Rainforest QA (CLDRDR, Inc. d/b/a Rainforest) is a company looking to change the way that companies approach software testing by adding a human element. The company uses a stable of human workers fielded from crowdsourced manpower to provide parallel automated testing and quality assurance as-a-service.
In the current world of software development getting a good product to customers is just as important as getting it released quickly. That means testing new functions, changes to code, patches and hotfixes before they go live as not to disrupt customers and this process works best by matching the real-world use of the software, either by emulating the devices it will run on or actually running it on real hardware with banks of test devices.
The benefit of going to humans for testing is that it provides an extra layer of authenticity to the test chain: software changes not only go to particular devices but also into the hands of human customers. The problem of going to humans for testing is that it’s difficult to manage and it’s not as clean or swift as automated software testing, which can be done by and on machines and scaled at the pace of technology.
Rainforest gets around the slower-than-machines problem by buying human power via crowdsourced manpower platforms such as Amazon.com Inc.’s Mechanical Turk program and CrowdFlower, Inc. Workers are hired in groups, given instructions on how to run the tests and log into Rainforest virtual environment to perform the task as provided by the client.
On the backend Rainforest uses an algorithmic system to parallelize testing and input the results into a data pool that can be easily read by the client. Using crowdsourced human testers, Rainforest can provide rapid test results in as a little as 30 minutes making Rainforest an extremely competitive service.
“Companies today can’t wait days or weeks for QA,” said Fred Stevens-Smith, co-founder and CEO of Rainforest QA. “Agile and Continuous delivery demands continuous QA that becomes part of the development process, is more thorough than automation, and gives you reports that your teams can immediately take action on. And that is what Rainforest is all about, QA at the speed of development.”
QA-as-a-service enterprise solution
Earlier this month, Rainforest QA launched an enterprise product for QA-as-a-service that greatly increases the number of testers available to any organization when using the service. The basic service, at the “Staff,” level costs $5,000 per month, paid annually, and provides a powerful pool of testers. Enterprise customers would negotiate their own price points based on needs.
Clients put together test cases via a natural language test creation form (after all these tests will be run by humans following simple instructions), which means that the test makers have the ease of just speaking their native language. The clients also receive a backend that provides on-the-fly information as the test is run, screenshots from testers and a visual guide to the results of the run.
The Rainforest QA-as-a-service backend gives DevOps teams a visual guide to test results.
Rainforest tests can be run on all current browsers, all platforms (Windows, OS, iOS, Android, web) and also desktop software testing.
Rainforest’s service also integrates with common team communication and tracking tools such as Slack, Pivotal Tracker, and Jira. And for power users, and any other 3rd party software integration, Rainforest has a command line interface for access to the service’s API.
The DevOps angle
Rapid development and deployment of new functionality, patches and hotfixes introduces strain on developers to produce bug-free code and deliver it as rapidly as possible through testing to live. Rigorous and efficient testing that can generate as many possible outcomes and stress conditions as possible as would occur in the real world. To do this, DevOps teams employ automated testing services both in-house and -as-a-service that employ emulated hardware to run their new software on before delivering it to production.
Aside from providing the “human element” to testing, Rainforest QA also provides a secondary service which is extremely useful: human testers can test for user experience in a way that a computer-automated test cannot. DevOps managers can see after a test suite is run if certain (or all) testers had trouble understanding a UI by looking at the test result screenshots. This information can be used by the development team to assist in user experience changes before end-user customers end up complaining about hard to understand navigation or ill-explained app functions.
Staying ahead of quality assurance from a testing angle means less down time for developers fixing problems not caught before release. Of course, this also means customers get a better product sooner and the company looks better in their eyes. The quick testing turnaround done by humans also means that problems that might not appear in cookie-cutter automated testing becomes visible before it becomes a problem that’s harder to fix.
Much of DevOps relies on increase communication between development and operations teams (oftentimes mixing the two) and operations deals most often with customer complaints and issues generated after a product is deployed. Human-driven testing in a way virtualizes the experience of the operations team in a controlled environment when it comes to the experience of would-be customers giving DevOps teams an early warning.
Featured image credit: via StockSnap