Speech Analytics Challenges and 'That Slap'

0:27
Welcome back everyone to the next episode of Throwing Shade. And we've had some interesting feedback and lots of responses to the last episode. So thanks very much if you downloaded, if you subscribed to the button, make sure that you do hit the subscribe button if you'd like to hear more of the show. And today's episode, we're going to focus on a couple of different bits and pieces as usual. The first of which is going to be around speech analytics, shameless plug for for what we do, but let's be honest to me. Like there's there's a real danger when working with speech analytics that you can get it wrong, right? Yeah, 100%. And I think the, the only reason why we do speech analytics is because people have done it wrong. Like, that's like the big driver. So I, I came from, you know, when I, when I was working as an analyst, I came from, like, kind of a marketing background. And through working in contact centres, I was always interested in coaching, and how you get the best out of people. And, and it wasn't that I'd seen lots of things wrong in those different fields. It was it was really around well, how can you how can you improve and optimise the different processes, but with speech analytics, there were just some glaring errors that I saw. If you go back to when I started first working in speech with speech is 10 years ago, about 10, 10-11 years ago. Now, when I started doing like, the first speach my first speech analytics projects, that was mainly in kind of voice-led IVR started in voice led-IVR. And then and then some around, and then some around speech, speech recognition and speech to text for within within service, and sales. And back then, like, it was terrible. It was really bad, you got really, really bad results. Most of the time. The speech to text transcription wasn't wasn't particularly great, And I was like, pretty badly burned by it, like, I was I was always hopeful that the the technology would catch up. But I was just astonished by like, the, the the ineffectiveness of the applications even, you know, I always say like we did, voice led voice lead ID and V. So voice-led identification. And still to this day, like, I absolutely bloody hate that. Because like, if I'm, if I'm at home, fair enough. Okay, but if I'm working in my office, do I really want to shout out my sort Code and account number? Or do I want to shout out my date of birth? I try and remember, like, is it like for it? Is it? Is it six digit or is eight digit kinda thing? Yeah. And that's what we were trying to do. But then the application wasn't great. And the other the overall solution, because when you went then went through to the agent, it was so untrustworthy, the agent always had to double check anyway. So you still have to say, so you still have to kind of say it twice. Anyway. So I always thought that was really strange. And then And then, you know, moving moving through into like, the, the speech analytics part, you know, I, so it was the the automated kind of identification was the one first, which I always thought was really strange. But then, a few years later, we started moving into speech to text. And we were we were doing text analytics and speech analytics. And we will look at we worked on on different systems for different customers. And 99.9% of the work that was being done was categorization. So like categorising calls, but like, you can do that with like, a bit of paper. You do it or not with a bit of paper, but you could do it with a data capture tool. So like, when in one of my former roles I built, I built me and the team built, I shouldn't say because somebody else built it. But maybe the team built like access kind of front end that said, you know, what was the primary call reason and what was the secondary call reason? And what was the did, was there another do another transact type of transaction on the call? What was that? And you just clicked on it? So the all those clicks don't take any time whatsoever. So why would you spend an obscene amount of money on speech Analytics as well? And that was another thing. But the speech analytics like the actual transcription cost an absolute fortune, right, so the benefits cases never really stacked. I don't even I don't even I don't even think I was particularly privy as to what those benefits cases were or what was sold, but then some sort of dream must have been sold. I think it was, I actually think it was, we'll help you understand what goes on in your calls. I think that was what what people were saying, however, like, you then went into a project where transcription accuracy was like 30%, or, you know, 20% 30% 40%. And then, and then an industry around speech Analytics was created. So then you had companies who had your 5, 10, 15 speech analysts, just categorising calls or listening to a call, and then categorising it because, obviously, like words were mis transcribed. So to create a category, like a grouping of words, you had to like listen to it and go, oh, like, yeah, HandBrake is handbrake for, for this particular category, though, he's been Miss transcribed as pancake. So I need to also include everything, every time it says pancake and all that type of stuff. And some projects would last 18 months or so. And therefore, like, everybody just lost the will to live with them. And I was probably about five or six years ago, I was there like scratching my head going, like, is this what speech analytics is like? Surely these like there's better? There's better ways and means to gain insight? Yeah, I think when you're when you're, especially when you're going back to like the subjectivity of the agent disposition in the