Best ChatGPT Prompts to Improve Your Contact Center!

In "Revolutionizing Call Centers with ChatGPT Prompts," we unveil the power of AI in redefining call center operations. This episode is a goldmine for managers and supervisors, as we share the best prompts to optimize and streamline your operations.

Dive deep with us into the crux of Overall Operations to decipher key performance metrics and understand their implications. Learn how Workforce Management can be demystified with the Erlang C formula and ChatGPT. Uncover the transformative potential of targeted Training simulations based on real-world scenarios. 

Enhance agent responsiveness with the Agent Assist system, drawing from the most common queries and challenges. Explore the nuances of Analytics, using ChatGPT to identify trends, bottlenecks, and areas ripe for improvement. 

Delve into QA to understand how to ensure consistent, top-tier service quality. And finally, discover the art of Coaching, using AI insights to provide pinpointed feedback for agent growth. 

This isn't just about harnessing AI but maximizing its potential for tangible results. We're giving away the best prompts to supercharge your contact center. 

Tune in, apply, and witness the transformation!

If you are looking for USA outsourced customer service or sales support, we here at Expivia would really like to help you support your customers.
Please check us out at expiviausa.com, or email us at info@expivia.net!

Follow Tom: @tlaird_expivia
Join our Facebook Call Center Community: www.facebook.com/callcentergeek
Connect on LinkedIn: https://www.linkedin.com/in/tlairdexpivia/
Follow on TikTok: https://www.tiktok.com/@callcenter_geek
Linkedin Group: https://www.linkedin.com/groups/9041993/
Watch us: Advice from a Call Center Geek Youtube Channel

Transcript

Speaker 1: 0:25

This is advice from a call center geek a weekly podcast with a focus on all things call center. We'll cover it all, from call center operations, hiring, culture, technology and education. We're here to give you actionable items to improve the quality of yours and your customer's experience. This is an evolving industry with creative minds and ambitious people like this guy. Not only is his passion call center operations, but he's our host. He's the CEO of Expedia Interaction Marketing Group and the call center geek himself, tom Laird.

Speaker 2: 0:59

Welcome back everybody to another episode of advice from a call center geek the call center and contact center podcast. We try to give you some actionable items. I got some actionable items for you today. Take back in your contact center, improve the overall quality, improve the agent experience, hopefully improve the customer experience as well. If you do not know me by now, my name is Tom Laird. I am the CEO here at Expedia Interaction Marketing and Expedia Digital Got to say that now that we've kind of added that through in added that in we are a 600c contact center outsourcer and we're doing some things on the digital side that I can talk a little bit about as well.

Speaker 2: 1:36

I hope everybody is rocking it. I wasn't planning on doing this today. I just got off a great podcast with the, the, with CX, cx live, and then have a I don't know, have about a 45 minute window here. So I wanted to talk about this because I was really excited about getting this out. And I want to talk to you guys about chat, gpt and everybody loves the AI aspect of this. I want to give you some of the prompts that we have been playing around with, that we have been utilizing that we are that I personally use in the analytic piece, in the workforce management piece, in the overall operations of our contact center. I want to show you what we're doing with QA and trying to fully automate quality assurance using actual scorecards, and talk that through a little bit. I did a full podcast on that a little bit ago. But I want to get a little bit deeper, as we're as we're a little bit more into this. I will have the physical prompts in the show notes when this is done. So if you guys want to play around with some of the stuff, you can, and let me dig right into this.

Speaker 2: 2:41

So, first thing is the cleaner your data, and let me, let me set this up. I'm going to do this as easy as possible, right? So this is for all of you. This is for the call center manager, the call center supervisor, the QA person, right? You need no programming for this. You don't need APIs for this, right? Everything here can be done in the chat, the 3.5 or the 4.0 on the desktop version. You don't need to be any. You can play around with this on your phone, right? So I want to set that, the table with that up. That's for. This is for everybody. This isn't? You know some crazy in depth, behind the scenes stuff that you can do. We'll do a podcast on that, because that's what we're doing for our QA piece, but for this I want this to be for kind of everybody.

Speaker 2: 3:24

Second, well, this is stuff that I actually use, right, I have played with this so much that it's become kind of part of me, even from the financial aspect. I'm a BPO from the financial aspect. I don't know if any of you guys know this, but, like, if you have a budget at home, if you have again our financials, I just literally copy and paste the spreadsheet, throw the spreadsheet into chat GBT and start hammering it with questions. Where can we be more efficient? Where can we save money? Where can I increase my gross revenue? Where can our bottom line get better? Where are we inefficient when it comes to our certain programs? What client should I raise prices on? What client should I add more of to? You know what I mean. You can add all that stuff in there. From a budgetary standpoint, you can do that too, but I want to talk again about the call center.

