February 4th, 2012

The Missing Part of the Online Marketing Puzzle : NLPComments Off

One of my hobbies for many years now has been the study of NLP or Neuro Linguistic Programming. This is the study of how people communicate, there are better explanations and facets of NLP but communication is what I’m interested in.

NLP has a number of presuppositions, which are things that may not be true but might be nice to pre-suppose. These include ‘everyone looks at the world using their own map of the world created by their own experiences.’ and ‘everyone tries to pick the best option in any given situation.’.

brain

So what does this have to do with online marketing?

Well, let’s take the situation that you have a website and you track your visitors to the site using a web analytics tool. Your site receives traffic as a result of your SEO and SEM activities. So, you can drive traffic to your site and you can analyse the traffic to learn insights from traffic patterns etc.

Now up until this point you will have been looking at various metrics and KPIs but have you lost sight of the fact that these numbers actually represent real people. It is often very easy to fall into the trap of forgetting that your not trying to attract more numbers but in fact you’re trying to reach out to more ‘people’. This is where NLP falls into place in online marketing as it emphasises the ‘people’ aspect and makes you think about how it is best to communicate with your website visitors.

This all ties up nicely with visitor segmentation, this is taking your website visitors and splitting them up into groups based on commonalities. These common aspects may be geographical location, age, gender, job title etc. or they may actually be based up the communication style of your visitors.

A neat way of looking at this is to create personas of your target market segments. Personas are example people within your target segments and are a good way of ensuring that your marketing activities are aimed at the right sort of people. So, you might have a persona called Sam who is a marketing bod in a small company in England who is given a small marketing budget to work with. Sam enjoys playing basketball and socialises regularly with his colleagues from work. With the added NLP edge we might add that Sam is a very visual type of person who learns and understands best when concepts are explained using pictures. This aspect may be common within that one particular market segment and will allow you to communicate better and ultimately sell more to that group of people.

You can test to see which type of communication method works best with a market segment using standard A/B testing, maybe with an email containing images and one that only has text. But remember that communication methods are not always about visual, auditory and kinaesthetic ways of looking at the world, preferred tone of voice and style of writing can be different between market segments.

In conclusion, I just wanted to open up your minds a little bit to another aspect of online marketing rather than pure metrics and numbers. Remember that you are selling your products and services to people and where ever people are involved you have a psychology aspect which can prove to be extremely useful in increasing conversion rates! :-)

If you want to hear more about how psychology fits in with web analytics and online marketing then leave a comment below and I will come up with some more examples of where I’ve used NLP to improve websites.

What Can My Web Analytics Tell Me?Comments Off

I have had many people come up to me at internet marketing trade shows with the usual set of questions:-

  • What’s all this ‘web analytics‘ stuff then?
  • What’s the difference between your product and Google Analytics?
  • What information can web analytics tell me?

But its not just passers-by that ask that last one, no, its some of our new cusomters aswell.

One of the things that we ask at the beginning of any web analytics implementation is “What information do you need from your shinny new web analytics solution?”. Sometimes i’m suprised when the customer brings out a long list of required reports and KPIs (Which is great!), but most say “What information can I have?”.

So in preparation for that moment when you need to pull out that list of reports and KPIs I have prepared a few questions that a good web analytics application should be able to answer. Have a think over these questions which should hopefully get your brain cells in the mood for coming up with your own specific list of requirements.

Lets start off with the basics :-

  • Which sites are my visitors coming from?
  • What is the most popular keyword people are using to find my site in the search engines?
  • Is the number of visitors to my site going up?
  • What is the most/least popular path through my site?
  • Where do most people leave my site?
  • What is the most/least popular page on my site?

For the content writers :-

  • How long do people read each article?
  • What is the most popular topic?
  • What content causes people to go on and convert on the site?
  • What topic should I stop writing about?

For the marketers :-

  • What is the best converting keyword/page/referrer/landing page/marketing campaign/path etc..?
  • Where do people drop off my sales funnel?
  • At what point do people stop filling in my sign-up form?
  • What is the return-on-investment of my marketing campaigns?
  • Which converts better, paid or organic search traffic? (Clue: Organic mostly!)

For the web developers :-

  • Which part of the website is broken today?
  • Which web browsers should I adapt the website for?
  • If I create a new Flash demo what percentage of our visitors won’t be able to see it?

For the boss :-

  • Is the web site making me more money than it is costing to maintain?
  • Which member of the web team should I promote/fire?

