Just been reading Thom Kennon and find myself in complete agreement with him.

He talks about looking for info on Google’s recent changes to trademark policy. “I first searched in Google for ‘google trademark’ and came up with a mix of old or irrelevant algo results on page one, first timely results below the fold. So I turned to Twitter and searched ‘#google trademark’ and voila — nothing but timely results with a wealth of links back to rich, hot-off-the-presses and diverse content.”

This is exactly Twitter’s strength.

Twitter has become a real-time search engine populated with human-reviewed web links (as opposed to Google which uses non-human search indexing). For contemporary matters, Twitter often produces much better results than Google.

It amazes me that Biz Stone has publicly stated Twitter will not pursue an advertising model (eg. AdWords). That is the PROVEN BUSINESS MODEL you THICKHEAD! See all that screen real estate you’re not using on a Twitter page?? That’s what it’s for! To make you MONEY. Sheesh.

Although they were bright enough to buy the leading Twitter search engine (Summize), Twitter have completely missed the boat in the way they’ve integrating it on the home page. It’s a key feature but has been buried.

Google had better hope that no-one smart buys Twitter. Despite what you might read elsewhere, it’s Google’s only serious challenger.

Google’s made two changes to its search results. The first is an attempt to give you more useful suggestions for your search. For example, when you search for ‘oil’, it will come back with ‘did you mean unburned Venezuelan crude oil?’ at the very bottom of the page under the heading ‘Searches related to oil’. This is practically useless to me, partly because Venezuelan crude was NOT what I wanted to mix with Balsamic Vinegar but also because I have Google set to return 100 search results per page. I usually don’t scroll all the way to the bottom. Lazy. *Thinks* Why can’t my browser make scrolling to the bottom a one click operation? We would call it ‘Page Down’.

Chris Crum from WebProNews describes this as Google trying to improve ‘intent-based’ searches and makes the point that they still haven’t solved the ‘Java dilemma’ that Bruce Cray talked about last year. Bruce said some people searching for ‘Java’ are looking for the programming language, some for Indonesia and some for coffee beans. Hint: most are looking for the programming language.

I would guess there is enough intelligence in Google’s algorithms to detect dramatically different contexts. Perhaps Google could give you a disambiguation opportunity, like Wikipedia do…

When the page loads they could give you their best guess as to what you’re searching for, but also give you a line of disambiguation options. Choosing one of those options would then re-do the query but add in whatever word is necessary to remove the ambiguity in the original search query. So that’s another issue I’ve over-simplified to my own satisfaction.

The other change is to increase the length of ’snippets’, the (normally two) lines of description that appear in each search result. As of now, Google extracts longer descriptions if the user’s search query has lots of words. Which makes sense because people are writing longer search queries now than they used to. And two lines is often not quite enough to give you context.

There are two ways that snippets are chosen. If you write a Description tag on your web page, in most cases, Google will use what you wrote in that tag as the snippet. If you don’t write one, Google will pinch whatever phrase/sentence it thinks best matches the keywords searched for.

So if you’re trying to get people to click on your web page, it’s a good thing to get a four line description rather than a two line one, right? Question is, does this mean you should re-write your Description tags as four lines instead of two?

My guess is that it will not be necessary. Mostly what’s happening under the new system is that if Google finds lots of matching keywords it is ignoring your Description tag and pulling out the phrases surrounding the keyword. Indirectly, Google are increasing their control of the way search results are displayed and lessening the influence of the webmaster.

Book in for Web Promotion SHOCK – it’s aimed at businesses who want to increase their web traffic. The cost is $75; designed to give usable knowledge that people can implement themselves or pass on to their webmaster.

Most Australian businesses, including some very large ones, have no idea about the basic principles of search engine optimisation. (I recently wrote about advertising agencies as an example). So part of my agenda is to share that knowledge. Here is the SHOCK: it ain’t that tricky.

There are still businesses that view their web site as one of those things you have to have. Like a Business Card. But the Internet has become fundamental to how people collect information. Much more powerful than just another advertising medium. Some of the companies I work with are generating ALL their new business off the Internet because they have web stategies that deliver relevant information to people who are looking.

You don’t need a web site. You need a web strategy. Part of that is usually keywords-based SEO that puts you at the top of the Google results for certain terms. Like I am for marketing consultant Perth and like my clients are for most of their target keyword phrases.

I will probably pick up some work optimising people’s web pages, but for me, it’s not so much about editing the Title tags, it involves working out the best use of the web to deliver a business result.

