The Contact Center QA Score is OVER- How to Score QA in 2026

00:00Tom Laird here. You've heard me break down contact center technology, technology on the podcast, time and time again. Now, it's time for me to answer your challenges. As your contact center technology advisor through our new company, Expivia Digital, I can help you with things like CCaaS selection, any type of AI implementation, and nice studio and integration services. honest advice you hear on the geek applied specifically to your operation. Visit Expivia.com to get started. That's Expivia.com. Welcome back, everybody to another sort of advice from a call center geek, the call center contact a podcast. So we try to give you some action, brought items take improve the overall contact center, hopefully improve your customer experience as well. My name is, Tom, Laird hope everybody's doing well. I got some hot takes here and I want to to try to, to make us think a little bit differently here.

00:55And over the two and- and half, and going on three, three years of, doing our auto QA tool, weve learned a lot. And the number one thing that I want to talk to you today, and I'm to be doing more on this is the boring rate that 80. That's scoring, right? That 80%. [QA coverage remark omitted] They don aint. It's the way that our human brains had to score things when we did things manually. And now with the power of AI, I would argue that you are doing your customers, your contact center agents in your company- company a massive beings, and, and also on Apple or Spotify, check it out on YouTube, too, because there's I'm kind of doing with this, but I- I'll try to make sure that I that I'm hitting both audiences, right. So the score was the output. [QA coverage remark omitted]

02:25That's how you're scoring your quality. But what we're finding with being able being able to do quantities, that that data point is just a small byproduct of the overall The end all be all and I'm going to show you some of the ways that I believe that it is misleading you in some of the ways that you can kind of handle this and in a much. much better, more the productive way, and actually make real progress in the customer experience. So, again, this is not, if you want to buy some of this, I'm going to show you how you a lot, which is why I think it's more important that you look for a tool, but if you're looking for an AI-powered QA a lot of things, but they're just doing the same thing at a scale. So now you're getting, I think, more the misleading complexion information if you're not looking at the proper things.

04:00And today, today disadvant to go over what I think those proper things are. I think that this is really, really good. so for most of you and again for those of you listening, I just have a little Excel spreadsheet up on the on the on my screen and this is how you guys were the majority of us are still scoring calls, right? this is the number one thing that now kind of goes away and I'm really literally just jumping in head first with this I know so many of you out there have because I get this And that poses so many problems, right? The first thing is, we're going to look at the dispute. This is the number-one thing that now kind, kind of, goes away. and I'm really, literally just jumping in head first with this, I know so many of you out there have, because I get this question every time, hey, can, can the agent actually dispute The call from Otto, QA, and again, disputes were important when you were scoring and maybe incentivizing your agent off of three calls per agent per week, maybe nine calls Now, we don't have to dispute things because we have mass quantities of data and with the mass quantities of data, we can look at the bigger picture.

05:11So one call doesn't mean anything anymore. One call is just an insignificant data point. So side, you now in the macro, so we don't disappear anymore, we look at all of the data. so again, if you had an agent scored one call and you know you that agent said hey you know what I did verify that customer's address and maybe you have a conversation Now, you can the main point, there's no reason anymore And that's why I think it's doing It's much cheaper and it's much more informative. other cool things that an auto QA kind of tool can do, that a human being can. And I'm going to go kind of on some of these. Again, I'm not going to spend a lot of time on this. [QA coverage remark omitted] Right? So, again, how many times would you maybe have a... They'll write up maybe two sentences on what happened.

07:20Well, you don don't do more. Now, now we can utilize a call summary, and the call summary S in itself is, S again, all of this stuff is building data and that's the change that QA form that we think is important. All that-that, that is, is, is something to now feed all of the data. of the data. It's just one little set of data points, and the more information that we can collect on the form, the more powerful. So this call summary, now, maybe if there's an exception, this was an auto fail, you want We're. Not. Going to read all these summaries. These summaries are for building data for some of the things that you can do down the road or kind of as we kind of build this thing out so that you and kind of start to see what is actually happening in the overall call center, what is happening with your teams teams, what is happening. With that at scale, not just kind of for one data set.

08:26How many of you guys have a, have a question on your form that says, you know, the agent was the agent empathetic, right? Did they use these empathetic terms? Like quote, be more em K, but the problem is when we're doing human being human scored AI, human scored QA, we're scoring words and phrases, we're scoring what they've said. And even Rif an agent says something like, Mrs. Smith, yeah, I'm really, really sorry to hear that. Because So, they did say itiasco they were empathetic. That doesn't have to happen anymore with AIPowerQA, That whole I'm really sorry to hear that phrase that we're scoring and we're calling that empathy should be done. Why that has hesitated. or did the agent show empathy? We Did the agent then respond in kind where raise or lower the tone, raise the lowered sentiment. We can look for many outcomes We're not scoring phrases that don't have any impact So now you So again, it's more important, it's starting to collect data on the overall sentiment.

10:42So looking at, you know, what percentage of the time does our agent have a customer who

[Gap 10:50–11:05: no reliable transcription]

11:05of what one of your agents are actually turning negative into positive, and then what ones are turning positive into negative? Those are important,.. Those are important data points, things that you can very easily pick up with AI, look at the sentiment, and this is the stuff, instead of just giving an overall sentiment score that I think really moves the needle and is really, really important. This is the stuff, and again, I don't want to make, this is sales on auto QA, but the things that we thought were really important I think is really cool. And when you're looking for, a tool, make sure you ask these type of questions. How many of you have to, you know, take those three, three calls that you scored for Tim or for Susie And So again, you're not summarizing three calls now you're summarizing 300 calls for an agent

[Gap 12:30–12:46: no reliable transcription]

12:46last month, yesterday, and

[Gap 12:53–13:16: no reliable transcription]

13:16we're wasting time coaching off of that don't mean anything when we have mass amounts of data that we can now use to really help. This is another another amazing thing that you can do with AI Power QA that you should be asking all of your vendors that you're looking for is after call surveys These surveys are a joke in my opinion. I think that they cause a lot of frustration. you can use AI to predict CSAT and NPS, right? So again, we've taken about 300,000 different QA scores from our customers that were manually done and found out and said, hey, AI, what would this agent or this customer actually probably do, do on that call based on what happened on that transcript, based on how these were calibrated. So you don't have to send out after-call surveys. You can now predict NPS and C-SAP with AI, right?

[Gap 14:23–14:31: no reliable transcription]

14:31Looking we see that on these refund requests and these billing b disputes, he was actually at a 78, or a 79 and a 74. He really struggled. but he rocked these password resets and order status calls. So again, this will weigh your scoring so that now you understand what is actually happening by to