Speaker 2: 4:15

So let me start with this. Have as much data as you possibly can. Spreadsheets are fine. So if you have, you know, csat scores and NPS scores and average handle time and all these different metrics. You know, take some time and kind of put them together and I'm going to tell you exactly what you need to do each of these prompts. So let's get this going.

Speaker 2: 4:36

So, from an I guess I called this first one like overall operations, right, and you know, this just is kind of and again I'm just looking at my notes you know streamline, optimizing the function of the contact center, the things that you know that you would need, right. So, historical performance metrics like average handle time, first call resolution, csat scores, nps scores, service level, any of those metrics right. The more metrics you have right, the more robust your outputs are going to be. If you have current feedback, right. So if you have done CSAT surveys, if you've done NPS surveys, have the feedback actually in there for each of the reps. You know abandonment rates, all of those kind of metrics that you that you can think about. And here's some of the prompts that you can play around with, right.

Speaker 2: 5:23

So you know, given the historical performance, the of, I'm sorry, given the given the historical performance metrics and feedback, identify three areas of our contact center operations that needs immediate attention, right? So it's going to parse through all that data and then, four seconds, talk about some things that that you are really struggling on. You know you can do things like you know. One that I wrote was you know, based on the tools and platforms mentioned, are there any efficiencies or redundancies that might be affecting agent performance or customer experience? That goes especially back into all of your feedback from the customers and your your CSAT scores. You know, considering hold times and abandonment rates, suggest strategies and what metrics should we really be looking at to improve these and improve the caller experience? So, again, more high level, not really digging in the weeds. I'm going to dig in the weeds here in a second, but those are some things from a basic standpoint. At least it gets you thinking of the the actual data that you're going to need.

Speaker 2: 6:25

All right, let's talk about WFM. I think this is really cool. I know a ton of you guys are using spreadsheets for everything, right? So many contact centers, from a QA standpoint, from a workforce management standpoint, from a staffing standpoint, are using spreadsheets. We have found this out in all the research we have done for our QA AI platform that that we're trying to build out for everybody that specifically has spreadsheets. But workforce management is a huge piece of Excel right is a huge piece of the contact center you know, especially for the smaller contact center. So you know some of the things that you you you can utilize this for this.

Speaker 2: 7:00

If you do not know what an Erlang C calculator is, there's a lot of them on the internet. But it basically allows you to put in your handle time, your service level, that you would like put in your the calls that you want per interval, and it will spit out the calculation that you need for the amount of headcount for each specific interval. But you can ask chat GPT to do the same exact thing without doing that and you can kind of make it a little bit more your own. So you know, if you have you need the hourly call volumes, right, I mean the average handle time shrinkage. You can add in there as well, which you probably should the service level, right, and then you can even have in your average wait time and you can say things like, using the Erlang C formula and the data provided, calculate the optimal number of agents required per hour to meet our service level, and it's going to spit out by interval what you need. If you are struggling with time, to how to figure that out, if you're not really sure, even if you are using an Erlang C, this is a way that you can make it a little bit more your own and you can even, you know, make this even more specific. To say, we also need to schedule our lunch between 11 and 1. 20% of our reps go at 11., 50% go at noon and 30% go at one o'clock and then have it, then spit out all that, so you can get almost as specific as you can in workforce management. Now, it's not a workforce management tool, but if you are a smaller contact center, it can be extremely powerful. Right To be able to utilize that as an early on see and as a workforce management tool. Another thing that you can utilize that I know some companies are building out on a robust standpoint.

Speaker 2: 8:38

But the training aspect, right. So, looking at knowledge gaps, looking at real-world scenarios that you can kind of role-play in I don't wanna say virtual reality world right. So, again, if you have agent performance metrics, especially pre and post training, if you have any of those scoring and either surveys that you may take during training, if you have getting feedback on what people struggle with. What are some of the topics that you make sure that you need to train and educate on and what are some of the common customer complaints right and what are some of the real-world scenarios that your agents face and again, really cool, like prompt one. Given the performance metrics and post training feedback, can you look at some of the current gaps in our training and can you recommend some of the new modules that we need to make sure that we focus on right and have it really kind of almost look at your training? If you wanna throw in your entire training deck right From a copy and paste standpoint, you could do that, have a kind of look at that and research it.

Speaker 2: 9:45

Things we don't really think about, but I love this is based on common customer complaints and scenarios. Can you generate potential training scenarios or simulations, right? And then you get to talk back and forth and I know there's a bunch of companies that are building stuff like super robust. But let's just say, hey, we have a scenario where chat GBT, there is a coaching scenario. We wanna focus on a customer that does not wanna give their social security card or, I'm sorry, their number. Can you please role play that with our agent, right, and then have the agent talk, the more robust you want to get with your prompt, you know you can say hey, we have a 65 year old male from Pennsylvania who is a little bit crotchety and does not want to give his social security number. When we are verifying, we need to. If we cannot get the social security number, we need to verify three other pieces of information which is dot dot dot. Please take us through that scenario and then, boom, you have a built in kind of scenario that you can have those conversations with with a fake pretend customer. Really cool. Now again doing that just with chat. Now voice is coming, but at this point you know you could do this tomorrow, right, with the couple of your agents, which I think would be really cool.