calls, you can see why speech analytics will be an improvement on that. But that relies on the speech analytics, being able to actually understand what's going on in the conversation, and not taking an eternity to kind of build out like your business moves fast, right? So like, what does it have in a speech analytics platform that you can't adjust yourself, where you can't keep up with the business? So you've got up here, you know, professional services fees to kind of get it where you need it to be? And it still takes months to get there? I like it all on this date. Yeah, so sorry. Like my data capture tool, when we first when we first rolled it out? The the biggest the biggest stair transaction type was over. So other, other other, right. So like, I understand that people would just like press wherever it was just to go through. So there has to be some validation and stuff. And you are right. But I think that I think he was like, speech analytics to me, it always been either oversold or was under delivering on its promise. And I just was, I was just so frustrated all the time. And I think that's what you can see in the industry. Like, there's a lot of, you know, senior people in it, or procurement, or CTOs, who we speak to. And they are like, don't, don't talk to me about speech analytics, like I wasted three years of my life, that'll never get my promotion. I had to leave a job because of it. So yeah, there's a whole heap of stuff. There's a whole heap of reputational damage that went on. And that's what, that's where I think our starting point was around around speech, it was like, Well, how do you fix it? And how do we do things differently? Yeah, I think for me, what got me really excited with speech. And this is what really sort of turned me on to it was when you were talking about building business cases, open it, the AHT is an interesting one, you know, like, you know, that by increasing your AHT, and you can improve your your FCR rate, then what your business can make that stack up without having the stats to understand what that business case looks like. And because they can't prove it, they never end up doing it. Which means that everything just stays as it is or gets worse. And then the idea that you could use like speech analytics to truly understand what that business case actually looks like by being able to correctly divide those calls off. Yeah, if the speech analytics to the right price, why wouldn't you do it? Yeah. 100%. And I was like, when I think I'll always remember, like, we did it, we did the speech analytics project that went okay. And I'll always remember, I was working with Sam Phin, who's, who works for who works for TMAC. And I was speaking to him and I was talking to him about, like, what he thinks is important when it comes to speech. Now, he's had a lot of experience when it comes to like different using different systems. And I'll always remember, he was like, let's do some analysis on acoustics. So let's do some analysis on talk over and silence time. And let's not let's just first of all, forget the not forget the speech to text and then natural language processing. Let's just look at The acoustics and see what happens within the acoustics. And Mate, it was an absolute revelation. Because what you can then do is that we build a model, which was the likelihood to be a promoter or a detractor based on the acoustics of the call out, sorry, based on all of different metadata, handling time transfer rate and stuff. But one things that were built in, were acoustics. And the number one thing, the number one, like most influential thing was talk over talk over was highly predictive, when it came to NPS and silent sign was highly predictive. So that just goes to show that sometimes it's not even the words that you're saying. It's how you're saying them, and the tempo through which you say them. It's almost as important as the words themselves. Yeah. But if you're only QA like, 1% of your calls, how do you know what your talk over time is? Like? You can't it's not even something even if you were listening to the calls manually. It's not even something that you could you could measure, as a team leader, he just got a finger in the air gesture, right? Like, I think your customers end up talking, you end up talking with your customers quite a lot of the time. How do you how do you can't coach somebody on that? Because it's subjective. It's your, like, you're going to get a lot of pushback from that, right? 100%. And that's actually yeah, that's where, like, I was going, Oh, I personally was going Oh, s***, like, so. So there's a whole heap of influential stuff, probably the more influential stuff that number one is not about quality and compliance. It's about quantifying the dynamics of the call. And then number two, it's how, how would you ever quantify that? If it wasn't for some sort of speech analytics or acoustic analytics event? Like, I'll give you another really interesting thing. So like, you know, since we met, Jez groom, from, from Cowry like, you know, speaking to those guys was, was always really, really interesting. And they talk around cognitive overload, cognitive overload, like, you know, really interesting, really interesting to me now, because, obviously, like, I'm probably really, really bad at just talking very indiscriminate amount of time. And then everybody sort of trails off. So I do try and like, get people's attention now and get people's feedback and stuff. And get people's feedback. And it drives my wife, Lauren Absolutely nuts, She's just like, I'm still listening to you. She's like, All right, well, your eyes are closed. But anyway, so this cognitive feedback stuff. So even