Noise or Music?

Why web measurement is easy, yet gaining insights is hard

Categories: Implementation Fundamentals, Metrics Understanding / Comments: 2

Collecting data is very straightforward – you simply paste a few lines of JavaScript to your pages and data will start to stream into your account. I am specifically referring to Google Analytics here, but the principal is the same for all the main web analytics vendors.

Superficially that’s all there is to it. If you just wish to view visitors and pageview counts you don’t need an analytics specialist to help you – all you require are basic webmaster skills. However, products such as Google Analytics have 100+ reports so that you can analyse much more than these – in fact, regardless of how much traffic you receive, those can be covered in a handful of reports.

So why so many reports…?

If all you require are traffic volume graphs and a site-wide conversion rate (i.e. the number of transactions divided by the number of visits), then you don’t! That’s the point. Traffic volumes and site-wide conversion rates tell you very little about the success or not of you website. They are blunt metrics, as Neil Mason recently wrote. In order to be effective in optimising the performance of your website – be it in how you acquire visitors, or what happens once they are on your site, you need to be able to answer two fundamental questions:

  • What is the value of a visitor to my site?
  • What is the value of a page on my site?

Being able to answer these two innocuous questions opens the door to the world of insights. That is the role of a web analysis – to provide you with insights so that you can continuously improve.

For example, the value of a visitor allows you to determine which medium/channel provides you with your most valuable visitors. Be it AdWords advertising, organic search, email marketing, social media efforts and so forth. You can then take action on that information such as increase or decrease your activity in those channels to gain more high value visitors or become more efficient in acquiring traffic.

Taking this further, you can drill down into a specific campaign, keyword or even a tweet to access its individual impact on your site’s performance.

The value of a page allows you to identify your best and worse performing pages. Your best performing pages are great targets for A/B and multivariate testing as they can give you the greatest impact. Knowing your poor performing pages allows you to fix errors, improve content or even cull the page in order for you to focus your web efforts more effectively.

Dave Chaffey posted recently on the use of “value”  in web measurement and the 3 key value measures within Google Analytics.

Why is this hard…? Because users just have the basics…

ga-hacks.gifThe hard part is going beyond a basic install of your web measurement tool so that you have a more complete picture of visitor activity, that crucially, includes value. This is not rocket science, but it does require product specific expertise and experience. A best practice installation of Google Analytics for example, requires a knowledge of what can be achieved “out of the box” and what requires further thought.

Non-standard Google Analytics items:

  • Data Structure
    If you own more than one website domain or subdomain e.g.,,, etc. How best to structure the data so that reports are easy to interpret? separate profiles, separate accounts, roll-up reporting…?
  • Tracking File Downloads
    e.g. PDF, XLS, MP3, DOC etc. are not tracked by default.
  • Tracking Events
    In-page actions that are not a pageview. Many add-to-basket, blog comments/ratings, fill-in forms do not generate a pageview.
  • Tracking Flash
    Interactions with product demos, videos etc.
  • Defining Goals & Monetising these
    You need this to determine value.
  • Tracking E-commerce Transactions
    Often these take place via a third party payment gateway and you will wish to track this as one continuous session.
  • Page and Keyword Grouping
    Rather than look at reports on thousands of URLs or thousands of search engine keywords, you can group these along a theme e.g. all pages from section1, all brand search keywords etc.
  • Labelling Visitor Types
    Differentiating visitors who are members, subscribers, customers etc. from other anonymous visitors.
  • Segmenting Visitors
    Sub-sets of related data e.g. social media visits, different levels of engagement, geographic regions etc.
  • Tracking Error Pages
    These are not tracked by default
  • Tracking internal site-search
    Sometimes tricky if the visitor’s query term is not contained in the URL. Also tracking zero results (a very important KPI…)

In the book, I devote 26 pages to getting the basic setup right. The above, more advanced implementation considerations are covered in 96 pages – and the other 400 pages on all the other things you should be aware of if Google Analytics is your thing!

My point is, if you wish to go beyond the basics of web measurement, either get a Google Analytics Certified Partner to help you, or, if you prefer to do-it-yourself, read the book Advanced Web Metrics with Google Analytics. That way you can stop counting and start analysing!

Share Button

Comments (most recent first)

  1. Les says:

    @ Jay: I had a look at the post you reference above….. The product looks quite good – especially with the weekly updates. Nicely done.

  2. Jay says:

    Dear Brian,

    Your blog post was a high-level perspective on how gathering data and viewing it in simple, aggregate form did not provide a clear picture on the intent of a customer or the benefit of a visitor to ones’ web site. I thought it would be pertinent to mention a product that specifically adds value to Google Analytics.

    Next Analytics is an Excel add-in that it exports data from Google Analytics and allows a user to work with their data to create valuable reports that would be difficult or impossible to make otherwise. There are a number of reports created by Google Analytics-certified partners that show, among other things, visitor behaviour, to help an analyst or businessperson get insights from their data. The product allows segmentation metrics such as unique visitors to assist in determining the value of the visitors to ones’ web site.

    Since this particular blog post seemed to be aimed at people interested in getting started in web analytics, it seemed to me that your readers might be interested in a product that made working with Google Analytics easy to work with.

    To learn about one of the reports that focuses on customer behaviour, check out this blog post:

Leave a Reply

Your email address will not be published. Required fields are marked *

Anti-spam question (required):

This site uses Akismet to reduce spam. Learn how your comment data is processed.

© Brian Clifton 2020
Best practice privacy statement