Web Analytic Matt

Web Log Generator Plugin Version 27
Matt Hopkins posted in Web Log Files on March 30th, 2007

If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!

The first version of the web log generator was a quick fix to my problem of not having any analytics for this site. But i’ve had some time to sit down and improve it slightly by adding a 1st party cookie.

The plugin is a bit heavier in size now but that doesn’t appear to have an impact on page loading times. If you want to test this plugin side by side with its previous version then this is possible and should not cause any conflicts.

You can download the logger2.zip file containing the PHP script at http://www.webanalyticmatt.com/plugins/logger2.zip

As usual just unzip the file into your plugins directory and activate it in your wordpress administration area under plugins. Once activated the screen will go blank but don’t worry about that. If anyone knows how to get round that little bug then please let me know.

Thanks

Top 5 Performance Boosting Techniques For Your Website Using Web Analytics2
Matt Hopkins posted in Web Analytics on March 30th, 2007

You have a website, its ticking along nicely, you get some revenue from it and it provides you with a reasonable web presence.

I’m going to make the bold claim that web analytics can boost the performance of ANY website. But it can also just provide some pretty numbers and graphs that give you no value whatsoever.

In order to achieve the former situation you need a little expertise. Hopefully this post can provide you with some real techniques to start improving your website performance. Importantly, this post assumes that you have a pretty good web analytics application like Google Analytics or Unica NetTracker.

Technique #1 - Conversion Metric
This shoud be the basis of most of your basic web analysis. Your analytics tool should allow you to define a conversion metric for your website.

A conversion is where a visitor to your website has carried out an action on your site which benefits your campany goals. This could be that they have purchased a product online, they have given you their contact information and become a lead, or even that they stayed for longer than 5 minutes.

So a conversion event should be based upon some form of trackable visitor behaviour.

Once you have this conversion metric you can then use it in conjunction with some of your web analytics reports to fine the best converting traits of your website visitors.

Technique #2 - Referrer Report
The referrer report shows you where your website traffic has come from. This is useful information as you can define which web links drive the greatest levels of traffic to your site.

For example, if wikipedia links drive hundreds of visitors to your website then you may wish to add more links on wikipedia or other wiki sites to drive greater levels of traffic to your site.

Quantity is not necessarily quality
This is where your conversion metric can help fine tune your link building. By creating a report of your referrers then applying the conversion metric you can see which referrers give you the most benefit.

It may be that wiki sites give you the most traffic but that directory sites give you the highest levels of conversions.

Then your actions would be to increase the number of links to your site on web directories. This is because a referrer that provides 50% converting traffic is better than one that gives a 2% conversion rate. So build up the levels of the best converting traffic.

This same concept of using conversion rate as a Key Performance Indicator can be used in conjunction with other wb factors such as keyword, page, content area, geographic region of visitor, visitor path and advert landing pages.

Technique #3 - Path Reports
A path report shows the most popular visitor paths through your website.

By using your path report in conjunction with conversion rate you can decide which path you want to drive visitor traffic down.

Another good way of using path reports is to identify the most popular path overall. Once this has been identified you can then use this information to plant calls to action along this path in order to increase your conversion rate.

In my experience I have found that it is often easier to increase the conversion rate of your most popular path than it is to drive more people down a high converting path. This is your ideal path may not be the ideal path of your visitors, and changing their behavioural patterns is often very difficult.

Technique #4 - Scenario Funnels
A scenario funnel or funnel report is one that look at a single path through your website or a website process like a checkout process on an e-commerce website.

Using a scenario funnel you can identify where your visitors are failing to reach each stage of a website process or path. You also get provided with a conversion rate for each stage within that one website process. This can be very useful for finding out what stage of the process is failing at to what extent.

Lets use the checkout process as an example. The diagram below is from Unica NetInsight.

The 3 stages that we want to look at are

1. Entering the checkout
2. Entering credit card and personal details
3. Confirming your delivery address

As we can see from the graph above there is a huge drop in conversions from stage 2 to stage 3.

Using this basic information we can then decide whether to investigate why stage 3 is causing people to drop out of the checkout process. Or you may decide to remove stage 3 entirely which increases your conversion rate immediately from 4.9% to 22.3%. That alone could be a huge jump in website revenue.

