What this episode is about
There's a disconnect in a lot of contact centers. A team lead walks up to coach an agent and the agent says, what do you mean, I'm at 96. And the agent isn't wrong, because the form says they're doing great. The customer just isn't having a great experience.
That happens because most forms were built around compliance and checkboxes. Did the agent use an empathetic statement? Did they say the company name twice? I call that false hustle. It doesn't move first call resolution or the customer experience. What matters is the agent's intent and the outcome of the interaction.
With AI you can look across 50 or 100 calls from an agent and see patterns instead of one mistake. That changes auto fails too. One auto fail out of three scored calls looks like a disaster. One out of a hundred is a different conversation.
Then coach small and often. Give agents a snapshot of the handful of behaviors they're struggling with and work on those in short sessions. QA should be a development tool, not just a policing tool. Compliance still matters, and you can watch it across far more calls than a person could ever pull.
Key takeaways
- If agents are scoring in the 90s and customers aren't happy, the form is the problem.
- Score intent and outcome, not whether a phrase was said.
- A better tool won't fix a bad form. You'll just score more of the wrong things.
- Read auto fails in context. One in three calls and one in a hundred aren't the same problem.
- Coach in short, frequent sessions on the few behaviors that show up most.
In Tom's words
“what do you mean? I'm at 96%. I'm killing it.”Tom Laird · 04:15
“the intent of the agent and the outcome of the question”Tom Laird · 08:42
“you're scoring things that don't make an impact, you're just scoring more of it”Tom Laird · 09:43
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.
[03:07]When you go up to start to coach that agent, right, a lot of times they'll just be like, what do you mean? I'm at 96%. I'm killing it. Right? And there's a disconnect that I think happens a lot of times with the forms that we're utilizing, what people think are important.
[06:43]Again, you guys have heard me talk about empathy, right? Every single person that we see when we get an OttoQA customer, they have a question: did the agent show empathy, or did the agent use an empathetic statement, which I hate, right? And most of the time, we're just checkboxing it. Did the agent say something like, I'm really sorry to hear that.
[08:05]I think when we can move from just scoring calls and giving scores based on these checklists, to finding out the two things that we like to say: the intent of the agent and the outcome of the question, or the outcome of that interaction with the customer. Those are the two important things, not if we used the company's name two times.
[09:00]Everybody asks me, well, will they improve our CSAT and our NPS? I get that question all the time. And my answer is no, if you're using a garbage form. If your form isn't great, you're scoring things that don't make an impact, you're just scoring more of it.
[16:14]Auto fails are still going to be there. But if you have a hundred calls from an agent and they auto fail one because they didn't say something, that's a whole different deal than if you only scored three of them and they auto failed one of them. You think you have a disaster.