If you know someone who’s not making the most of the web, I’d appreciate your mentioning the seminar to them. Here is the Web Promotion SHOCK flyer if you’d like to print it out. Register here.

Google’s changes to their Search Engine Results Pages (SERPS) let you change the order of your search results. I’m so excited I could breathe. How often do you re-do a search? Those searches would be a very small fraction of my total searches so why would I invest the time removing rogue sites and re-ordering the results?

Unless Google plans to take into account my re-ordering, frankly my dear, I don’t give a damn. Perhaps Google hopes that by aggregating the re-ordering they can assemble more useful SERPS but I doubt even that would work. The people with the vested interest in shifting results are black hat SEO operators, ie the baddies, so I doubt Google will pay any attention to the re-ordering.

They are allowing you to view what other people say about sites and their re-ordering of them and that is a step towards what I’ve described as a next generation search engine. But it’s a very small step. Would be nice to see third party comments on a site if people bother to annotate. But what’s in it for them?

Listening to Bruce Clay and Matt Cutts discussing the future of Google, the key trends are increased personalisation and localisation of search results.

In other words, Google will know where you are, so it will automatically give you the local laundromats if you type in ‘laundromat’. And it will profile your searches (and perhaps your gmail) to help deliver results that match your own particular twisted personality. Bruce’s claim that ‘ranking is dead’ is just headline-seeking. There will still be benefits in localising your pages and optimising them for your target market, it’s just that it will be a bit harder to prove to a client that you’ve done the job right, because there won’t be a single overall search result page. But as Cutts points out, this is already the case. Australian sites produce Australian pages and so on.

The move towards intent-based search and local search are logical extensions for Google. The only people entitled to panic are the companies trying to flog local search directories. That would include Yellow Pages.

The interesting thing for me in Clay’s talk was the notion of Google interrogating audio/video files and weighting search results towards sites providing video content, what he calls ‘engagement objects’. I suspect that process is fraught with technical difficulties but it makes a lot of sense for Google to go down that path. If you’re not currently doing video on your site, go buy a green screen.

I do agree with Clay and Cutts that people offering SEO services will need to broaden their service offering; the future is not just about ranking, it’s about (1) an effective reach and (2) an effective web presence when they find you. I’m quite happy about that trend, it’s what I already do for clients.

The new search engine, Cuil, has underwhelmed people so far with its performance. I guess it’s early days but their apparent business strategies, ‘bigger than Google’ and ‘better privacy policy’ got me thinking about what a better search engine than Google would look like. What are the areas that might offer a competitive advantage?

Let’s start narrow, with two little ones. When you’re looking at a normal search query, you’re getting the highest ranked results. But these are often out of date. You can use the Advanced tab to get recent results but this is a bit kludgy. I’d like to be able to easily order results by recency.

Second, I think Google seriously underuses Alerts, the email notifications you can set up on any search query. With Alerts, you get new results as they are indexed, so over a period of three months or so, you become very well informed on your subject. My point is, running an Alert once a week for a three month search period can be far more powerful than a one-off search that generates 400,000 results. Some searches you do will be one-offs but some will be ongoing areas of interest. Current search engines don’t distinguish between the different classes of search.

Google doesn’t promote the Alerts; I’d be putting the Alert button next to every search query you run. And I’d be offering it as an RSS feed as well as an email.

Now a broader approach.

Google is a bot. An automated search built in a pre-Web 2.0 world. Perhaps the way to beat it is to take advantage of (1) the fabulous people who are already out there categorising and rating content and (2) people doing new searches who are encouraged to rate/tag the sites they visit.

Imagine a service that aggregates searches done on Web 2.0 sites like StumbleUpon, Delicious and WordPress then indexes their tags. Aggretagging. To me it’s not about getting 400,000 results, it’s about getting 20 better results than Google. To do this, you need the discernment that real people bring to the equation.

The other tactic is to turn your search engine into a social engine, using networks of existing and emerging friends. Google has no soul. Compare it to Twitter or Friendfeed; already being used by people to share quality sites. People use social media to build knowledge networks. That’s a threat to the Googlebot model.

Experience with Delicious and Twitter shows that people are prepared to share some of their search results. Our social engine would allow you to share as many of your search queries as you wanted. Cuil was exactly wrong. Less privacy, not more. Combine the social networks with aggretagging and Bot’s Your Uncle. Social aggretagging.