when you even even old school speed, even old school audio recordings, where you've only got a single channel, if you've got diarisation algorithms, which can do speaker separation. You can even with that, you can do How long is bloody Jimmy Hosang talking versus how long is Paul banks talking and what's the what's the what the difference between those two things and what's the optimum amount of advisor talk to, to customer talk, and so, so even like the, the with speech, speech, transcription is rubbish stage, there's still all of that value that you could be getting from acoustic analysis. And I think that that, that really kind of opened my eyes opened my eyes to and actually, you know, we we do, we do, we do speaker separation and stuff. Now, speaking, separation isn't easy. It's really, really hard. It's really like super intensive from a, from a computational point of view. But once you've done it, like it is, is more kind of accurate and more binary than that than the than the all of the different words that you could possibly do and all of those permutations. So if you, if you kind of went to a client, like just the acoustic analysis, the talk over the silence, and all of this type of stuff that you could get from going through a speech product is highly valuable, right? Think about how much that could have been been done, you know, five to 10 years ago, which hasn't been because all the while people have been worried about word accuracy. Yeah, that's that's one of your big bug-bears right? Word accuracy. Oh, my God. Like I hate it. I hate it because like, so this, this like, what? It's one of the first things I get asked, it's one of the first things that I get asked when I speak to you do but you disappear up your own backside, on word accuracy. And it's because like, it's because of, I'm just being honest, like, it's because it's been missold, or And or, it's the kind of the luck of imagination of the analyst like to think of other ways in which like, they can use that data. So how I, how I always put it is, so first of all word accuracy, if, if your transcription engine is accurately transcribing 'the', 'and' 'or' all that all that type of rubbish, right? We could do all of those, he could do all of those, but those are indicative or contextual, of anything other than just kind of just just just normal words, stop words, really. So like, first of all, you need to strip all the strip all of that stuff out, like regardless. So that that could be, you know, that could be part of you, you were you, you your Word Accuracy. The second thing is, why would I want to put it because the focus has been around key word accuracy there, like key words, like I want to put complaint with washing machine, or I want to put refund next to trainers, right? Because there's been such a fast fascination of that, like people are like completely forgetting, like, who they are like, and what they should be doing and what they should be driving. Right. So I like to focus more on the predictive model side of things. So for me personally, like, I like to build predictive models, which is like the likelihood to be a promoter, or the likelihood to be a detractor or the likelihood that this conversation was compliant and things like that. Now, that does different things. The different thing that that does, right, is it makes the keyword transcription, the transcription and things like that, a little bit more naive. So you're not trying to look for keywords and phrases. You're just trying to say what words and what clusters of words are likely to predict, you know, a detractor or a promoter and things like that. And that that almost, it doesn't, it doesn't take away word accuracy. But it makes it less important because if you've if you're trying to predict who's a promoter, and what did I say before HandBrake is Misstranscribed as pancake? Well, it doesn't matter because it's kind of going, well pancake is a pancake is predictive of a detractor. Yeah, that's what's what you're trying to predict. So that type of that type of analysis really, really helps because it kind of takes away some of the word accuracy, just have the key word stuff. And those keywords are important keywords and categorization and building those such things is important. But you need another tool to your armour and another tool to your armour is, you know building, building predictive models, and you can build some really, really powerful like, models off the back of it. And then also, just time, like, everything takes an inordinate amount of time, like when you build in events. And so that's why I've always been a big proponent of topic modelling. So what topic modelling is, it's completely machine led way of categorising conversations, and it looks at all of your calls. And it looks at the clusters of words within those calls, and then tells you how many clusters there are, and then tells you what the words associated to those clusters are. Like, that's kind of it, that's what it does, right? And you can do some really, really clever, more clever stuff with it. Like, it can tell you, you know, you can, you can force a number of topics that you want, or it can tell you the self, there's loads of different stuff, but there's loads and loads of things that that can do now, where, where building categories, you get a lot of false negatives, because you've not been able to find like specific words and phrases that are associated with it say, complaint has been mis transcribed, doing doing topic modelling does two things. So number one is machine-led. So it happens like this, like day one, you could have topics and you could have you know, five topics, certain topics, you could have