How Assessing Call Quality Stops Problems from Reaching Contact Centres

Patrick O’Connor of Spearline  discusses how testing the quality of incoming calls can protect the contact centre.

In this article, call quality refers to audio quality and not quality assurance (QA) through scorecards, as is traditional in contact centres.

Would you consider call quality to be the last step of the project pipeline before releasing the product to the customers?

Yes, the last thing that happens, any feature, any bug fix, anything like that, gets call quality verified, which basically means that someone from your team is taking a look at it and fixing any issues.

One of the big things we do is make sure that if the development team adds a new feature, it doesn’t break existing functionality, which can sometimes happen.

We have what’s called a regression suite for that, which means even though the fix/new feature is in the number section, we’ll also check the campaign section as well as checking the manual tester.

We need to be careful as we never know what happened under the hood or whether some unusual knock-on effect has happened.

A big part of what we do is making sure that the new features don’t impact the existing ones.

When you test the platform to ensure the customers have optimal customer experience without any issues, is there any danger of being too careful?

While there’s no real danger of being too careful, you could be too pedantic, for sure. We have a complex platform, so there’s millions if not billions of ways of making numbers and making campaigns.

Even with the best automation in the world, you’re not going to hit every single one of those, so it’s about picking our spots and making sure we get the best coverage in the time we have available to actually do it and not go down rabbit holes at the last second.

What types of programs and languages do you use?

Similarly enough to what the development team use, we use Python. We do that just to keep things simple, meaning we’re not having to chop and change between various different languages and technologies.

Python is a great language: it’s easy to understand and it’s very well supported. With any new features we want to try, it usually has great support for it. We use various different modules or plug-ins to do our work.

We rely heavily on a module called Selenium, which is how we interact with the browser. If you were to see one of our programs running, it would look like the computer is erratic because the mouse would literally be moving around the screen on its own. This happens because we want to simulate what the user does. Selenium allows this to happen.

We would manually test if it was something small, but we want the computer to do our work for us because there are thousands of tests inside there and it would take me two weeks, or more, to run that, while a computer will do it in two hours.

It’s less time-consuming and more efficient. When I’m away from my desk, the computer is doing my work for me. So Selenium is that bridge for us.

It takes our Python script and it’s able to feed those commands into Chrome and actually do those button clicks and enter text and much more.

For further information about Spearline, visit www.spearline.com

Author: Robyn Coppell