All 8 entries tagged Internet Search
No other Warwick Blogs use the tag Internet Search on entries | View entries tagged Internet Search at Technorati | There are no images tagged Internet Search on this blog
July 25, 2007
Writing about web page http://www.johnston.co.uk/2007_06_01_blog-archive.html
I happened across a couple of ‘long tail of search’ quotes recently that confirm the rule that half of all search terms are unique.
Chris Anderson’s ‘Long Tail’ blog quoted a senior Microsoft executive’s letter confirming this ratio.
Steve Johnston’s blog confirms this ratio for Google and goes on to quote an even more amazing claim.
Udi Manber, VP of Google Engineering said at a recent conference, that there are three reasons why Search is only going to get harder in the future. One of these I already use – “scale and diversity are almost beyond comprehension” – one of which is not particularly relevant to this point, but the third will replace my previous reference: “20 to 25% of the queries we see today, we have never seen before”. I will convert this into ‘1 in 4 of the expressions typed into Google today have never been seen by Google before’. Ponder that for a moment. Google is tapped into our collective consciousness. It’s astonishing.
Truly astonishing indeed! My experience, where no matter the size of the server logs that I processed, 50% of the search terms were unique also confirms that these figures are taken from unprocessed logs. So extraneous characters, ignored by the search engines, have helped reach these easily remembered ratios. The other route causes are synonyms and word ordering, see word order counts.
Most of these ‘very long tail’ search terms can include words from the ‘short head’. My findings that 80% of visitors to a well titled page will have used one or more words from the title yet 50% of the page’s search terms will still be unique! Having looked at all of the long tail terms for a page titled “klaxon horn” (and many others) I can also believe that 20-25% had never been seen by Google before!
March 29, 2007
An earlier entry described how important webpage titles can be in attracting visitors. Visible, bold, larger font, title keywords in search results encourage searchers to click. The closer titles get to echoing searchers queries, the more they will click.
I collected this list of the five most common errors with examples of useless titles.
Have I missed any?
- No titles at all.
- Useless titles; “New page 1” “Welcome”
- The same title for every page.
- The same very long title for every page.
- Template titles; “About”, “Contact”, “Location”, “Products”, “Services” etc.
January 12, 2007
Writing about web page http://www.skillspin.co.uk
Many pages that I have analysed in my data mining exercise have attracted over 80% of their visitors with keywords in their titles.
The power of a good title for your webpage has been well documented in the SEO websites. When you look at the search engine results pages, SERPs, on Google, Yahoo! & MS Live, you can understand why. The top line of all the results, whether 4 or more lines per result, is the title.
These are in a bigger font and the words from your search terms are in bold.
Searchers are more likely to click on a title with more of their search words. It shouts out that this is a page that answers their question.
So how does this line up with the 'long tail' of searches described in my previous entry?
A single title keyword "tracker" brought in 71% of visitors to one page. This word was also in 915 different search terms, 3/4 of them entered by only one searcher. So these two facts do line up.
To bring visitors to a page it is essential to have a good title that accurately describes the content.
To attract the visitors from the 'long tail' the page should be written in natural language with real information, enough words and variation. How many words are needed could produce another post.
December 11, 2006
Writing about web page http://siteexplorer.search.yahoo.com/
I have found the most accurate Inbound Link tool is the Yahoo! Site Explorer. It also gives a clear indication of the pages from a site that are cached on Yahoo!’s databases.
Inbound link numbers are not as critical these days with quality and relevance being more important than quantity. All the search engines need to find a link or be notiified to be able to discover your page.
The site explorer allows you to select “except from this domain” to reject links from your own site and links to the whole domain or to a subdomain.
October 24, 2006
There are many ways to use the search engines. All of the list, below, have been observed in recent years. Some are more efficient that others but all can habitual.
- Search Site = Home Page – to search again use the Back Button
- Bookmark the Search Site – to search again Back Button
- Use the search panel – Searching again is easy but takes the most screen space.
- Type into IE Address Bar – Search Again = Yes – Defaults to MSN or toolbar – Smallest footprint.
- Start Bar ‘search internet’ – Search again Yes as Search panel – 3 clicks to start!
- Search Toolbar – Search Again Yes – Small footprint.
- Deskbar search – Search again Yes – Precurser to OS desktop.
- Search on OS Desktop – Search again = Yes.
- Select text & right mouse button – Search from anywhere – choice of local computer or internet.
- Site specific search. e.g. eBay and Amazon.
- Specialist search e.g. Google Scholar, Yahoo Images, Blogfinder.
- Non-search finding e.g. History, Down arrow on Address Bar
The growing choice are the search toolbars, where Google and Yahoo! are dominant with 95% share. IE7 and Vista could change all this by putting Microsoft search as default in the desktop, (with the option to change to other preloaded search engines). Toolbars account for only 12% of all searches however, comScore (2006).
From my visits to SMEs in the midlands the evidence of policies on the computers reveals the following options.
• Company policy. All computers with standard builds, mostly networked.
(This is conscious and planned.)
• Installer & maintainer policy. Often the default search engine lines up with the market leader at the time when the computer was installed. (Conscious but ad hoc.)
• User policy. Adding toolbars, default search providers etc. Google wins here.
(Conscious and ad hoc.)
• Software updates policy. These are agreed, once, for virus scanners, acrobat readers, etc. These can introduce ‘enhancements’, Adobe’s Acrobat update pre-checks ‘Download Yahoo Toolbar’ on the left when the user needs to click on the right to continue. (Unconscious and ad-hoc.)