a 450 topics we built for, for one company across loads of different skills, across loads of different departments across loads of different quintile of the call and things like that. But you have them day one And what quintiles do, it's almost a bit more of a catch all. So they are a little bit more like you've got a bit a few more false positives, they're a bit more flaky, you've, you're capturing stuff that you wouldn't necessarily have captured. But they then inform how you build the events. So they'll tell you some interesting things that are going on in different parts of the call. And you'll see a cluster of words that are associated with it, and you'll go, oh, s***, this is really interesting. And when this topic is prevalent, man, my sentiment is negative, or my talk over is high. Right? Now I'm going to go and build an event. And this event is going to be, you know, high talkover, high negative impact, high negative sentiment. And these words and phrases, right, bring that all back from it, let's let's see what that actually is and what people were talking about. So it's a real neat way of a machine, giving you things to focus on. And it's all about expediting the categorisation element, which I believe is low value. And getting to the Insight element, which is around the the anatomy of the call, where you can drive value, how you can do coaching, and where you're leaving value at leaving money on the table. So if I'm ahead of customer services, customer experience, right now listen to this, I'm sat listening to you thinking, Jesus, the outcomes of what you're seeing sound really important, I'm really excited about. However, I'm not a speech analyst, I'm not a I'm not a I'm not an analyst person, like I can understand it, or I can get it, but I'm really passionate about the outcomes for the business feels like that's a really complex thing to do. That what you're talking about there sounds like it's gonna take me like weeks or months to deliver that. Yeah, and I think Well, I think that's another I think that's another thing with the, with the the State of the State of speech analytics in the state of it, that's a lot of software, just in general, within contact centres. So another bugbear that I've had. And I, in fact, this isn't just to do with speech, actually, because this is the problem with WFM systems it's the problem with telephony platforms and reconfiguration and things like that goes all the way through. It is they've been designed by engineers and analysts, for engineers and analysts. And they're way, way, way, way, way over complicated. So it's almost like you have to have like training and, and experience within the platform to make any use out of it. And it becomes like, it becomes an industry in and of itself, a cottage industry being good at, I won't name any of the names, but some of the speech platforms, and you make a kind of a career out of it. Whereas what my believe is, like, we were not speech, right, you know, TMAC, isn't a speech company, but I don't, I don't believe in analytics, even really, now, in fact I was talking to you the other day, like, I'm not even sure like, I'm an analyst or a data scientist, I'm probably a customer experience person now. And what I care about is having, having, having the identifying the right experiences in the most simple way possible, so that I can action them. And so we design our platforms and our outputs to be number one, really, really simple to get to simple to get to the answers. But number number two, like they've got to be like outcome focused, like what's the so what, and not to be like a bit of like, nerdily in like, you know, belly but bellybutton gazing, it's got to be kind of, it's got to be action orientated. Therefore, like, again, historically, a lot of speech analytics projects has been backward looking, like what happened last week, that's just like, MI. Like, that's not a great use case. Whereas I, what you want to use speech for is forward looking stuff, like, what behaviours are happening within the contact centre today that I want to fix for next week. And then I want to compare last week, last week to this week to see like, you know, the quality of my calls improving, or me becoming better at identifying vulnerable customers, or the amount of alerts going to a second line support team for escalations, decreasing all these types of things I want to I want to reduce and I'm doing it because I've got a very simple configurable system in the middle here that has speech powered with it, but really, it's like a, an alert based action based system that is driving this value and it's driving this continuing, continuous improvement programme. It sounds like there's so much that you can do with Speech analytics, I'd imagine it's quite hard to keep control of that within the business. If you do bring, if you do manage to get the business to put the business case forward, get speech on sits in. I feel like everybody would want a piece of it. Everybody's going to be like pulling up me like, I want this information over here. I want that information over there. But it's it's actually it's my project. And I wanted to originally do this with it. Is that Is that fair? Yeah, well, I'd probably say, the lifecycle of a speech analytics programme goes a little bit like this, the speech, the speech, stuff gets procured. Nothing happens for like 12 months, while your analysts and stuff, figure out what the bloody hell they're doing with it and start building categories and stuff. Maybe 12 months is a long time, but maybe like six or seven months, it just takes like, an inordinate amount of time to get anything going. You've not been able to hit the business