With online business growing by leaps and bounds, today’s corporate management, along with those responsible for the day to day running and effectiveness of an organisation’s online presence, have long appreciated that in order to maximise their online business they first have to be able to analyse and understand it.
In other words, you can’t fix or improve something if you don’t know, and know with total certainty, what’s wrong with it, and in order to do that with complete confidence, impeccable data providence is mandatory.
Up until recent times, the wherewithal to attempt to gather and glean this core data was generically termed web analytics and was made up of many approaches and methods, all of course, with different pros and cons.
Historically, web analytics products have been from two camps, ‘logging’ and ‘tagging’. Logging being a process of analysing webserver- log files which, due to its limitations, was extended to include the adding of custom html tags to the page so as to generate custom log files i.e. extended logs; an approach designated as tagging.
Tagging, in spite of its wide application, often causes consternation and aggravation for many concerned due to its inherent inflexibility, the very long deployment time involved and the attendant, unavoidable high costs.
Furthermore, the eventual resulting data often gives cause for concern for several reasons besides just the time and cost i.e. the dubious accuracy of the eventual results. So in the main, analytics remains trapped at an immature stage of development, with cumbersome tagging and even more ‘remote’ methods such as network packet sniffers and web log analysers.
Implementation as has been said, takes many months and maintenance is ongoing and costly – even then, the best an organisation can expect to receive are masses of aggregated statistics and indicative measures only. Such approaches are inflexible, fail to account for offline fulfilment, and require analysts to decide on the processes they want to measure before they can even begin to measure them. At the same time, concerns over detail and accuracy still remain.
A new dawn; a unique approach to this increasingly important requirement is a method designated Dynamic Collection™. Dynamic Collection utilises patented technology from Newbury based speed-trap Limited and is increasingly winning hearts and minds throughout the online fraternity.
As many organisations’ online business presence becomes ever larger and more important, and in many instances is often anticipated to become the most profitable part of their business, being able to accurately and quickly collect and analyse online data with unwavering confidence as to its accuracy and timeliness is crucial. speed -trap’s patented Dynamic Collection solution uses a totally different approach to data collection.
This difference reflects the fact that while speed-trap’s solutions use a single html-insert to collect their data, they do not build log files with the data, but use interaction between active collection servers and the single insert to control and organise data collection. Therefore, speed-trap’s technology is unlike that of other solutions tied to tagging and overcomes their shortcomings.
While Dynamic Collection has significant analytical benefits in many applications, for many types of user the real differences are in cost of ownership and flexibility, that is to say the speed and ease of its implementation. Deployment takes a matter of hours, compared with many months or in some instances years of a costly and potentially inaccurate old style tagging process. The longer it takes to tag a site, the more out of date and less relevant earlier tags become – an emerging and ongoing maintenance nightmare.
An example; in late 2005, Yorkshire Building Society replaced its then current web analytics supplier and implemented speed-trap on its website. Commenting upon the outcome Nick Gander, Yorkshire Building Society’s online marketing manager explained: “Our website has two overall aspects, transactional and non transactional, i.e. where our visitors either have or do not have live interaction with us.
When we were using the previous product we had to put a unique tag on every page of our website and that was a very difficult process to have to manage. It was perhaps easier on the non transactional parts of our site where we had total control of it but on the transactional parts of the site, where we had the application processes, all those deployments were done in phased releases, so if we needed to just add a page or add a tag to it we had to wait for a long period of time before we could do that”.
Explaining further Nick Gander continued: “Further more, our website uses templates too, so ideally what we needed and were looking for, was a ‘one insert fits all’ method which would mean that we could put a small single insert of html code in a template page deploying it just the once in the knowledge that it was then on every single page, and that is exactly what speed-trap’s unique technology does.
So we have one insert that sits on the transactional part of our site and we have another that sits across all of what we call the static part of the site and that’s all we need to do; we don’t need to do anything else”.
“The single speed-trap insert”, Gander explained, “allows us to capture all the data and information that previously required at least one, but usually very many tags on every single page which, as I’ve said, was of course very time consuming, especially in view of the fact that we have thousands of pages on our site.
Here’s an example; if we have a mortgage application process, which could be about thirty or so pages, and we wanted to do a customer analysis, i.e. if we wanted to look at the actual process, following visitors through the application procedure, previously that would have involved tagging every single page with a different tag, and on key parts of that journey we would then have had to put other variable tags in as well for capturing specific types of data - a very arduous process to say the least.
Now we have the freedom and flexibility to look at these things on-the-fly, previously if there was specific information that we wanted to look for then we would have had to tag up our pages accordingly, what speed-trap allows us to do is to have just one ‘insert’, i.e. a single and always the same piece of code, sitting on the page, it’s easy, we can pull the data we need out of a page without any fiddling around at all.
This is because a single ‘insert’ on each web page automatically discovers what is in that page and then collects data on whatever the visitor to our site does; literally everything they see, everything they do, all whilst timing every user action down to the millisecond.
We can even totally recreate a user’s session via a replay, replaying the customer’s online session and their every action just like a video. Here, the page is the source of information rather than artificially added tags. The speed-trap code insert, is identical on every page but, as the page loads, so does the code/insert, collecting dynamically defined data in real time.
All data can be recorded, from the loading of a page to the movement of the mouse, from the clicking of a button to the content of a search list. Accuracy is assured because we’re collecting from source, i.e. the user’s browser and because there are no third party issues to worry about, cookie blockers and fire-walls are untroubled”.
“Majoring on accuracy for a moment” Gander elaborated, “at the end of the initial pilot with speed-trap, they really caused us some severe introspection – when we first saw their results we assumed that they had it wrong, but their confidence convinced us to look at the back-end data, and once we did that we realised that their figures were accurate to within 1 to 2% of the real numbers; the figures we had relied upon for years from two previous vendors were overstating results by over 100%. Furthermore, we have no security concerns because privacy, security and Data Protection issues are rigorously managed, particularly because this is an entirely 1st party solution”.
Malcolm Duckett, speed-trap’s VP of operations and marketing highlighted the now increasingly rapid take up and implementation of Dynamic Collection by organisations from all types of industry and commerce: “For all organisations from whatever sector, be they private or public, the starting point must always be the data; without good data no amount of analysis will deliver the insight or information to allow them to drive business success. We solve the data problem - we do exactly what it says on the tin - our technology delivers more detailed, accurate and complete information than any other solution on the market.
As the Yorkshire Building Society and many other organisations have found, our Dynamic Collection technology avoids the tagging headache other technology forces on you. Deploy our code item in your page template and, as Nick Gander at Yorkshire Building Society has said, it enables you to watch every item of data you need flow in, from click to mouse-over, from submit to basket values, from secure, dynamic, Web 2.0, Flash and static pages. As Forrester Research said a while back, ‘the truth is out there, but not in your web analytics package’ - it is now”!