There we have it, a basic and general list of questions that your web analytics system should be able to provide for you. For more contextualised information either look inwards for the questions or ask your local web analytics consultant.

Web Analytics Methodology Part 1Comments Off

So you’ve just read or heard about a technology that can improve your marketing bottom line, or maybe you’ve known about it for a long time but haven’t been ready to take the plunge as of yet?

Of course I am talking about web analytics, but what is this technical term that you hear spoken about at marketing conferences?

Well, web analytics is simply looking at where people go on your website and how they interact with your site. The reason you might want to peer into the world of your website visitors is simple, to learn from their actions so that you can improve your website conversion rates.

This is achieved in many ways, for instance you can find the paths through your website that convert the most visitors, or you could find out what content is attracting your target market segment. You may think that the answers to these questions are obvious to you, but there are always lessons to be learned from your visitor’s behaviour on your website.

Once you have made the decision to proceed with web analytics there are a few things you should know.

  1. A web analytics implementation is not a 1 hour process, in fact to do it properly a project should be created to ensure that the process is properly managed from start to finish.
  2. Web analytics is an ongoing task and not just a one time fix for all your marketing woes. Web analytics may start at just one hour a day but in my experience I have found that once you start providing detailed web information, people in your company will sniff the coffee and wake up to the realisation that this is real and actionable information. At that time the web analyst becomes the provider of information and becomes worth their weight in gold. We have seen this happen in America where web analysts with only a few years experience are being hired for $100,000 a year. The UK is following suit with a growing market in web analytics.
  3. The other important thing to know before you embark on your web analytics voyage of discovery is to outline your desired outcomes.

Outlining your desired outcomes means that you have to write down somewhere what you want to do with your web analytics information. Do you want people to interact more with your brand? Buy more of your products? Do you want to grow your website traffic?

You should create these outcomes and put them in a table with three columns as shown below.

Overall Objective Desired Outcomes Who Cares?
Generate more income from the web Increase website traffic CEO
Marketing Manager

In the first column you should define your overall objectives for your web presence.

In the second column you should put all your desired outcomes in order to reach your objectives.

In the third column you should match up the desired outcomes with the person who cares about that outcome. Let’s say your outcome is “Write content that attracts your target market share.” I would imagine that the marketing department and especially the person/people who write your website content would like to know what content is attracting the ‘right’ traffic.

This table will be useful when customising your web analytics solution to your business and sending the right information to the right people, as in the people who can take action.

The way in which this information reaches people is vital.

If your content writer is sent the number of views, visits, visitors, bounce rate and view:visit ratio then they may get overwhelmed with the technical jargon. However if they are sent the number of readers, the reader to comment ratio and the article relevance numbers then they might mean something a little more obvious to the content writer.

So information is near useless unless it is given some context, this is where Key Performance Indicators come in. They allow information to be presented in a way that is familiar to the business. Importantly, these metrics should be ‘key’ to the business and provide an ‘indication’ as to whether the website is ‘performing’.

Part 2 of the web analytics methodology will continue with how to define KPIs for your business along with the next steps of your web analytics journey.

Designing a website with web analytics tracking in mindComments Off

As a web analytics vendor consultant I see all types of websites every day in all different shapes and sizes. But the one main difference between what I think are the good ones and the not so good ones are their trackability. By that I mean the effort required to track visitor traffic throughout the website.

A quick note to remind you that no matter what type of website you have, be it Flash, AJAX, dynamically driven using Coldfusion, all website traffic can be tracked its all about how much effort you wish to put in to tracking.

I have compiled a short list of what I see as the most trackable types of site with the least amount of effort required.

  1. Flat file html web page sites with really clear directory structures and page names, the kind your SEO (Search Engine Optimisation) company would/should create.
  2. Dynamic websites in PHP, Coldfusion, ASP, .NET etc. where all user entered values are output to the URL. Eg. http://www.bookassist.com
  3. Dynamic websites that use one page like index.aspx and pageid values for every single page.
  4. AJAX websites.
  5. Flash websites.
  6. Websites that mix any number of AJAX, Flash, flat file and dynamic elements.

Most of these types of websites you can use page tags or log files as your data capture method of choice however it is only possible to capture data for the following site types using page tags. Be aware that when implementing a page tagging solution that you should plan what data and where it should be captured very carefully to avoid misinterpreting your web analytics data.

  • Flash
  • AJAX

So why are flat file web sites easier to track than Flash files?