case for which was like, you know, 5-6 100, a million pounds, then you're already spending 200 grand on it. So then, then everybody then starts going out to the business going, right? How can we use speech analytics to help your part of the business? And loads of people then go? Well, I've got absolutely no idea. And they go, right, well, let me spend some time with you. And we'll write down a load of things and a load of use cases and then we'll answer them for you. And it then becomes like, well, an emperor's new clothes. It's like, we've spent a load of money on it. Now, every single contact centre project of various projects, or any project has to somehow involve like speech analytics, and then it becomes like an absolute vacuum. I think where I've seen it work really, really nicely as it is, for the to be like a lot of structure around it. Is a guy Rich, Rich mobile at home homeserve. I think they've got a really good rhythm about the stuff that they do. I think I've been really impressed when I talked when I talked to those guys around like just the structure and it just being very, very tight in the briefs. And just going through it. I think that's, I think, number one, if you're going to if it's going to be insight based action points, I think that's what the way you've got to do it. I like I like more around, like, how do you? How do you have better conversations and monitor the agents having those conversations on mass, and push those into coaching? Like that, to me, makes more sense, like so you've got a great script, or you've got a great overview about what good looks like train, train models, and create insights around what what good tempo is, or what events happen, you know, if they're apologising correctly, or are the identified vulnerable ability, correct. They're, you know, making sure that the FCA regulation has been done or is likely to be done. And creating a scorecard about what great looks like for that. And then pushing that into agent coaching actions and correct. That's how, for me, you create the right rhythms. And, and actually, actually, it's probably a bit more top down as well. I think it's more agreeing those those that stuff at this top line level. With the CEO, and the Chief Operating Officer and things like that, these are the metrics that are going to drive your business forward. Not all of them are going to be about speech, but but some of them might even be underpinned by speech. You know, if you want to upsell and cross sell Omar, then the mechanism to upsell and cross sell might be a next best action for an end our CRM, but the underpinning technology might be your speech or text analytics, making sure people have said it, or getting the feedback, why they're not upselling and cross selling from the customer. Maybe the what they're trying to upsell and cross sell isn't, isn't required. Okay, so the second half of the show, we usually pick up on on recent events. And I think it's probably hard to get through the last couple of weeks without having seen what went on at the Oscars. Jimmy, what, you know, there's there's various stances on all of this. And I'm a little bit surprised that some of them, if I'm fair. What's your take about what happened with Will Smith and Chris Rock? I mean, it's accepted you. You will you send me the link. Didn't You ! So. So I woke up, and you sent me that you sent me the link to what had happened. And I will. Honestly, I thought it was a joke to begin with. And that but then maybe like half an hour later. I am. I obviously went on to Twitter, and Reddit, and had and had a look. And I think it was the reaction afterwards. So I think just to explain this, just to explain the situation. So what happens is Chris Rock walks up walks out onstage. And I'm not going to say everything verbatim, but he, he said hello to people. And then he said that he made a joke that was referring to GI Jane. And I'm talking about Jada Pinkett Smith - Will Smith's wife. And there was kind of a bit of a pause. I think it might have snapped to Will Smith, who was laughing at some point, but I don't know whether or not the timings were off. But then then it cut back to the long shot. And Chris, so Chris rocks on stage, Will Smith just strides up slops and slaps him. And everybody laughs because they think he's right, surely. Yeah. But then because they're like, "oh, surely it's scripted". And then, and then it cut what was interesting downwards, so then it caught. And then it was only on another, I think it was on the Australian kind of feed or something like that you saw, but he erm Will Smith goes back to his chair. And effectively says, My keep my wife's name out of your f*****ing mouth. And he says it multiple times, like increasingly increasingly angrier than it cut. So, yeah, so that was so thanks for waking me up in the morning. To go through that. But I found I've, um, you know, having reflected on it, like, found I think there's I think there's some interesting, some interesting things to talk about around this subject. Because I think I think he covers things around free speech. And I think it covers things around like I think these things around kind of masculinity I think these these topics around this topic topics around Will Smith in and obby say itself and Will Will Smith his wife's like relationship and the things around that. I think it's worth, ya know, in around like how the, the academy the academy got in got involved because later on that evening, like Will Smith one is when the best actor award as well afterwards. And so, I think it was there's a lot of different there's a lot of different things to discuss on. On something that actually I had a particular point of