Speaker 2: 11:07

So how about agent assist? Right, and we know that agent assist is a huge tool, but you know we can kind of do this the, I guess the poor man's agent assist as well. If we're looking at, if we want to equip with our agents with instant right information, right, so we need to know, you know what are the some of the questions, the most common questions that our customers are asking what are some agent feedback and how they answer these challenging you know issues and what are some of the points of delay right looking at. You know we always struggle with this. This is a question that we struggle to train with and there's a lot of mute and hold during this right.

Speaker 2: 11:42

And then you know, ask, say you know, using the frequent customer queries and agent feedback, suggest a quick, quick reference scripts or guidelines to assist our agents in resolving these issues. And you can kind of just kind of script them out. So that way again, whether you have it in your database, whether you, they just have that information, it's a quick and easy way to create scripts because, I mean, you guys know this, sometimes it takes forever. You got to go through marketing, you got to have people print things out, you have to have people think things through and it just drags on forever. And I think that this is kind of an easy way for some very basic piece of information that you guys struggle with, that you see, from a QA, from an analytics standpoint, that you can give some instant feedback and instant information. And I think you know that's really cool. Analytics, my baby, my favorite, right. You guys know that.

Speaker 2: 12:32

So first things that you can do is, if you have any platform so, let's say, you're a little bit bigger you know we utilize, we build out our analytics right using chat, gbt, chat GBT can you please give me all the category, subcategories, trending keywords and phrases needed for a retail toy company that is looking to improve C sat nps and get any type of relevant marketing data that they can find to give to their sales teams? Right? And then it will build out the categories, the subcategories, the trending keywords and the phrases that we can put in there. And now we're looking at all of that information, right, and we don't have to look through all kinds of previous data to try to come up with some of those things. Now, if you have, if you've already started your analytics right, you can print out all the trending keywords and phrases, throw that in there, make sure that you're asking for some specific things and it will pop out. You know what are those keywords, what are those phrases. How many times was that said? Really, really cool.

Speaker 2: 13:29

So, again, anything that has to do with the data that you can get trending keywords, trending phrases you know, really, giving a in depth report of what we do, this for clients all the time and we're using chat GBT word before we would just build this ourselves. So, you know, I will take, or in our, our team will take, all the trending keywords and phrases. You know, whether that's 2000 of them or 1500 of them or whatever that is, we will throw that into chat GBT. We talked to our clients about you know what specific things are they looking for marketing, data, pain points, issues with logistics, any of that, those types of things and we ask those questions and that's how we build out our reports. We'll build our reports and say, hey, here's the things that it came out with. We say question answers, question answers, question answers.

Speaker 2: 14:17

Try to make it actionable. So if you have an analytics platform, you could make that stuff extremely actionable by asking chat GBT to review all of those. You know those bubbles that we all love or hate. Filling all, pulling all those out and having that really do analytics on your analytics right, and it's made the tool extremely more useful for us from from that standpoint. Really really cool use case, really easy, really cheap. I think it's pretty easy. Qa, all right. This is something again that I think I probably have the most amount of experience on now recently, as we are trying to build out a full QA platform SaaS product for contact centers that are 50 seats and under that are using Excel spreadsheets.

Speaker 2: 15:11

Having said that, you can do a lot of this stuff right now on your own. You can, number one, have it, review your scorecard, talk about what the most important pieces of your culture or the aspects of what that scorecard should be showing. So, for example, chatgpd. Here's our scorecard. Copy and paste it, throw it in. We want to focus on giving a high-end customer experience, on making our average handle time as short as possible and creating as much value as we can for our customer, while making sure that we have a cross-sell upsell component or whatever this is. So, whatever your aspects of what you want to do with your scorecard, it will rewrite or change the nuanced scorecard to make sure that you're hitting those things from your cultural aspect. We want to make sure that, again, to be able to use technology to figure this stuff out is amazing.

Speaker 2: 16:04

And then the second thing is that and we're doing this is you can ask chatGPD. Now you need the transcripts of the call. So this is a big issue. There's a ton of companies out there. One is, and I'm going to tell you is DeepGram, which is this crazy, super cheap. They will give you a ton of free credits as well so that you can test this and play with it, but it will take your recording in literally two seconds, give you a full transcript, take your transcript and then ask chatGPD to score your call based on the form that they just created, and it's going to score your card, score your calls, and I think that this is where I say that QA is going away here extremely quickly. But a really, really cool way to use chatGPD as well, I think for supervisors you front end on the floor, supervisors that the coaching aspect, this can really come into play as well.

