The short answerStart by listening to calls and finding what needs fixing. Break the call into sections, train agents on how each section should sound before you score anything, then build a QA form around those same sections with a clear benchmark. Hold people to the standard only after you've told them what it is.

What this episode is about

This is one of the first episodes of the show, from January 2019. The point I wanted to make: most centers train agents on where to click, and very little on tone, delight, empathy and engagement. Then they wonder why customer service is poor. We need to flip the education and the QA system so tone comes first.

If you don't have analytics, start by listening to a lot of calls. You'll get frustrated, and that's okay. Those frustrations are the list of things to fix. Then break the call down: opening and greeting, verification, issue handling, resolution with some empathy, a call to action, any cross-sell or upsell, and the close.

Train each of those sections before you score anything. Role-play how verification should sound, how to handle your top five issues. Then build the QA form on the same sections. I don't care if you use A through F, one to five or a 100-point scale, as long as there's a standard and you benchmark to it. Record calls so you can show agents what a one or a five sounds like.

Once supervisors are scoring and you're comfortable with the numbers, tie incentives to quality. And use both live and recorded calls. If I could only pick one, I'd take live, because I'm fixing issues the same day.

Key takeaways

In Tom's words

“We need to flip this whole education in this QA system on its head where we focus mostly on the tone”Tom Laird · 03:42
“It doesn't really matter as long as there's a standard in your benchmarking to that”Tom Laird · 10:53
“But first you have to tell them what that standard is”Tom Laird · 14:57
Full disclosure: I founded OttoQA, so take this part for what it is. We built it in our own contact center. It scores calls against your own QA forms, there are no seat licenses, and if you want to see it on real calls it's a 15-minute demo. If you'd rather build something yourself, the episode covers what to watch out for. Book a demo.

Transcript excerpt

These excerpts come from an automatic transcript and are lightly edited for readability (filler and obvious transcription errors removed). For exact wording, listen at the timestamp.

[03:42]We need to flip this whole education and this QA system on its head, where we focus mostly on the tone and then we focus on where to click. We need to get our agents to speak properly, to understand the need the customer has at the time, and to interact with the customer on a personal level.

[10:34]I don't care how you score. We have clients that use A, B, C, D, F. We have one that goes one through ten, one through five, some of them have to add up to a hundred. It doesn't really matter, as long as there's a standard and you're benchmarking to that.

[14:42]You have to hold people to a standard. But first you have to tell them what that standard is. So many centers have low quality, and that standard was never told to the associates. They don't really know how to speak to people.

Related reading