Well data from Flash files has to be collected by specifying the user events to track, this is done using page tags. Flat file websites store all their data automatically in web server log files by default.

That is not to say that page tags cannot be used for flat file websites, there are advantages and disadvantages for both page tags and log files.

In conclusion not all websites are created equal but they can all be equally tracked with varying degrees of effort.

Compete introduces new web 2.0 measurementComments Off

The competitive analysis tool Compete has recently announced that it has introduced a new metric to its data called attention. From what I can see this ‘web 2.0′ metric is simply the total time online for the average visit rebranded.

But this idea of taking our plain web 1.0 metrics and turning them in to something funky and web 2.0 like got me thinking about other useful web 2.0 metrics or KPI (Key Performance Indicators) I can think of.

So we have:

  • Attention – time online for an average visit
  • Visitor participation rate – a conversion type metric used to measure if a visitor participates in web 2.0 activities such as posting a comment or submitting an article to Digg.com
  • Visitor consumption – how much content a visitor consumers which could be like the number of page views they see, but obviously it could also equate to AJAX/Flash events
  • Feedback rate – the rate at which a piece of content receives feedback via user participation, for example this could be based on average comments per blog post
  • Subscription rate – the rate at which content is subscribed to via RSS/Podcast/etc based on the total number of views of that content
  • Activity rate – the average time a visitor moves from page to page or event to event during their visit
  • Consume -> participate rate – a value based on the number of times a piece of content has been consumed which has led to visitor participation
  • Content loyalty – a rate based on the number of types of content a visitor consumes over the visitor lifetime

With any good web analytics package all these metrics are fully achievable, obviously given the data to analyse. But the questions are:

  • “Are they any good as web 2.0 metrics?”
  • “Do they hold enough meaning to base business decisions on?”
  • “Can you think of any better web 2.0 metrics?”

Go on, post a comment and in a few days i’ll update this post with my “consumer -> participate rate” based on your activity.

Update: after 14 days this post has 0 comments, I would say that the consumer -> participate rate is 0%, I guess no more talking about Compete metrics then :-)

Web Analytics MethodologyComments Off

I know that web analytics is a relativley up and coming field but it still suprises me how there is so little information on the web when it comes to a defined methodology for carrying out web analytics. So I thought I’d write one out that I use in my role as a web analytics consultant.

Once you scratch the surface of web analytics you will find that its a big subject with lots of facets. As such I have split this methodology into managable parts that will be posted on my blog over the coming weeks.

The contents of the methodology looks like this:-
Part 1

  • What is web analytics and why do I need it?
  • Defining your desired outcomes
  • Measuring your outcomes with KPIs

Part 2

  • Defining your KPIs
  • Who wants what information?
  • Your tracking plan

Part 3

  • Defining your data capture plan
  • Picking a web analytics tool
  • Hiring a web analyst

Part 4

  • Implementing your web analytics solution
  • Monitoring using your analytics
  • Turning analytics insights in to actions

Subscribe to the RSS feed now to avoid missing any of the 4 parts.

Read part one of the web analytics methodology
Read part two of the web analytics methodology

Emetrics Summit London 2007Comments Off

I was lucky enough to have a day at the Emetrics Summit in London this year. I had not been to one before so it was interesting to see how it compared to the other web analytics/marketing events.

It was definately not like the other events such as TFM, Adtech and SES. There were I think no more than 200 people in what appeared to be a well organised networking event. As well as the opportunity to swap stories about how your company deals with web analaytics challenges, there were seminars throughout the event.

The seminars ranged from competitive analysis to behavioural targeting and some of the topics were really very indepth like the web analytics methodology presented by the web analytics team for Nokia.

Personally I found it good to put some faces to the names I have previously heard in the web analytics market such as Jim Sterne, Avinash Kaushik and Miles Bennett.

The main things that I learned from the event were about the behavioural targeting space. With Touch Clarity just being bought by Omniture it was suggested to me that Omniture had just taken the one partner in the market that everyone wanted to partner with, leaving a gaping hole in the market. It was also suggested that the people to take the place of Touch Clarity are the multivariate analysis outfits such as Optimost seeing as Touch Clarity were doing behaviour targeting combined with multivariate analysis.

I wonder how long it will be before Google Analytics bring in their behavioural targeting? Especially as they already have the multivariate analysis piece in place with their Google Optimizer.

Anyway, the event was good from what I saw and the food at the Russel Square Hotel was absolutely amazing, the best cake i’ve had in quite some time.

If you went along, let me know what you thought by adding a comment below.