Speaker 2: 17:03

If we have some performance metrics on where agents are struggling on whether it's CSAT, nps, whether it's their QA, whether it is getting more specific. They're not listening for closing signals, they don't read a certain part of the disclosure all the time, they're not verifying the proper amount of information, getting that type of feedback, having notes that you've already typed up from previous coaching sessions with them having two or three transcripts that kind of show this as well, and then you don't even need the transcript if you want to just utilize like, hey, listen, here's the last 15 times I talked to this agent we're focusing on. We have these three, four issues and considering the in chatGBT say, considering the feedback, the metrics that we need to hit the feedback that we've given outline a coaching framework that targets this specific agent who doesn't take coaching really well and can be a little bit more rough. Put all that in there. It will give you some really good, at least thoughts and outputs to kind of figure that out. Provide actionable steps or exercises that coaches can employ that make specific changes to agents that are not doing X.

Speaker 2: 18:11

All of this stuff that can help your, your training aspect. It can get really specific with some tips and tactics. You can ask it to use specific coaching methods. Right, there's the grow method, there's the Oscar method. All of those Take a transcript and say, hey, can you please use the Oscar method? Look at this call and give me coaching points to talk to my agents about. Right, it will do that for you and it will do that instantly and it will be. It's really on point. It's really, I think, really helped. You know a lot of the coaching aspects of what we're doing, and then I want to talk to you guys too.

Speaker 2: 18:49

Then about the hiring aspect, and let me get to my notes here on hiring. Okay, a lot of you guys keep do exit interviews and you, you kind of keep track of all that stuff, right. But if you, if you can put that into an Excel and ask, chat to you, hey, what are the main reasons, what are the top five reasons that agents are leaving our organization, and can you give me three or four you know tips to improve that, to improve our retention? Right, have it. Look through. You know any type of data that you have on why agents are struggling? You know, maybe there's something from a specific supervisor that you're not seeing, maybe there's something from an overall trend of a program, maybe you're hiring wrong, right, so I don't know. There's some things that I think you can, you can do there. You know, looking at the agents that work as well, right, so you know, looking at the, the best agents that you have, and in kind of talking through some of the feedback things that you've given them, you know how is that agent looking at kind of their reviews, why are they good, right, and kind of giving chat to you to see that, to kind of see and build your overall profile of you know what makes a great sales rep, what makes a good customer service rep in your organization and kind of really, you know, kind of build that out a little bit more. But I think looking for common traits is what chat, gbt and AI is really, really good at and I think that can be some some things that can help you. So again, this is kind of you know where my head is at with a lot of this stuff and I don't want to make it hard. I want you know frontline supervisors, I want call center managers, I want team leads to be able to you know, or at least to be able to give you guys some stuff that I think can help, can add to the productivity that you're doing and help you get through your day and improve the context and improve the agent experience as well. I mean there's so many ways that we could go with this and so many other things that I'm sure I will talk about in the future, but, you know, really excited.

Speaker 2: 20:52

A couple questions here I have on on TikTok Blenya. I am a CEO of a contact center, so I run and the name of the company is Expedia, so we're about a 600 seat call center outsourcer located in ERPA. So we work for a lot of different clients that have outsourced their customer support, their customer service. So that's kind of you know, I guess, my 30ish years of experience with this and trying to do it a little bit differently than how a lot of people think of the contact center. The call center is hopefully what's coming across with a lot of the content that I do.

Speaker 2: 21:30

Again, this is an episode of the. No, it's not, it's Expedia, exp IVIA. I get a lot, I get Expedia a lot. I'm surprised they probably haven't sued us, but we've been around for about 14 years now so I don't think it's happening. Well, thank you, I appreciate that. You know, I appreciate the you know kind of the shout out a lot.

Speaker 2: 21:48

I know a lot of times on TikTok you don't know what you're going to get from a. You know if somebody's going to say something silly, somebody say something mean or something nice, and it's nice when people say something, say something nice. But again, if you guys are into the call center, contact center world, you can check out the podcast. But we're doing right now advice from a call center geek. Check that out.

Speaker 2: 22:06

Check me out on LinkedIn as well. My profile on TikTok has my link tree, so all the content that I have on TikTok here and on LinkedIn and on Reels and all over the places is there. So you know, I appreciate if you guys check that out and you know, let me know what you think. But if that's it, that's all I got guys. I hope that this added some value to everybody. I think that. You know I did not really announce this. I just kind of had some extra time through this up here, so I know there wasn't a lot of people live here on LinkedIn, but I think that this one will get some traction here as we go in kind of post production and I'm excited to see how this one goes. So thank you, guys. I'm going to hang out here on TikTok, but LinkedIn We'll see you guys.