08/17/2022

Moss Drake

Panel on the Trends and Future of QA

• Tariq King
• Srilu Balla
• Greg Paskal
• Mark Bentsen
• Philip Lew (moderator)

---

Adding Value to the Business

• Efficiency versus effectiveness -- it doesn’t matter how fast you’re testing or how many tests pass if you are testing the wrong thing

• KPIs should be predictive and driven from data coming out of testing.

  • Cost of unplanned rework - based on historical data

  • Lack of research and analysis in dev before promoting to QA test level

• Value for the business, shift up, what does the business want?

  • Moving quality out of testing and into the requirements. Integrate quality upward... Shifting left and shifting up to test right

  • Testing to test, versus testing to pass

  • Test the product, fix the requirement

  • Test the requirements

  • Don’t just test or validate requirements but shift left further in the development of requirements by analyzing user behavior and learning behavior

  •  Utilize AI and machine learning to learn and quantify user behavior

 

The Role of QA and Testing

• Even the Test Engineer is in sales, needing the ability to communicate and demonstrate legitimate value to the customer.
• A Quality Engineering team can consider themselves valuable to the organization when they would consider it foolish to not include the skills of the proven Quality Team to assess and speak on behalf of the risk identified in the product to be delivered.
• Good and bad testing can appear identical to those not trained in the Craft of Quality. To be a Quality Engineer requires a great deal of integrity on the part of the individual contributor.

 

Automation

• A term used today, “Hyperautomation” can communicates an excess or beyond proportional. This term implies the idea of automating as much as possible. Unfortunately, what can happen is that rather than an assessment and evaluation of what should be automated and when, the goal of automating everything in sight takes priority. 
  • Propose a change in this terminology to Macro Automation with the implication to apply sound automation implementation across the business including testing
• Automation of IoT will take new technical skills in hardware and tools. This is a skillset diminishing since about the 1990’s and could be the next technical bump for already skilled automation engineers.
• Automation can be pushed much further than we’re currently using it.  For example, there’s automation that can test the requirements - automatically checking for inconsistencies, vagueness, and missing requirements.
• Regarding use of AI and ML in automation
  • They can play a role in the analysis and trend detection of risk. 
  • One area where AI and ML can make a significant impact is in Data testing – looking for anomalies. This would be similar to where they are already being used widely to detect fraud.
  • AI & ML are getting bolted on to many tools these days yet bringing very little added real benefit other than using the buzzwords. Pursuing the craft of locator development may be an alternative.

 

Takeaways