Lies, damned lies, and statistics.

I’m measuring my web traffic using 2 different services – and getting dramatically different reports. Google Analytics vs WebTrends: which should I believe?

The variance in the reporting is substantial, and each gives quite a different slant on reporting how well my web site is performing:

• Google Analytics reported twice the number searchers clicking through for the keyphrase “internet marketing training”
• WebTrends reported 12% more unique visitors
• Google Analytics says my visitors are primarily from UK, Ireland and Sweden
• WebTrends says my visitors are from UK, USA and unspecified European places

I would always advise clients not to naval-gaze on their web statistics, but instead to use them as trend indicators. And as marketing guru Jim Sterne is fond of saying, if your yardstick measures 39 inches instead of 36 inches, it’s still great to have a measurement tool. (nb – long live imperial measures)

But why are contradictory trends reported even at a macro level by these two systems?

The answer, of course, is that each system is using different ways of handling cookies (deleting and non-acceptance), different ways of defining the word “daily”, and indeed the use of different technologies like JavaScript. It’s really for me to get a better handle on both packages and get a deeper understanding of what the statistics are saying to me.

If you want to learn more Stone Temple Consulting’s 2007 Web Analytics ShootOut gives a very useful overview.

I also think it is time for a reality check, and a reminder that it still isn’t possible to measure the effectiveness of other marketing campaigns like direct mail, radio ads or any of those quaint ways of marketing (!)

I will stick with running both web analytics packages, and together they will serve to validate each other’s results and help me to test the quality of the data.

Why am I using both Google Analytics and WebTrends? To help advise clients:

1. Google Analytics is free, whereas Webtrends is a service you rent, costing typically £20 per month plus setup fees
2. Your data is confidential with WebTrends, whereas I’m sure Google are peeking at my data

Aftervote is a customisable Social Search Engine that is also a great little SEO research tool. Aftervote takes results from Google, MSN and Yahoo, and combines the data into one big search mashup.

And it incorporates the all important Web 2.0 element: vote for the sites you do like, or blacklist whole sites if you don’t want them appearing in your own personal search engine results.

Add onto that dynamic widgets that display PageRank, Alexa Rank, links from del.icio.us, technorati, and digg, and you’ve got an excellent SEO analysis tool.

Be sure to customise your Aftervote experience using the My Settings tab

For the geeks amongst us, it make nice use of Ajax

It’s worth spending some time exploring all the functionality AfterVote has to offer, particularly in terms of customising your searching experience.

Google may command a whopping 78% market share of searches in the UK, but there many other search engines that are superior to Google for specific searching tasks.

Charles Knight has produced a guide to the Top 100 Alternative Search Engines and I tried out some of his recommendations.

ChaCha is a search engine that has a real person waiting there to help you out (I guess they used to be called librarians…) I gave my personal guide Marna the task of finding out driving license requirements in Ireland. The ChaCha gave the search results, and then Marna very helpfully hand-selected two results which were perfect for my needs. Wonderful!

If you haven’t played with Ms Dewey then you’ve missed a treat. She’s tough and sassy and a unique search partner.

My favourite meta search engine is Ixquick – it searches across multiple search engines simultaneously and shows you the relative rankings for your team in each engine.

Charles Knight will be updating his list of search engines regularly, so well worth keeping an eye on those developments.

One of the problems with blogging is that your best content might be buried away in the archives of old articles. Oldies but goodies, but nobody can find them.

But your web statistics will show you your most popular postings. It’s been a year since the launch of the new website, and I’ve had a chance to review my most popular postings.

I’m going to add them as a side bar to my blog in my forthcoming update, but in the meantime:

Susan Hallam’s Top 10 Blog Postings

1. Writing for the Web: web readability tests
2. Internet marketing trends: local search
3. Internet usage in the UK
4. Search engine “findability” index
5. Top tips for your Title tag
6. Social bookmarking and search engine optimisation
7. Pay Per Call: Touch Nottingham
8. Internet marketing myths: Googlebowling
9. Internet marketing toolbox: iWebTool
10. Personalising PPC campaigns using Microsoft AdCentre

What’s worth reading if you want to keep up with the latest developments in search engine optimisation?

Well, why not let other SEO fans suggest what they’re reading?