There are of course certain limitations to scenario funnels which should be understood when looking at these types of reports. An explanation of these limitation can be found here. Understanding Funnel Reports

Technique #5 - Content Groups
A content group is a way of clustering related website content together. This adds value to your web analytics by creating a form of business context around your website content.

This means that you can compare the content areas that matter to you and your business.

The advantage of this is that it too can be used in conjunction with your conversion metric, showing which area of content contributes the most towards conversions.

Obviously the action to take from this information is to create more of the best converting content. Bare in mind that could be articles, but it could also be product categories, products or even popular downloads.

Tracking the Effectivness of Marketing Campaigns0
Matt Hopkins posted in Web Analytics, Search Engine Marketing on March 28th, 2007

I have written previously about the tracking of email campaigns. Tracking any other marketing campaign works on exactly the same principle.

This post aims to show you how to effectively track your marketing campaigns and more importantly, show you what is possible with this information.

As with all information analysis you need to define up-front why you want information, what its purpose is, what you will do with the information and what information you require. Once you have written down these things, you can then begin to plan your marketing campaign tracking implementation.

Lets take an example of a television advertising campaign which is designed to drive traffic to your website. TV adverts don’t come cheap so you will want to know if web traffic increased as a result of your advert, did the traffic convert into sales and is it worth running TV ad campaigns in the future?

To gain answers to these questions you will need o identify the web traffic as being generated as a direct result of the TV advert. Once the traffic has been identified it can then be tracked through your web analytics tool to find out if it converted into sales which will answer whether the TV campaign is worth repeating. You could also glean the return on investment (ROI) of the ad campaign by then taking those web conversions and seeing how much revenue they generated in comparison to the cost of the ad campaign.

Here comes the tracking
Every ad campaign will normally contain at least 2 elements. The first is a marketing message that you wish to convey, something like “My product is the greatest!”.

The second element is a call to action which specifies the viewer to carry out an activity, something like “Go to my website at www.webanalyticmatt.com and buy my stuff!”.

It is this call to action that we are going to focus on. We need to make this call to action unique to all other website traffic and other marketing campaigns. To do this we use a number of name = value pairs in the query string of the URL of your site. A query string is everything after the ? in a URL.

An example tracking URL might be www.webanalyticmatt.com?campaign=salesgeneration&channel=tv
&segment=sky1&advert=3&dateofad=22-march-2007

Now, the first thing I think when I see that URL is “No way is anyone going to type all that in to their web browser!”. Luckily we can do some HTML redirection magic to turn that complicated URL into www.webanalyticmatt.com/free. (Heres a PHP script to get you started.) So on the call to action you just need to tell people to visit the shorter, vanity URL which will automatically redirect into your tracking URL thus making your ad campaign traffic trackable.

Breaking down the tracking code
In the above example I used the following codes in the trackable URL:-

• Campaign
• Channel
• Segment
• Advert
• Date of Ad

These codes should be used across all of your marketing campaigns so that they can be compared on a like by like basis. Consistency is key.

Campaign defines the overall marketing objective for that marketing campaign.

Channel defines the distribution channel for that marketing objective to be realised.

Segment defines the individual marketing message or audience for that message.

Advert is the individual advert shown which has driven the traffic to the website.

Dateofad defines the data on which the advert was released, as the same advert may be released a number of times to difference audiences and on different dates.

Using the above tracking codes and the right web analytics tool you can have the tracking granularity to ask most business questions about the traffic your marketing campaign drives to your website.

These tracking codes are specific to adverts but hopefully you can see how they could be generalised and then used for any marketing campaign that had a web element. Example campaigns include TV ads, pay-per-click advertising, link building, affiliate schemes, banner advertising, email campaigns and offline print adverts.

If you come up with a truely generic set of tracking codes the please share it with your fellow marketers by posting a comment below.

Understanding Funnel Reports5
Matt Hopkins posted in Web Analytics on March 26th, 2007

A funnel report or scenario funnel is a report that looks closely at a single path that visitors take through your website. This path could focus on a particular website proves such as newsletter signup or a checkout process for an e-commerce website.

A typical graph shown on a funnel report is shown below.

Elements of a funnel report
Funnel reports are defined by identifying each stage within the funnel. This identification is normally based on profiling your website visitors, possibly by the pages they visit but it could be by any online behaviour.

For example in a checkout process you might have 3 stages:

1. Entering the checkout
2. Entering credit card and personal details
3. Confirming the delivery address

It happens that these 3 stages are based on page views within the checkout process.

