Make sure your tech organization doesn’t suffer from a disastrous defect with these three tips to affordable and effective app quality!
By Stacy Kirk
The recent Southwest Airlines and FAA groundings are a harsh reminder that software quality is an afterthought until something goes terribly wrong. IT Quality Assurance is often only recognized when the worst-case scenarios are experienced. Over the past few weeks, we have seen Southwest’s net loss for the quarter of $425M due to cancellations. The worst impact, however, is on their brand and customer allegiance, which is estimated to cost as much as $800 million.
Last week, travelers were impacted by thousands of cancellations when a corrupted database file sparked an unsuccessful reload of the FAA’s NOTAM system. The system outage caused over 9,000 flight delays and 1,300 cancellations.
Even if your organization is not in aviation, such an outage use case can be found in every industry. Here are three tips to make sure your organization is prepared for a disastrous system meltdown.
Assess your testing maturity
augment with a managed testing service
Control costs by setting aside a tiny budget to augment with a managed testing service. Doing this can help catch user-based scenarios and challenge your team or testing partner to save every penny you invest with automation.
Keep your team iterating the features you need to be competitive and meet your customer’s needs. This requires test professionals to go deep and possibly not have time for the big-picture user testing. It is therefore vital to ensure all bases are covered by having an external team that conducts external end-to-end testing.
Plan for the worst-case scenario
Plan for the worst-case scenario by investing in quality and ensuring you test for edge cases that could cause major disruptions. Ask yourself who should be on your disaster recovery testing team. This should be a diverse team that includes developers, testers, IT, Support, product, and even customers.
We often feel comfortable with ensuring that the “core” functionality has been tested. This is called the “happy path”. Yet the happy path only covers the functionality of the system. You must make sure you also test the “very very unhappy path” (when your worst nightmare is realized). For Southwest, their “very very unhappy path” was their system overloading with weather-related flight cancellations.