Steve Johnston
The Google Blog of a Google Consultant

Following on from my February 21st post, we are now testing the effect of locale outbound links on the Florist near Bath section of the Near directory. Currently a search for a Florist in Bath returns the relevant Near page in 17th position. By adding a few outbound links to high ranking sites about Bath we will test whether they have any beneficial effect. We have abandoned the Manchester and Brighton Bank pages as they are both now at no.1 of the benchmark searches we were doing.

Google have updated their Cheat Sheet. It is always worth a look: http://www.google.com/help/cheatsheet.html.

The ever generous and inspired Ken McGaffin over at Linking Matters has just published an excellent crib sheet on 'How to get links without asking'. Given that the age of reciprocal linking is well and truly over - but that links have as much value as they ever did have, both in terms of customer acquisition and in terms of search engine rankings, particularly on Google - here is some stuff to contemplate. And here, with his permission, is the full copy:

Sending out link requests is a time-consuming business. So wouldn't it be wonderful if other sites linked to you without being asked? Sound impossible? Well, it can be done and here are ten strategies to prove it. Why not start 2005 by making sure you use them?

In a link building campaign, you have to find target websites in the first place, review them to decide if asking for a link is worthwhile, identify the contact person and send them a customized link request. If you can do all that in 15 minutes per request, you'll be doing very well. On that timing, you'll be able to send out 160 requests in a 40 hour week and with a success rate of say 10%, you'll get 16 links for your efforts.

Perhaps there is a better way to spend those 40 hours.

Sometimes the best way to get links is not to ask for them. That doesn't mean you should sit back, do nothing and wait for links to appear. Spend time creating content and making your site as attractive and easy to link to as possible. There are lots of things you can do to build your link popularity without sending out link requests.

Here are some suggestions:

1. Make it clear on your own website that you want links - ask for them directly and make it easy for people to link by writing the linking code for them (using keyword phrases in linking text of course). Explain why it will help: "If you've found this site useful, please link to us so that others can benefit".

2. Play an active part in online discussion groups and forums. Not only will you learn, but you'll become known and will spot the movers and shakers - and if your people post good opinions and helpful advice, people will link to you.

3. Publish a regular newsletter and republish the content on your website. Encourage people to link by asking, "if you've enjoyed this newsletter, you can link to the permanent version at (insert URL)".

4. Publish articles on other websites - ezines, information sites, media sites, even article banks. This works well. You can find site to submit articles to by doing a Google search, e.g. - intitle:"submit an article" business - this produces over 1,000 results. Spend time looking for niche ezines and newsletters that serve your market, scan the type of content they publish and adapt your writing to their style.

5. Create an interactive tool. My favourite of old was 'The longevity text'. You answered about 20 questions and the tool predicted at what age you would be likely to die. Irresistible - I still remember that my prediction was 83. The test was designed and published by an insurance company - could they have been trying to sell me a pension? A little bit of creativity in thinking about such tools will be time well spent and a good programmer will be able to create a tool in just a few hours.

6. Create great content. (OK, it's an old trick, but it still works) Just keep publishing great stuff. One of the best at this is search engine marketer and prolific writer Jill Whalen. Her weekly newsletter from http://www.high rankings.com is a model of consistency and genuine value to her many thousands of readers.

7. Submit your website to legitimate award sites. Mike Corso's Cool Site of the Day, http://www.coolsiteoftheday.com is a must together with any others you can find. You'd be surprised at how high your chances of success are.

8. Praise and link. When you find genuinely useful resources, write a short, complementary review together with a link. Then publish on your website and click on it just to make sure it ends up in your target's referrer logs. Larry Chase of Web Digest for Marketers publishers a monthly newsletter that reviews about 10 web resources - you'll find it at http://www.wdfm.com . A few days later he writes to each reviewed site and allows them to quote his comments in return for a link.

9. Become a source of quotes. Provide sharp, useful, timely quotes for the media. You can get yourself known and you'll find journalists will seek you out. Sign up for services such as http://www.prleads.com ($99 per month) or the premium priced http://www.sourcewire.com (£1,200 per annum).

10. Volunteer to become an editor for several directories. You'll be able to submit reviews of your own sites and you'll have opportunities to get to know your market better - that inevitably leads to more linking opportunities. Go beyond the obvious - as well as http://www.dmoz.org, try http://www.joeant.com, http://www.skaffe.com and http://www.web-beacon.com.

Sometimes we can get too wrapped up in sending out link requests. Steeping back and thinking of some creative ways to get people to link without being asked will be time well spent.

All the copy above is copyright Ken McGaffin, but the advice is yours for free to get on with. Thanks Ken.

Following on from my post of the 15th Feb, and the reindexing of both pages, it seems as if the classification influence of outbound links is relatively important. Both the Near pages have climbed much higher in their respective SERPs. Banks in Brighton result is now No.1 globally from No.6 - 100kb screenshot - Banks in Manchester is now No.4 from No.28 - 60kb screenshot. We are going to make a few more changes, wait to see if the results stabilise and then see if we can discover the relative influence of locale on the Manchester result.

Google have just released a new beta version of the Google Toolbar. Has a couple of interesting new features. The new AutoLink feature looks for patterns of data that appear to be address details, book ISBNs, Carrier tracking numbers, etc, and invites you to click through from there to look at Google Maps, Amazon.com, etc. It does this by changing the cursor when it moves over the data concerned and includes the link in a new drop-down button in the toolbar itself. So it is hard to ignore, and if you were another book retailer, paricularly, you'd be pretty upset that the data in your web page is triggering Google to invite your visitors off to Amazon.