Another element of the funnel report is the visitor drop off value. This value is shown for each stage within the funnel and shows how many visitors have dropped out of the process at each stage.

The final element of the basic funnel report is the conversion rate metric. This metric is shown for each stage of the funnel except for the first stage and shows the percentage of visitors that have converted or reached the stage from the previous steps.

Common misunderstandings
Funnel reports are not always as they seem. Lets take the example of analysing a web form with multiple fields. You may wish to know where visitors are dropping off of the form.

The funnel report shows the number of visitors who have completed all stages within the funnel. Importantly there is no explicit order in which these stages have to be completed and the visitor may have gone to other pages within the site between stages of the funnel.

So perhaps a funnel is not the best visualisation for this kind of report. If we use the example of the web form then lets use the data visualisation shown in the Venn diagram below.

The above diagram shows all the people who filled out field 1 of the form in red, visitors who filled out field 2 in green, visitors who filled out field 3 in blue, visitors who filled out field 1 + 2 orange, visitors who filled out field 2 + 3 in grey, visitors who filled out field 1 + 2 + 3 in black.

From this diagram you can see every possibility of the 3 form fields being filled out but the funnel report will only show data in red, orange and black. This is assuming that the funnel is ordered by the field numbers.

So why don’t the web analytics tools show this visualisation? My excuse would be the complications involved if there are over 3 stages as the diagram would have to be in 3D to show all possibilities. A 2D version of a Venn diagram is only ever possible when you have a prime number of stages.

My personal feeling on the best type of visualisation for analysing form fields would be a line graph as shown below.

You can see that there are 9 fields and that people drop off the form at field 2 and 4. If you can think of some better visualisations then please post a comment below.

What are the uses of this funnel report information then?
Despite the downsides to funnel reports you can still get a lot of useful information from them and use that information to make accurate business decisions.

As long as you use funnel reports to track exact paths such as website processes where steps cannot be bypassed the visitor drop off and conversion rate metrics are very useful in working out what steps of the process need improving. From this data you can investigate further to see if there are commonalities with the visitor drop offs and and work out how to improve the offending steps.

How Do You Track The Effectiveness of Your Email Campaigns?1
Matt Hopkins posted in Web Analytics, Search Engine Marketing on March 24th, 2007

My first question to you is “Do you track the effectiveness of any of your marketing campaigns?”. I deal with lots of companies on a daily basis that have no tracking for any of the marketing that they carry out. To me this seems very silly as you don’t know which marketing activities are working and where you are wasting money, but today I will focus on tracking the effectiveness of your email campaigns.

So, why would you want to track your email campaigns? What are the benefits?
You need to track your email campaigns so that you can differentiate the email shots that work from the ones that don’t, only then can you attempt to identify why some work and some don’t.

If you know which email campaigns drive traffic to your website, you can then start tracking which links within an email shot drove people and which didn’t. Ideally this means that you can, over time perfect your email campaigns for your market and get the best possible return on investment (ROI).

As well as knowing which email campaign drove traffic, you can also find out which campaign converted the most traffic and so track that all the way to how much money you made as a result of that one email shot. This is a true reflection of marketing ROI.

And how do we do this magical tracking?
Firstly, it isn’t magical. It really is very simple and embarrassingly easy.

Lets take an example of a newsletter sent out to 3000 current registered website users. Importantly you know who each one of those users is, well you know a name, an email address and maybe some other interesting data about them such as geographic location etc.

Using their name you can create a unique identifier for them, lets call it a UID code. Store this UID in your email database along with the users record.

Now, constructing your newsletter. Of course you need to write the content, that’s the hard part which may take many days or weeks. But the important bit is including the tracking.

You will probably have some headlines in your newsletter that link to the whole article on your main website. In these links is where the magic happens. For example, lets say that one of your links is http://www.webanalyticmatt.com/article1.html. That’s all well and good but a trackable example of the same link that takes them to the same page would be something like http://www.webanalyticmatt.com/article1.html?UID=765&Source=email&EID=22-03-2007&link=4.

Lets pull that URL apart now, notice that the only thing I added was ?UID=765&Source=email&EID=22-03-2007&link=4.