view at the beginning. And now I've got like quite a different point of view and probably different to yours. So yeah, I think I think what surprises me most is that If we take it all as is as it as it as it wasn't pre planned and like I'm still sceptical of the initial assault if I'm honest like I still think that looks staged. Yeah, like let's assume that it was a genuine you know thing and let's be honest, we know that that Chris and Will outside of outside of this are allegedly quite good friends. So yeah, out of character it's certainly out of character for Will who to my memory as far as I can remember from being a kid, never put foot wrong in the public light in the slightest. No, nothing I can put my finger on anywhere. But for me like the biggest biggest surprise out of all of this is like all eyes have been on well for the entire time. But there's no point being any attempt by Chris Rock to actually apologise for what has actually offended His friend and his friend's wife in public. And I get your point about free speech and and I'm not saying that violence is an answer, but at the same time like two wrongs don't make a right and at some point he's obviously offended His friend and nobody seems to be bothered about when's Chris Rock gonna make an apology for upsetting and you know, like, and I think it goes beyond offending his friend and it's it's it's kind of that whole thing around is is he's offended erm everybody who has alopecia basically norming like making light of the situation if it was if it was a racist joke, I'm pretty sure that like that wouldn't have gone that the way that it has. But but so actually, I think we might be on the same side here but what we can I think we can I can argue kind of both of both of the sides are look look like Chris Rock has told many racist joke. Like, like many people have laughed at them and bought tickets to see him and stuff like that. So I don't think I don't think it's necessarily like, you know, joking about somebody joking about someone's race or, like we can go through it like, you know, it's a joke. So joke joking about, like race or disabilities and things like that. Like, you know, Dave Chappelle has gotten to like quite a lot of bother around, like Trump, with the trans community around the jokes that he was selling around the drop trans community. But I think, like, there was a lot of interesting things that he actually said, If you listened to what he was saying, and you can get past some of the initial like, you know, should you really be saying he was? Yeah, he was saying like, some quite interesting, some quite interesting things. And Dave Chappelle and Chris Rock, and people like that always kind of say, say interesting things. And there's a certain commentary, but like, I've always, I'm always of the opinion, like, you know, you can, you can make a joke about anything, like you can, but I'm, I'm not particularly I'm pretty, I'm pretty thick skinned. Like, I think you can make jokes about anything, I think you should actually be able to talk about things that other people people don't want you to talk about, because I think that's how you find common ground. And I'm of the opinion like, you know, humans have way, way, way, way, way more in common than they have. Not in common. But we argue over the edge cases now. Like, and I see that across, like, you know, the, you know, the we've now got, like the feminist community and the trans community to certain people for arguing over things. People who've got like, way, way, way more in common than that they have not in common, but they're arguing about edge edge cases and stuff. And I don't agree with it. And I think that you can make a joke about anything, right? However, right? What that proved was, you can make a joke about anything, it doesn't stop you from getting smashed in the face. Because of it. Like, and I'm not a violent person, right? I've never, I don't think I've ever had to fight in my life. I don't think I've like I don't think I've raised my hands, like on to anybody, like a cry almost all the time. But like, it's not, it's not built, like I'm just not, not violent. And I've got a pretty big mouth. And I've said, like I've said loads of stuff, and I express my opinion and stuff. So I'm saying this, at the point where I'd probably never hit anybody, I'd probably likely to be hit about something that I'd said if I was joking, or if somebody took me out of place, right? However, like, when you joke about a community, you're joking about, you know, an amorphous blob of things. But when you're, when you're going in on an individual person, right? You've got to be very, very, very, like, careful about that their own situation. And, you know, if you like, and, and even then moving on from that if you're going in on a particular person, and they're five metres away. And maybe a few drinks, right? You've got to be really, really, really careful. Now, I've, I have friends who will stand up comics, actually, oh, let's, let's not say friends, but let's say acquaintances. So I went to I went to Salford University. I was just last weekend, I was out with some of my university friends to two of those guys used to do stand up and I went to see those. There's a guy Dominic Woodward who I went to university with really great compare. He's a great comic. Phil Buckley. He's He's a great comic. And the most famous person is Jason Manford, who was really, really popular now. He's been on he's been on BBC on ITV. It's kind of everywhere. He's omnipresent. He's really popular. And Jason Manford. Like has made jokes about me when I've been in the crowd. Right? And I can't remember the full the full jokes. I won't tell the