How Poll Everywhere Uses Exploratory Testing to Develop Critical Test Coverage

Picture of Ashley Dotterweich
Ashley Dotterweich, Wednesday March 22, 2017

Building a foundation of critical test cases can be tough. According to the 2016 HPE World Quality Report, 33% of software development teams have trouble knowing what they should test. As a result, teams often test too much -- and waste valuable time running tests that aren’t mission-critical -- or too little -- and miss major issues.

Identifying critical tests can be challenging and time-consuming. Both developer burnout and over-familiarity with the product can make identifying critical regressions hard. Poll Everywhere faced this problem last year as they were ramping up development on several new features simultaneously. “Ramping up testing for these projects would have required me to come up with a set of new regression tests for features that were apt to change at any moment,” says Brian Goodman, product manager at Poll Everywhere. “I didn’t want to be writing and rewriting lots of tests while we’re still in development phase, where the UX could change dramatically.”

In order to keep QA fast during development, Brian and his team turned to Rainforest QA’s crowdtesting network. Poll Everywhere uses Rainforest Exploratory to identify unknown issues in their new feature set, allowing devs to recover development time from exploratory testing. Here’s how Brian’s team has leveraged crowdsourced exploratory testing to find critical test cases faster.

Outsourcing Bug Discovery

Poll Everywhere’s lean team of developers owns their own testing processes. But exploratory testing can be challenging to manage, as it can be hard to predict how long it will take for exploratory tests to yield valuable insights. Ad hoc feature exploration takes developers away from writing new code, and can easily result in th the team getting off-schedule.

Poll Everywhere offloads bug discovery from their developers using Rainforest Exploratory. Any issues found by Rainforest tests are thoroughly documented, giving Poll Everywhere everything they need to reproduce issues and write regression tests to prevent them from resurfacing. Crowdsourced testing gets human eyes on the product, which is key for fast-changing new features that are challenging to write automated tests for. Poll Everywhere was able to discover 36 unknown issues for a recent feature using Rainforest Exploratory -- without taking developers away from other work.

Getting a Fresh Perspective

Poll Everywhere able to identify bugs quickly and more comprehensively because Rainforest testers are relatively new to the feature. In the past, Poll Everywhere pulled developers and team members from other parts of the organization to help with testing. It was easy for the team to experience developer fatigue when testing these features and overlook issues.

With Rainforest, Poll Everywhere is able to get a fresh perspective without adding work to anyone else on their team. “Rainforest Exploratory testing helps us get someone who has not seen any of this before in front of the feature,” says Brian. “They’re familiar with the project because it’s usually the same set of testers, but have fresher eyes, and they’re looking at it with a new perspective.”

Building Coverage Where It’s Needed Most

Because knowing what to test is key to keeping testing agile, fast-moving teams must be able to spot and resolve issues quickly. Rainforest testers provide video recording and detailed notes for every issues found Exploratory test runs. By working with the bug reports produced by these exploratory test runs, Poll Everywhere is able to assess areas of their product that might be problematic. They can then develop regression tests to ensure that these areas stay stable in production.

Learn More about Exploratory Testing with Rainforest

Read our case study to learn more about how Poll Everywhere uses Rainforest Exploratory to discover bugs quickly and speed up deployments.