The ? shows the start of the query string for that page, importantly this does not usually impact the content on the page unless it is dynamic in some way. The UID value identifies the user, Source identifies that the visit came from an email, EID is the unique email identifier and link identifies which link within the email was clicked.

You can gain finer granularity of data with your analytics with the more specific values that you record. So if you wanted to record the users postcode in the email link as well then this could be analysed, allowing you to segment your market further.

Within your analytics application you should be able to perform look ups of your UID values against your email database to bring in the users actual full name and any other information you have on them. This gives your data real business context, especially if you only have a handful of registered clients, it allows you to monitor their website movements very closely should you wish.

Once you define a particular marketing campaign based on your tracking codes within your analytics, you can then see which email converted the most users, which email link or type of link converts the best, and event which dates are best for sending out email shots. Obviously if you have other Key Performance Indicators (KPIs) that you use to determine the effectiveness of your campaigns then these should be used instead.

It is then up to you to calculate your return on investment (ROI) from your email campaigns based on the number of conversions it produces and decide whether email shots really work for your market segment.

The Online Search Trilogy0
Matt Hopkins posted in Web Analytics, Search Engine Marketing on March 22nd, 2007

This is a post for all the online marketing people out there who know about Search Engine Optimisation and Search Engine Marketing but are a little bit hazy when it comes to how it all fits in with Web Analytics.

To demonstrate one example of how these three parts of the online search story can work with each other I have drawn the diagram below.

How Does It Work?
This trilogy starts with a website. Lets say you have a lead generation website about your carpet cleaning services.

As an online marketer, when building and populating the website with content you would create fully SEO’d copy that would be nicely indexed in the major search engines and hopefully get you a nice pile of visitors that need their carpets cleaned.

Now it is common knowledge that the easiest way of getting traffic to your website is to buy advertising space. Particularly good advertising space can be found at the side of search engine results pages (SERPs) as these will be associated to the search term made by the user of the search engine. Your SEO’d content pages should make reasonably good landing pages for your adverts as long as your marketing message is coherent between the advert and the landing page.

Once you have some visitor traffic to your website you can analyse it in your web analytics application. Things that you may wish to look for are:

• What are your visitors doing once they land on your website?
• Are your visitors converting into leads?
• Which organic keywords are visitors that convert using?
• Which organic keywords are causing visitors to bounce straight off your website?
• How can you shorten the path from your landing page to a conversion page?
• How can you introduce opportunities to convert visitors along the most popular visitor paths from your landing page?

All these questions and more can be asked of your analytics in order to refine your Pay-Per-Click (PPC) advertising campaigns.

Also you can find out what organic keywords cause people to convert, once this is known you can write more content using the good converting keywords. This means that your visitors with a higher conversion rate (targeted traffic) will reach your website via organic search methods. Once your organic content is attracting the right level of traffic you can reduce the PPC campaigns for those keywords and work on targeting other keywords.

What is Web Analytics and Why Should I Care?0
Matt Hopkins posted in Web Analytics on March 21st, 2007

The short answer:
Web analytics is the process of looking at the visitor behaviour to your website and then using that information to make business decisions that will improve business performance.

The long answer:
If you have a website then there will be a reason for its existence, it may be that you run an online shop where visitors buy products from you. It may be that you have a business and require a web presence to describe your products and services and thus generate leads. Or it may be that you’re just creating web content because its fun.

What ever your reason behind having a website you will have, or should have a goal for that website. In the case of the online shop, the goal is to make more sales. For the lead generation site the goal will be to generate more leads in quantity and quality. And for the fun content site the goal may be to entertain as many people as possible and gain more traffic to your website.

Now, you have your website with your current content and you have your website goal or objective. How do you get from your current situation to a place where your goals are being achieved?

This is where web analytics comes in.

At the lowest level websites are hosted on web servers and these web servers record requests made to them for resources. Resources can include images, Flash files, MP3s, PDF documents and web pages. These records are generally grouped into time periods such as days or weeks into web server log files.

It is possible to process these log files and produce meaningful reports on the way in which the web server resources are requested. When the web server records a resource request it will also note down who made the request (IP address and Internet Browser), what time that request was made, which internet page caused that request (Referrer) and some other information about that request. This means that when it comes to looking at the meaningful reports created by processing your log files, there will be a record of who made the requests (unqiue visitors) and in which order they made them (visit clickstream).