Google’s personalised home page has a new feature that allows you to define “Tabs” for subjects you’re interested in . Google will automatically populate it with popular content from other people’s Tabs. It’s another step forward in terms of social networking.

So, I created a Tab called “SEO” and the personalised home page gave me these recommended readings:

(… have you also noticed you can now “skin” your desktop with interactive images. In this case, the sun rises and sets according to the time of day…)

The big guns in SEO are all there in my new Tab: Matt Cutts, Danny Sullivan, SEOMoz.

But if you own favourite isn’t there, you can “add stuff” to your Tab. Google tracks as new stuff gets added, and as new content becomes more popular it will become part of the standard “I feel lucky” setting.

Take a look at OutOfMyGord’s posting if you want to read more.

Wordtracker have announced the launch of a UK version of their popular keyword research database. The new service will be launched at the SES conference in London on 13-15 Feb.

Keyword research is an essential first step in developing your plan to improve your website’s rankings in the search engines.

The lack of UK specific data has been a significant weakness for Wordtracker, and this development means they should be able to capture a significant part of the UK keyword research market.

Wordtracker costs from UK 4.20 per day up to 140.00 for an annual subscription.

Other keyword research tools for the UK specific research:

Keyword Discovery which costs US70 per month, or US600 per year
Yahoo Search Marketing’s Keyword Assistant which is free

Continuous improvement is an essential part of your web marketing strategy.

How could you improve your web design? How can you improve the user experience?

I’ve just discovered a quick and easy automated website review service. It gives feedback covering:

• marketing
• design
• accessibility
• experience

and a review of where you rank for what it perceives to be your key phrases.

I only get a measly 8.4 out of 10, and I think they’re trying to sell their own web marketing services. It’s a very useful little tool, but could I suggest you take it all with a large pinch of salt!

I was asked which tools I use to manage this website. I’ve pulled together a list of the software that keeps my website ticking.

My choices won’t suit everybody, and in hindsight I might have chosen differently. But these services are all working and sometimes it is just too difficult to change horses.

This website was designed as a demonstrator of how to easy it is to keep your content fresh and up to date. I use Senior Internet’s content management system for the Events, Library and Client sections of the site.

I didn’t realise how much money I’ve invested in Adobe products: Dreamweaver and Contribute are used to create and update standard web pages. Images on the site are edited with Photoshop Elements, and PDFs are created using Flashpaper.

The blog section is kept up to date with Blogger, and readers’ subscriptions to my blog news is automated with Feedblitz.

I monitor activity on the site using WebTrends website statistics software. Incoming links to the site with monitored with SEO Elite, and overall performance in the search engines is tracked with WebPosition and WebCEO.

Customer data is managed in the office using Goldmine, and I use Goldmine to send out small email marketing campaigns. Larger email marketing campaigns go out using Vertical Response or Intellicontact or Mailchimp.

Any thoughts on my choices?

I talk a lot about search engine spiders, and how they index the text on your website.

But what about us humans?

How easy is it to read the text on your website? How easy is it to understand?

A “readability test” is an automated analysis of the copy on your website. The test looks at relative difficulty of reading your text:

• the number of words in each sentence
• the number of syllables in each word
• the complexity of sentence structure & grammar

The test provides you with the reading age or years of schooling. If you want to reach a broad audience, then you need to be writing at a level equivalent to 8 years of schooling.

Use these readability tests to determine your Fog Index, your Reading Ease Index, or how many years of schooling a reader needs to understand your copy.

Juicy Studio: Readability Test will test your online web pages

Document Readability Test lets you cut and paste your text for testing, and makes recommendations for improvements.

What about this page? You need 10 years of schooling to read it easily, and it fails the recommended Reading Ease test.

Let me know if you understood it, and whether you found it easy to read!

iWebTools is another collection of SEO tools: search engine position checker, a backlink checker, spider simulator, etc

But what it also has are 2 unusual tools: a Page Rank prediction tool, and a Visual PageRank checker.

Today I asked it to predict the page rank for my Internet marketing in Nottingham website, and it predicts my PR will increase from 4 to 5. I searched for an explanation of how it came up with this prediction, but I couldn’t find out how it came up with this (good) news.

Hope it’s right in it’s prediction – I’ll keep you posted.