Why You Should Not be Scoring 100% of Your QA Calls

00:00Tom Laird here. You've heard me break down break down the podcast time and time again. Now it's time for me to help you with 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. It's the same honest advice you hear on the geek applied specifically to your operation. Visit ExpiviaDigital.com to get started. That's ExpiviaDigital.com. Welcome back everybody to another episode of advice from a call center geek, the call center and contact center podcast, where we try to give you action by items to take back at your contact center, improve your overall quality, improve the agent experience, hopefully improve the customer experience as well. I guess that's the whole point of this, right? For those of you who do not know me by now, my name is Tom Laird.

00:49I'm the CEO of Expivia Interaction Marketing. We are a 500 seat contact center outsourcer located here in the States. from home and on-prem, and also the founder and CEO of auto QA, where we're fully scoring contact center quality calls using AI. And I want to talk a lot about auto QA today. Obviously, we're using this with all of our customers at Expivia. But this question, and I do probably, over the last three years, I've probably done one podcast a year on this topic. And it's constantly, I don't wanna say evolving, but it's unfolding to be more and more true as we see costs of what some of these other QA companies are doing. And again, I don't wanna make this a sales, this is not a sales podcast for auto QA, although, you know, we're definitely gonna talk about some of the benefits of doing QA this way, right?

01:48using AI in the most, I think, efficient way possible to make sure that you're getting ROI instantly. So whether you use us, whether you use somebody else, I think this is a main topic that you need to be informed about and you need to be able to ask questions when you're talking to a provider or somebody that you're looking to possibly bring on as a vendor to do QA with. So, there's two main ways to kind of jump right into this thing, right? There's two main ways that people are scoring calls and billing for it from a kind of technology partner platform format, right? And the main one still is the old school, you know, everybody's always used it, technology and CCaaS and CX, which is your C license, right? [QA coverage remark omitted] I've talked to a head of people at CCaaS platforms. I know a lot of those type of individuals and they have picked my brain and we've had conversations on this on how to get the cost down for QA. [QA coverage remark omitted] So what we have found is that they do, they come nowhere near using using overall, they don't use the best models, right? They use, there's a lot of different, tactics with caching and things the cost down.

03:47But it. It's not going to matter when you have a, you know, 250 seats center, you know, taking three minute, four minute calls, right? The volume of call is just so a I guess technically that wouldn't matter, I mean, if you have a longer call, it's less, it's a. It's just the amount of people, the amount of scale, the amount of cost a. So, when you, go-on to these websites, you'll see for most of them, we kind of respect, right? They're honest.

[Gap 04:19–04:29: no reliable transcription]

04:29we're going to, you

[Gap 04:36–04:59: no reliable transcription]

04:59[pricing remark omitted] We're going to be more cost effective than a seat license. So what we do is we utilize a usage model. So you pay for how many calls do you want scored? [QA coverage remark omitted] And my answer to that is, yeah. I mean, that's not a bad thing. [QA coverage remark omitted]

[Gap 06:05–06:27: no reliable transcription]

06:27they can predict or become with a margin of error of plus or minus 3%. So meaning, if auto QA scores scores something at an 87, we're gonna be, maybe you're, and we have found that human beings, if I score a call and whoever's listening to this scores a call and your QA and your contact center manager scores a call, the average is five to 7% that you're gonna be off. So that's why we say we are more accurate

[Gap 07:23–07:53: no reliable transcription]

07:5383 right in that 85 so you can. [QA coverage remark omitted] So again, I don't know., hopefully I'm explaining thatciating that well a lot. [pricing remark omitted] You should be using certain percentage to give you the highest language. [QA coverage remark omitted] So it's, it's crazy crazy to, to kind of kind of bring up here too, and let me, let me bring engineer my notes here, um, a couple of things that I want to kind of bring up here- So it's, it's crazy to, to, to kind of bring up here too, but when you, you know, it makes, it makes total, I think, it makes, it, it, it, it, it makes, it, it, it, it, it, There's, there's a couple things though

[Gap 10:08–10:38: no reliable transcription]

10:38And I think that's kind of the- the biggest kind of fallback there, right? Is, we can kind of tell you exactly what you need to do to be in that kind of space to have your margin of error be really where it's supposed to be. So let me calm some people here too, right? So, and again, just quickly to get back to that, We can then just take the two questions, right? So you have two questions that have to be compliant. [QA coverage remark omitted] 5,000 seats, yeah, I'm going to be, or a usage model will probably be too expensive for you, and you got to kind of just stick with because the models just won't allow that. But for most contact centers, 30 seat centers, 50 RZ centers, you should not be paying for a seat license.

11:51And you should be just using a usage model, using a lot of large numbers. And again, we have, I have a full blog post that I did today that kind of gives an example. So a 60,000, call, monthly call center, right? So a call center that's taking 60,000 calls, a call center that's taking 60,000 calls, right? What we would recommend at auto, right? And we're saying that for this, you know and again, about a five minute confidence level plus or 3% margin of error, that would be just scoring 1789 calls per month. month. That gets you to that 99% accuracy. Our algorithm also spreads it over all of. Your agents make make sure that we hit that. Sometimes depending on the amount

[Gap 12:43–12:57: no reliable transcription]

12:57[pricing remark omitted] I mean, I don't know how else I can put that. have done a horrible job kind of the, like, with the content that we've been, with the content and

[Gap 13:51–14:17: no reliable transcription]

14:17is, too, is it 85%? Because if I do it smaller, I actually get with a better large language model, my accuracy is actually better. Look at the blog post that I put out today, it's on the auto QA site, really. It also gives you things, if you're going to do this on your that you have a sample of your own right making sure that you have a sample that is representative of kind of. Looking at stratifying your sample having automated selection and this is, and this is the stuff that So when we have an integration, we pull it in. All of that math is kind of already done for you. So you know that you're getting a fully scoped and accurate amount of universe of your QA. So the three things again that you're probably doing right now. Now, especially if you're not using AI power. QA rate that you're making.

15:13Making mistakes with us is when you kind of your QA cherry picks calls. right? We all all know this, right. They pick calls based on duration. They don, don't want the 12 minute. Minute call, they'll take that three and a half minute call, right? They'll lookIlook at things things, like, you know, we're just going to listen to customer complaints. [QA coverage remark omitted] and again, you need to score 100, but we have to have an algorithm set that matches how many, many calls that we need per month and the making sure that we need to score 100.

16:17So there's kind of two levels to this, there's is the overall contact center, then there's the agent. They're normally, within about 5% to 7% number-wise of number-wise, if the rumber is on your overall contact center, then Lied. [QA coverage remark omitted] Again, I hope these questions start to get asked when you're looking at these type of tools. that level see that they're going to give you, what happens if they're four if your form isn't accurate do they change it for you to the change questions and set up to make sure you get the most accurate QA and you're getting the most, I mean, let's just keep it there, that you're getting the most accurate amount of data, and data that DM, DM me with that, I can just kind of shoot you that and that goes pretty crazy deep into all the math that we've done with this, statistics, looking at the law of large numbers. [QA coverage remark omitted] What it's cracked up to be. And a lot of times you're getting much less accurate data than if you're scoring a sample size with the best large language models with the highest margin of error or the lowest margin of error, depending on how you.

19:08Look at it. All right, guys, thank you very much. Hopefully that just kind of at least get you thinking a little bit about this and I'll talk to everybody next week.