Ok, so you have some reports of who used your website (unique visitors), where they came from (referrer), what pages they visited on the site and in which order (visit clickstream).

Using this data there are a number of business questions that you can answer, things like:

• Do I get more traffic to my website on a Monday than a Wednesday?
• What is the most popular area of content that people visit on my website? What do my visitors want to read about?
• Where does my traffic come from?
• Where do people leave my site? Does one area of content cause more people to leave than others?
• What keywords to people use in search engines to find my website?
• Are people using certain keywords landing on relevant content that causes them to convert into paying customers?
• Is there a problem on my website that only Firefox browser users are experiencing?

The list goes on and on, and with different methods of capturing visitor data like page tags you can record more detailed elements of a users visit.

It is important when analysing your web analytics data however, that you have a goal for your website and so you should ask business questions of your data that will allow you to change your website in order to reach your goals and objectives.

But…Why Should I Care?
This is a nice easy one. If you want to achieve your website goals or need to prove your marketing campaigns are working then you need Web Analytics. Oh, and it will save you lots of money in your marketing budget.

If you still need convincing then speak to a web analytics vendor like SCL Analytics who will give you many reasons.

What Are Your Long Tail Keywords?3
Matt Hopkins posted in Web Analytics on March 20th, 2007

I recently read a post by Avinash Kaushik on How thick is your head and how long is your tail which got me thinking about how to best make use of your long tail.

The problem that I could see is that your long tail may be many thousand keyphrases long and to search through all these finding common keywords that you need to target seems like a mamouth task. So I thought maybe a little application that could take in these key phrases as group them by keywords showing how many instances of each word you have might be useful.

So that is what we have below in the little Flash application. If you put in the keywords/key phrases from your long tail in the box on the left, press the ‘FIND’ button, the box on the right will populate with all your keywords and a measure of how often they appear in the long tail.

Limitations: Can only process around 2000 keywords at a time. To process more keywords would require a proper application, this can be provided at a cost. Any enquiries to m.j.hopkins AT gmail DOT com

Once you have these important keywords you then have the task of writing content for your website based upon these keywords. This should, over time draw more people onto your website using these long tail keywords and so turning them into head keywords. Repeat this process every month or so to grow your traffic and hopefully your website income.

As always I am interested in any feedback.

Web Analytics Recruitment1
Matt Hopkins posted in Web Analytics on March 19th, 2007

So what does a company want in a Web Analyst? And why is it so hard to find in the current market?

A web analyst in my mind is someone who looks at the web analytics data from their website(s) and draws conclusions on what action should be taken within the business and on the website in order to increase Key Performance Indidicators.

In order for a web analyst to have any chance of improving a business they need to understand the web analytics terminology, understand the business that they are in and how performance is measured, and how web analytics can be used to improve the business.

From speaking with various members from the Uk Web Analytics industry and my personal experience I would suggest that the following skills are required:

1. analytical skills
2. the ability to make business decisions based upon data
3. knowledge of marketing, especially online marketing
4. understanding of the technology behind the internet
5. a passion for problem solving
6. programming ability in order to realise solutions
7. knowledge of the business

Now this is a difficult mix to achieve mainly because technical people are not often interested in the concepts of marketing and likewise marketing people don’t always have a very technical background.

My suggestion for recruiting those people is to start with a technical or analytical person and then teach them the required marketing concepts, this is because I have found that the ability to be analytical can rarely be taught.

Web Analytics Wednesday - London2
Matt Hopkins posted in Web Analytics on March 17th, 2007

On Wednesday the 14th of March I attended the Web Analytics wednesday in London. This was the first event of this kind that I had been to, as such I was not sure what to expect. There was a really good turnout in excess of 30 people attending.

For those of you not aware: a Web Analytics Wednesday is a Wednesday evening where people interested in Web Analytics meet to network and have a beer. I spent about three hours there speaking to various people from Web Analyst Contractors and Data Analysts to representatives of companies such as Google.

The main points that I came away with that night were:

1. The Website Linked In (where a large population of the Web Analytics community can be found)
2. Currently there is only a small Web Analytics population in the UK and the same names and faces continually turn up
3. Lots of the people who attended were there purely to recruit Web Analysts which clearly shows a niche opportunity
4. Most marketers are unaware of the Web Analytics terminology and simple concepts
5. Everyone loves a beer

Let me know if you disagree with any of those points by adding a comment below.