d knocked together when I first set it up and I decided that migrating it fully to WordPress would be the best approach. At the same time I thought this would be a great opportunity to do some serious search engine optimisation, since WordPress, combined with the right plugins, would make the process much easier.
(more…)

I had occasion recently to search on Digg, and was surprised to see that one of the words on the search string that was matched was the definite article, “the”. Try it. Even for the word “the” on its own, you’ll get matches. It even matches with other useful words like “these”.

This is surprisingly dumb for one of the most visited sites on the Internet. There are two reasons to strip out very common words like “the” and, uh, “and”: first the results are unlikely to be meaningful and second by removing them before doing a search you reduce the load on your (in this case, very busy) server. Load is evidently a problem, as Digg’s search is also liable to return nothing more than a helpful message saying:

Digg is experiencing a high volume of traffic right now. Please try your search again later.

I’m not the only one who finds Digg’s search unimpressive. It’s slightly ironic that a site whose purpose is to pass judgement on other sites - albeit collectively - is so flawed in one of its fundamentals. (Sensibly I tested my own site search before posting this… and — whoops — found a bug. Fixed now!)

Wordtracker launches free keyword suggestion tool

Wordtracker has launched a free keyword suggestion tool. The tool gives webmasters up to 100 keywords in each session.

[via Search Engine Watch Blog]

This is very handy for anyone wanting practicing search engine optimization without being paid to do so, given that most of the alternative tools require payment of a fee. Note that although the prompt says “enter keyword” (singular), you can enter as many keywords as you like to make up a phrase to be analysed.

How do you use something like this? You should ensure that as many as possible of the associated searches appear on the text of the page you are optimising. So for example to optimise for the phrase “toy robot”, you might want to include “toy robots”, “robot toy”, “best price on toy robot” and several other popular phrases on the page. If you want to be completely “White Hat” about it, it’s important to keep the page content as natural and useful to the visitor as possible. So you don’t want to present them with something like: “Look! This is a great place for every toy robot lover! The best price on toy robot guaranteed! Robot toys, we gottem! Green toy robots, blue toy robots, red toy robots!” You should strive to be more subtle about it.

Note that there may be a lot of value in what is called the “long tail”: longer, less-commonly entered search phrases. If you can score highly on many of these you can often out-perform those who are chasing more obvious shorter phrases.

(I’m certainly going to be using it to research some phrases I want to get a higher placing for. Unsurprisingly, I’m not going to say what they are!)

Google’s One Box Patent Explained

Danny Sullivan detailed the One Box for specific vertical results by Google maybe three years ago.

Bill Slawski does a great job of outlining Google’s recent patent application on the One Box.

This technology should be kept in the back of your mind as changes seem to be moving in that direction.

[via Search Engine Watch Blog]

Google is trying with this technology to return us not merely the page we want, but also to guess if we’re actually looking for an image, a video or a news item. A search for “Iraq” comes back with news items at the top, while one for “Mona Lisa” has images.

Something I take from this is the effort Google is putting in to making their search results more relevant. Anyone trying to make a living from so-called “Black Hat” search engine optimisation should not find this a comfortable read.

(What is Black Hat SEO? Simply put, it’s attempting to fool search engines into giving your pages a higher rank than would otherwise be the case, using strategies that break the spirit, if not the letter, of their webmaster guidelines. Black Hat techniques are rarely used to promote sites that offer real useful content and value since more honest marketing is generally just as successful in those cases.)

Another point to note is that Google is trying to give users what they want by second-guessing their intentions. With the vast amount of data at their disposal it is likely they’ll succeed a lot of the time. Second-guessing is a risky strategy if not done well: I doubt they’ll make the mistakes Microsoft did with its much hated Office Assistant, of being at once intrusive and patronising.

“It looks like you’re searching for pictures of naked cheerleaders. Would you like help with that?”

Garett Rogers writes:
Use your own domain with new Blogger feature
A brand new feature in the “New Blogger” lets you associate your own domain with an account without worrying about FTP settings or being stuck with a “blogspot.com” address.  My first blog about Google was Google Addiction hosted on Blogger — but it would have been really handy if I could simply point that domain [...]
[via Googling Google]

After Google became a domain registrar speculation was rife as to what they were up to. Recently they’ve started making new domain-based services available including a version of googlemail. With their acqusition of Blogger.com, therefore, the abiliity to associate a Blogger account with a personal domain seems like a logical step.

Domain pointing is nothing new. Years ago I set up my first site on Freeserve, bought the oddbooks.co.uk domain and pointed it to my Freeserve domain. But for pointing to work, the hosting service (in this case Google) needs to let you configure its settings to recognise the pointing domain. That feature is what Google have added.

If you have a blog hosted by Google you might consider this as a step towards creating a full-blown website of your own, by getting your domain out there and linked. Presumably Google’s search will not penalise such domains for duplicate content, though that remains to be seen.