Whilst I can see AutoLink upsetting a few businesses, I do rather like the new WordTranslator feature which translates individual words, when you hover over them, into a preset language of your choice.

I have been asked again recently, what exactly a Google consultant is? So here, for the record, and courtesy of the free encyclopedia, Wikipedia, comes a definition:

"A Google consultant is a person or company specializing in search engine optimisation (SEO) for the Google search engine.

Because Google is the most widely used internet search engine the commercial importance of achieving a good page rank on Google has grown hugely for many companies - particularly those businesses that rely on internet marketing to attract visitors or customers to their websites.

While it is possible for companies to buy position on Google using the Google AdWords system the results tend not to be regarded by those conducting the search as particularly authentic or as unbiased as the page rankings. Also popular words can get very expensive to buy.

As Google's popularity increased companies tried improving their rankings first by using Google bombing in which they created many different sites all linked to one another through a particular word. This gave the site a high ranking when the word was searched for. This "abuse" threatened the utility of Google as a search engine so Google responded by adapting its top secret web crawler technologies and ranking algorithms.

In turn SEO consultants tried to reverse engineer the Google technology in order to gain advantage in the page rankings. So started a struggle between Google and SEO consultants.

As it became more and more difficult to trick the Google system - so the Google consultants became more and more specialised in their work to improve the Google rankings of their clients (eventually to the point where they don't work on any other search engine). The careful use of copywriting, linking strategies and the frequency and timing of updated content can all help move sites up the rankings; the closer a site is to the top of the first page of results delivered by Google the more likely it is that the person searching will click on the link.

The boom in popularity of personal Blogs and Blogging has impacted on the way sites are ranked too and is reflected in the purchase of blogger.com by Google Inc

Google and the Google Consultants have more or less reached a Détente where Google gives a certain amount of advice to Google Consultants on how best to edit or engineer webpages to allow them to be indexed properly. However since Google is now listed on NASDAQ and its commercial success depends to a great extent on selling AdWords by auction it remains to be seen if the symbiotic relationship between Google and the Google Consultants will survive."

And I can't argue with any of that.

Last year I blogged some of the work I have been doing with Near.co.uk, particularly in relation to searches for local businesses. My work continues with Near, a business directory site based on Thomson Local data, where we now see some 200,000+ pages indexed in Google.

Large site optimisation is something of a unique challenge as it has to be rules-based, and frankly there aren't always rules you can create to meet some of the content-related tasks a site like this needs completing. One of the areas we are currently investigating is how outbound links, to sites of relevance to a particular page, influence their ranking. The Near site needs general outbound links in order to provide a good service to its visitors - there aren't many directories that survive on just the paid-for advertised links - but we are phasing in their introduction in order to learn which provides the most assistance in the effective classification of the pages concerned.

We are trying to learn - deep breath - whether classification or locale has the bigger impact on the calculated relevance of a page to a particular search for a business in a particular locality. First, we are building outbound links to sites that support the classification element of two particular pages. The pages concerned are lists of banking organisations found in the cities of Manchester and Brighton. You will see on these pages that the outbound links are to the bank web sites concerned, and a useful control on this experiment is the fact that none of the banking sites are unique to the cities concerned, which means their relevance is only classification related.

The benchmarks for these two pages are quite different and can be assessed by the following searches: banks in Brighton with the Near page already in 6th position at the time of blogging, and banks in Manchester with the Near page in 28th position. And part of the reason for blogging this activity in such detail is to prompt the re-indexing of these two pages. Oh, and if you believe it when Google tells you that 'in' is a very common word and was not included in your search, don't. Take it out and see the results change. Current theories go that stop words such as 'in' are replaced by some sort of wild card measure that looks for the incidence of a stop word between the searched-for terms. I will report on progress, as and when I can.

Call me cynical, but the recent update to Microsoft's MSN Messenger instant message application, seems to have been timed immaculately to follow the full release of the new MSN Search. On Friday last week, the 11th Feb, my Messenger application presented me with what appeared to be a manadatory update. It told me I had to update to continue. On Friday, I closed the warning window and Messenger opened up as normal, however, I suspect the vast majority of users would have accepted the update as presented.

What's this got to do with search engines, well, today, I accepted the update - I do find Messenger a very useful tool, so improvements are welcome - and the really interesting thing that prompted my cynicism, was what it also defaulted to as part of the installation. Two significant pre-checked boxes were presented to me - Install the MSN Toolbar (with MSN Search, naturally) and Make MSN Search my search home page. These options were defaulted to selected, and again, most users would have simply accepted them.

The corollary of this is that, regardless of their actual subsequent adoption of MSN Search, the vast majority of MSN Messenger users will now be exposed to it through the search panel of Internet Explorer. I wonder if this constitutes another abuse of Microsoft's monopoly position.

And the UK's Daily Telegraph newspaper doesn't think much of the new MSN Search either. In a piece today analysing the different performance of Google and MSN for particular queries, the Telegraph summarised that MSN was consistently less helpful than Google. Naturally we expect things to change - presumably for the better - over time, but maybe they let the cat out of the bag a little early.