Data Coverage Analysis

Testing is only as good as the data you have to test with.  This process will enable you to visualise the gaps you have in your test data and the areas of low coverage and saturation.  So you can ensure you have data to cover all scenarios and so improve your software quality by testing for situations your current data does not cover.

To be able to analyse your data coverage using the Curiosity Software Platform, carry out the following steps.

Create a data list to hold the data to be analysed

On the Data Lists page (1) click the New List button (2)

Create a list that will be used to hold the data for analysis.  

Note that the external provider needs to be set to yes.

This will add the Provider tab to the dialog, where you need to specify: the Server, Process (1), Connection to use (2), tables (3), and table columns (4).

By Setting Distinct Values to Yes (5), you can limit the amount of data retrieved, but ensure that all currently available scenarios are checked.

Once you have clicked ok, and the job to create the list has run,  you can open the list.

Open list

On the Data Lists page, click the list you have just created to open it, where you can see the columns that were selected from the data source (1).  

Clicking the diagram button (2) will display the spot diagram for the data, which will show you the data coverage.

Data coverage diagram

You can select which properties to view (1) and review the data coverage (2) for those properties.

You can see that in this example, we only have do not have any example data transactions involving GBP for Transfers or direct debits, and for USD, we only have direct debit data.  So, by looking at the other combinations of properties as well, you can see what areas you will need to generate test data for.