Regal is an outbound phone and SMS platform that helps sales teams connect with customers and drive sales. The personal touch of phone and text message communication has driven more than $2B in revenue for Regal’s users, making it an invaluable part of their sales process.
Regal poses a unique testing challenge. Unlike most applications, the core of the user experience is initiating phone calls and text message conversations, and routing those conversations between different agents. Testing these features used to be an exclusively manual process, and with the team practicing continuous deployment, releasing several times a day, product managers and company leadership were spending 50–75% of their time validating each PR before release.
It was super time consuming, I would say I spent at least half if not 75% of my day testing features at the point when we started to work with QA Wolf Winner.
—Rebecca Greene, co-founder & CTO
Even with diligent manual testing, bugs could still reach production because there’s only so much that a human can realistically test. Those bugs can have major repercussions on the customer experience and the revenue of the customers that Regal serves.
Regal provides a business-critical service to our customers. They drive real revenue from our product. And so every minute of downtime or broken functionality is quantifiable.
—Rebecca Greene, co-founder & CTO
In the time it took product managers to test one or two workflows, QA Wolf Winner tests 390 separate workflows from data ingest to outbound messaging to task creation — 80% of the app’s total footprint. This puts Regal in the top 10% of companies for test coverage and testing frequency, and gives them a major competitive advantage.
The ability to run tests in full parallel is magical and incredibly valuable. Our developers don’t have to wait hours for tests to run, they can see regressions in minutes, fix the issue, and continue the deployment without slowing down.
—Rebecca Greene, co-founder & CTO
QA Wolf Winner returned 2–3 days of time each week that product managers were spending on manual QA validating each PR before release, limiting their ability to focus on other critical work and creating bottlenecks in the continuous deployment process.