A lot of toolbars were installed on the computers of inexpert users in the unconscious and ad-hoc way. So are these ‘software policy’ driven installations the next market share battleground for the search engines?
There is a battle for these installations using the likes of Acrobat (Yahoo), Java (Google) and the browser/OS (MS) but the real goal has to be to educate the users to habitually use them.
My observations have revealed a lot of unused toolbars. The conscoius users of the latest tools and methods are likely to be the ones who are driving the growth of searching. These users are finding that these ever-present tools can solve most of their information finding needs with one simple mental model.
September 27, 2006
The strongest growth in search marketing in the US has been in location based advertising. All the main search relevance advertising channels, Google Adwords, Yahoo Search Marketing (was Overture) and Microsoft Search Marketing offer this service.
If a business offers a service in a specific geographic area it makes sence for their Adverts to show only to searchers who are in the area.
The searcher sees only adverts that are relevant to their query and some will be local. If a plumber bids on local cities in the West Midlands and someone from Coventry sees thier advert the word Coventry will appear under the display url. So this combines the advantages of relevance advertising, displaying only when the user is searching on the subject, with the Yellow Pages.
When I search here at the University and find location triggered Adwords the location shows as Warwick. The University is actually on the edge of Coventry but this is understandable. When I do the same from my Coventry home from a fixed IP ADSL broadband link it shows up as London!
Why is this? It turns out to be a limitation of the BT wholesale ADSL network. Their backbone ATM network can only resolve an IP address location dowm to three central hubs nationwide. Since BT’s ADSL has the majority of the UK’s broadband connections this is bad news for fixed line location based search marketing.
The Cable competition, Telewest and NTL, now combined as one joint company, do provide the correct location information down to the City and Town. (I am an NTL customer with their Analog TV service but unfortunately they will probably never upgrade my area to digital and offer broadband cable modems.)
Google Local and competitors will allow promotion alongside their Maps.
This will still work and allows users to look up restaurants, etc., where they plan to travel to.
True location based marketing in the UK will probably be delivered using mobile devices such as PDAs and smartphones as these really do need to know your location at all times to connect calls.
September 15, 2006
Writing about web page http://www.wmccm.co.uk/WMCCM/DesktopDefault.aspx?tabindex=9&tabid=2916
For the last three years I have been advising our SMEs that they should add their location information onto every page of their websites.
Often engineering customers, searching for a supplier to solve an immediate problem, will find many results from China, India, Eastern Europe and the US. If they are looking for a local company they will either use the ‘pages from the UK’ button if they have the option or add some location information.
The .co.uk, .org.uk & .me.uk high level domains have taught British searchers to add UK. They might also add Midlands, West Midlands, Birmingham, etc.
Recent changes to the WMCCM site made it clear that our own site failed the location information test. The best, long term, solution will be to add this to the Site Banner default style but this will take time. (The bottom of the page on www.wmccm.co.uk , and other ASP.NET sites can be a an indeterminate place.)
I have tried three approaches on the public tabs.
A single line at the top. This is my preferred approach being closer to the eventual solution. This is seen on the Case Studies Tab at http://www.wmccm.co.uk/WMCCM/DesktopDefault.aspx?tabindex=9&tabid=2916
Using a html panel on the left. See the Showcase Tab, http://www.wmccm.co.uk/WMCCM/DesktopDefault.aspx?tabindex=7&tabid=3142
In a larger panel in the centre pane as in the Directory Tab, http://www.wmccm.co.uk/WMCCM/DesktopDefault.aspx?tabindex=2&tabid=15
What do you think?
September 06, 2006
Writing about web page http://www.wmccm.co.uk
There are two ways of measuring the growth of the search markets, survey panels and web-server log data. Both these have been revealing significant growth.
Neilson//NetRatings Survey Panels report 55% annual growth to Dec ’05 against a 3% growth of US searchers. Hitwise, from their ISP & 500,000 web-site logs data, report Google’s UK searches were up 77% in the year to Nov ’05. Most indicators show Google gaining share with all the competitors holding position as they grow.
The search ranking sites differ as to the absolute market share of Google.
ComScore Networks and Neilson//NetRatings have Google at 43% and 48% respectively in March ’06 using survey methods.
WebSideStory using webserver log data has Google up at 55% in the US and 75% in the UK, Feb ‘06. Hitwise supported the UK figure showing 75% in March ’06 also using webserver data.
Two main differences stand out.
1. That Google is stronger in the UK is clearly shown in WebSideStory’s comparative data.
“Even more so in the U.K. than in the U.S., when people think of search, they think of Google,” said WebSideStory, Chief Marketing Officer, Rand Schulman.
2. The web server log file data shows a stronger Google share than the panel based surveys.
There are two possible reasons for the difference between panels and web server logs. Some panel results show unique visitors per day/week/month and underreports search volume. This is not true of Neilsons/NetRatings who project total daily search volumes. The pure server log file results will underreport visits to sites that don’t employ SEO professionals such as news and TV sites. Hitwise also use Internet Service Provider data to fill in this gap.
John Battelle in his book The Search, characterised Yahoo! as taking the human, contemporary, approach to Google’s technical, academic approach. Yahoo! has focussed more on Big Brother and reality TV shows. This supports an underreporting of Yahoo! referrals in server logs.
This analysis confirms that UK engineering companies should concentrate on Google.
This is supported by search engine referrals to www.wmccm.co.uk