by Peter Becker - Analyst for Corporate Web Strategy, Teradata
Hi. My name is Peter. And I am a webaholic. Wow, it’s good to get that off my chest. Once web analytics gets a-hold of you…
Of course, we analysts wear many hats. We’re flexible; adaptable. And that’s a good thing, especially now that Internet of Things (IoT) ideology is transforming web content and analytics, across time and devices.
Welcome to the Analytics of Things.
As a web analyst, I need my platform to discover solutions as quickly and as easily as possible. The development of reporting Application Program Interfaces (API) represents a great leap forward for integration with powerful open-source technologies and data warehousing to connect and store all kinds of data. But time allocation and the lack of two-way, real-time integration is hampering progress in web analytics. The more time I spend on configuration the less I can do in analysis; a problem which is compounded when web analytics is integral to multi-genre analytics.
The right information, in the right hands, at the right time
My data responsibilities can be summed up in three words: collection, warehousing, and analysis. And the aim is to get ready-to-use strategic information into the hands of decision makers with the right content-optimisation tools, in near-real time.
Many out-of-the-box analytics platforms (coupled with a tag management system and dedicated development) can help analysts accomplish these tasks simply – now, more than ever. This might work quite well for small business and enterprises, but you’ll run into problems quickly when you need to collect more data from more sources and leave enough time to analyse it.
For the past year we’ve turned to Celebrus, to capture clickstream data from a variety of Teradata domains. This helps us get-to-grips with what customers and prospects are doing on our websites (no-set-up, big-data collection, replacing the tag management system). Instantly, with the installation of a simple snippet of code, we’ve alleviated collection processes.
Next, all of that new data is sent to various Teradata platforms for warehousing and analysis. This is where I get to have some fun, as I can leverage those platforms for some advanced multi-genre analytics.
We’re using our technology to analyse our own websites
It’s fun – as they say – to eat your own dog food. Aster AppCenter, is a full analytics discovery platform that releases users, developers, and decision makers, from the normal enterprise web analytics and BI platform headaches. It provides simplistic functionality like normal web analytics platforms, but leverages an easy-to-use interface that makes data science just as simple. Our report of choice is nPath.
With nPath, we can create advanced visualisations of event sequences on our sites (aka ‘pathing’). With some of the more powerful web-analytics platforms that offer pathing, users can only follow sequences of a single variable – such as URL, for a common example. But nPath is structured quite differently. Paths can leverage any data-sourced variables within the app in the same report, while taking advantage of simplified query filters (not unlike traditional reporting platforms). This means I can configure the data itself within individual apps and leverage multivariate capabilities in a way segmentation never could. And the customisable interface is so easy:
Let’s start with something more familiar. I ran a query for the top 15 distinct web paths from our Cloud Overview page after it received a new location in our navigation menu. We’d like to test them against our other overview pages for predictable and (eventually) targetable behaviors, through our personalization tools (Celebrus has its own real-time servers). The overview page takes users to more in-depth pages, but could also suggest that we add a pricing page to our navigation. So, how big is our cross-section? Who keeps coming back and how often? That’s the appeal of pathing… well-spread icing on top of a big slice of cake. I get enough answers, immediately, with big indicators that tell me where to take my analysis next. I can re-run the same report with a few more constraints, or pull-in more parameters in either my filter or my report.
Example: I want to know which campaigns or channels were the most successful at converting through a gated asset-download. That’s easy enough with a goal-type metric – especially for landing page conversions. But what if I want to know everything in between, in one single report that’s also visually appealing? Welcome to multi-genre analytics; still a simple task inside nPath (although Aster has some more powerful tools at its disposal for this frontier).
We’re only scratching the surface at the moment but so far, so good. Aster gives me the flexibility to enhance the user role with extra scope and speed – beyond the web – while remaining wide-eyed at the prospect of using our own analytics products when I need them (and I do need all of them). This allows for a new kind of team collaboration, freeing web data from its silo and enhancing the output of the product (and the UX) through faster analysis and customised content.
All with the help of the 12-step web analytics programme, of course.