Speed to market is critical for the development of digital platforms.One of the essential success factors, is the ability to decide what to test in the shortest time possible; with an acceptable level of quality. Leveraging predictive analytics from various sources, to bring speed efficiencies in all areas is an untapped opportunity. Predictive analytics encompasses a variety of techniques from statistics to data mining, which analyzes current and historical data to make predictions about future events. One can intelligently use these techniques, right from the strategic identification of what to test, building a test matrix and test execution.

Here is my list of the top seven predictive analytics use cases for testing digital systems:

  • Use case 1: Omni channel (primarily mobile and web predictive analytics): Analytic tools provide metrics in the form of heat maps for most used business processes, features and statistics on operating systems used by channels across geographies. This is a key input for determining test coverage for functional, usability, reliability and volume tests. 
  • Use case 2: Defect prediction analytics: This refers to intelligent mining of data derived from historical Application Life Cycle management (ALM) defect data, which will aid in predicting defect injection and resolution rates while also identifying defects that are likely to cause escalations in production.
  • Use case 3: Test estimation predictive analytics: In this technique we use predictive analytics in conjunction with productivity data for future estimation sourced from ALM tools. This helps to determine the probability of meeting release deadlines and predicting productivity of teams.
  • Use case 4: Test coverage predictive analytics: In this technique analytics can be derived from ALM tools to correlate which test cases resulted in the most defects in past releases. This helps in creating a risk based test matrix.
  • Use case 5: Continuous monitoring analytics for building performance SLA’s: This means, leveraging predictive analytics from operational production data, to build business SLA’s against critical business processes. This in particular, serves as a key  input for performance modeling.
  • Use case 6: Security testing analytics: Security testing, penetration testing in particular, forms a big part of any digital strategy. Predictive analytics tools can help in evaluating known and identifying new vulnerabilities in network set up across firewalls, load balancers and server hardware as well application level vulnerabilities. This will serve as a key input to building the high risk test matrix for vulnerabilities.
  • Use Case 7: Data warehouse testing: This refers, to using statistical and OLAP tools to identify significant trends. For instance, a tester can identify columns, that contain distinctively unique values which are likely to impact the test case and data coverage. Analytics from these, used in combination with statistical techniques, such as orthogonal arrays, will further optimize the test set resulting in improved test speed efficiencies.

In the book PREDICTIVE ANALYTICS: THE POWER TO PREDICT WHO WILL CLICK, BUY, LIE OR DIE Eric Siegel appropriately stated“An organization that doesn’t leverage its data in this way is like a person with a photographic memory who never bothers to think”.

Be the thinking tester and leverage the power of predictive analytics in building your digital test strategy.