Webtrekk's Q3 (SaaS)

A web analytics tool like others?

Data collection is based on a Javascript-based tag that is very similar to most common solutions. The tag can be customized based on your needs and it is possible to create your own custom parameters – nothing new there.

On the reporting side, WebTrekk Q3 provides most usual features that any Web analyst expects to find nowadays in a Web analytics tools. These including the usual reports (visitors, visits, pages, sources), content groups, campaigns, path analysis, workflows, e-commerce, browsers & mobile devices, scorecards & KPI’s, overlays and many others.

WebTrekk offers the possibility between SaaS (Software-as-a-Service) or in-house solutions and it exists in two versions (light and full) – the light one already offering the most important features.

So what differentiates Webtrekk?

The main differentiator is that WebTrekk Q3 works with “raw” data that is neither sampled, aggregated nor pre-processed. Whenever you access what is called an “analysis” (usually called a report in other tools), the corresponding query is run against the raw data and the corresponding results are calculated on the fly.

But what makes WebTrekk Q3 really powerful and flexible is its “filter engine” - probably the most important feature of the tool. The filter engine is basically a real-time segmentation engine. It allows you to filter (i.e. segment) any analysis on almost any criteria available in the tool – be it a dimension (e.g. all visits coming from search engines) or a metric value (e.g. all visits with more than 10 page views). The filter is directly applied on all your data set. Data segmentation is retroactive so there is no need of reprocessing your data.

You can combine multiple criteria in order to build advanced and complex queries. And depending on the objects used in the filter, you can choose different scope (called “context"). For example, if you use a filter on page name, you can filter at page level (e.g. retrieve all pages with name that contains “product”), at visit level (retrieve all pages viewed during visits where product pages were viewed) or visitor levels (all pages viewed by visitors that viewed at least once a product page during any of their visits).

The filtering engine is not only used for segmenting analysis but also for creating custom metrics and formulas – providing a lot of freedom to create your own advanced metrics. Custom metrics and formulas can be added to any report at any time – here again, no need of reprocessing.

Additional features:

• Advanced campaign configuration with the ability to configure the attribution model (first, last, both, weighted...) and the possibility to see what are the preceding / following campaigns (for a specific campaign).
• Depth of engagement report is a kind of high-level funnel report where you can define different levels of engagement and see how your site is driving visitors across the different groups.
• Others interesting features include cross tables (allows the combination of two dimensions into a single data table - displayed in a visual way), built in heatmap report (similar as the overlay report but shows click density on top of the webpage), SEA and SEO data integration (Google Adwords API to integrate data from your Adwords account (at additional costs), reports with SEO data such as page rank, search volumes, backlinks for a defined set of pages).

Drawbacks:

• The user interface: It looks a bit oldish and it is not the most user-friendly interface nor always that intuitive. Some basic functions are not always that straightforward or require more clicks than necessary.
• The learning curve: WebTrekk Q3 is certainly not “easy” to use for non-expert users. It requires some practices and training to use correctly and to know how to leverage its “power”. This is partly due to the interface but also because it is more complex to "configure" an analysis. But once you get it, it proves to be very powerful and flexibe.
• Performances: Because it works on raw data and calculates results on the fly, it is slower than what I used to see (in WebTrends or Google Analytics for example) and for complex queries, you will have to wait.