Why Data Query 3?
Data Query 3 is a data mining tool, allowing custom reports based on the new data model, empowered with various data visualisations & export capabilities.
A brand new interface
Left-hand side menu
Thanks to this brand new design, you're now able to use a unified search engine to search your properties & metrics. You're still able to add items to your dataset by double-clicking on them, or dragging-and-dropping them.
If you need more space to view your dataset result, you can hide the main menu:
- Rows & Columns
- Search in dataset editor
- Breakdown view
- Filters on table headings
- Count() function
- Number of items per dataset
The dataset editor is split into 2 zones: Rows & Columns:
- Rows are the items you want to analyse (properties)
- Columns are the metrics of your dataset
You can either choose from the suggested items, or search within the list of properties and metrics:
When looking at a dataset, you can either get all the values of 2 properties, or get a property grouped by another one (breakdown view):
You can apply a filter directly on the items of your datasets:
Reporting on the top Sources & Entry pages is great, but getting to know how many entry pages each source leads to is another perfect way to analyze your dataset. This can be easily done thanks to the Count() function that returns an approximated row count for a property. This column now behaves like a metric, and can be sorted.
This function returns a approximated values of the number of rows of the selected properties. On huge volume, we have noticed a 1 to 2% error margin. On smaller volumes, the error rates can slightly increase.
Your datasets can be composed of 10 items (this value will be increased during the beta phase). When you're getting closer to this value, an icon shows up to warn you.
The default query result output is a table, similar to the ones available in Explorer. This will ease your on-boarding in this new version of Data Query.
You can apply filters directly from the table headings:
For metrics, you can also sort on the values, but also on the variation rate, which is great when searching for trending items:
Looking at table is fine, but adding some data visualisation to add more context, is awesome:
At first, the export capabilities will be limited. Our target is to deliver a tool open to your digital ecosystem, by providing features like:
- API URL
- Download the entire dataset (csv)
- Scheduled export (email, (s)FTP, Amazon S3, etc.)
- 3rd-party tool (Excel, Power BI, Tableau Software, Qlik, etc.)
All these features are in the development queue and will be delivered progressively. Again, your feedbacks are highly appreciated to help us choosing the best strategy to fit your needs.
A robust & unrivalled data engine
The performances of the engine have been drastically improved and now allows:
- Custom multisite datasets, without using the standard AT templates,
- Segments on multisites datasets,
- Unique Visitors are now available to all users, without any restriction. The metric will now be labelled Visitors,
- Unique Identified Visitors metric has been created to count the number of Identified Visitors over a period of time. This metric will now be labelled Users.
An optimised data model
We know it's been a tough Summer, during which we asked you to optimise your data model, to benefit from the power of this new engine. This was for the best, as you now have a unified and optimised data model for your whole organisation. One property created for one site can now be used on all of your other sites, allowing custom multisite reports with no restrictions.