joke, but it was relatively offensive, but not particularly not particularly offensive. And everybody laughed at me and I laughed because it was pretty funny type of stuff. Okay, but what you've got to think is like, so. You've got to I've seen like, kind of roasts and stuff where people go on and know that they're gonna have like, yo s*** said about them. But if you're is it jada or jada? I've always said Jada, Jada. If you're young you she gets she gets a lot of sh**. Like she's got she's got s*** over like a world events and stuff. Will Smith is got like a lot of ribbing at a lot of these events. Like, okay, you're there. You're there. You know that the Chris Rock is going to be there but you're expecting people to just go in on you. And are you expecting like as a woman who's got alopecia alopecia to have a joke set about? Yeah. Probably not and probably not on that night as well. No. Probably not on that night and not on the night when your husband's like gonna get an Academy Award. Like it just and there's just some things about it where it's like, again I think you can joke about anything but that was that was probably the closest to the line that you could ever get like you've seen people get you see people at a comedy show get ripped to s*** like are you know they Heckle and people go in in them but it's like that's part of it. But is it is it part of it like when you they're in the front row like with no hair, probably feeling you know not feeling particularly great about yourself. And Chris Rock one of the most famous comics on one of the most like watch nights in the world is seen something like that's, that's something that I found really, really weird. Like, on the back of like, you haven't really had an Oscars property for two years because of COVID and lock downs and things like that. There's been like kind of subdued Oscars and virtual things and like it's the first real one back everybody's out and really enjoy it really looking forward to night out. But you've got to ask yourself the question like, like, the Oscars know what Chris rocks like, and they still, you know, knowing his background, knowing all the stuff that he's you know that that he is quite provocative and risque. And like they're still asked him to present the Oscars so like, it's part of that blame with them? I don't I don't think I don't think the blends with them. I think I think Chris Rock has to. So if Chris Rock like, if Chris Rock knows, like, Jada Pinkett Smith and Will Smith, as well as people say, like he'd know that. What's a cool joke or what's not a cool joke to say? Like even like even for me who thinks you can say anything? Like I'm like, one of my favourite comics was Frankie Boyle. Right? He was blocked me on his walk through on Twitter. I'll tell you why. Because it's like he makes like racist jokes. Because he's like, he's, he's been he's been satirical. And he said racist words in satire. And that's fine. But then he's then people have a go at Boris Johnson about some of the things he said historically. And I've said, well, Frankie Boyle gets to say it. Why does it Boris? Boris Johnson when he was a journalist, like why can't he say and that and Frankie Boyle I didn't even tell you it it Frankie Boyle must have somebody who searches stuff and he blocked me. Anyway. I think you can say things but like, like the I don't think it's on the that bit is on the the academy, like Chris rocks got to have like a sense like, you've got, you've got a 100 200 whatever targets haven't You got a 300 400 You know, you've got to go all well, which which of the ones like look at Ricky Gervais. If you look at Ricky Gervais at the Golden Globes, the last time that he did it, like he went in on like loads of different people. And like, you know, he went in, you know, it's how you do the joke. It's the joke that you tell even it's how you construct it. Like, and how you practice it like you must you do it in a way which is it's that's that that is quite offensive, but I think it's okay. And it will just it wasn't even it wasn't even the best joke. And it was it was pretty offensive. Like Ricky Gervais went in on loads of different people he went in on it and actually, he made a joke about Caitlyn Jenner. And at the time when she just transitioned, but even then, Caitlyn Jenner adult Caitlyn Jenner wasn't there. So she wasn't like so it's a bit like, you know, second secondhand, but at the same time, like for me listening to the joke that he told it was like, yeah, it's relatively close to the bone. But it's like, it's it's relatively gentle. Like it wasn't, it wasn't too harsh. Whereas, like, somebody suffering from alopecia, making a bald joke just seemed like it was pretty just low and that and therefore like, I was like, You got every right to say whatever you want, was a s*** joke, you got slapped? Like, I'm like, I'm not a violent person. But that's that's how I felt. And then what what I thought was interesting, actually, was like, the aftermath. And the people who then commented on the aftermath. Did you see what Jim Carrey said? No, I missed it. So So Jim Carrey came out, actually. And he he kind of supported Chris Rock and said comedians and things like that. And then footage of him getting upon the stage and then accepting and accepting an award and basically sticking his tongue down Britney Spears' neck emerged. And then he tried to do the same with Will Smith and Will Smith was like, What are you doing? Like get away from me? But you're like, Is that appropriate behaviour? Like Absolutely not. That's that's not appropriate. So come on, there's got to be like these. There's a bit of hypocrisy and stuff there as well. And you've got I don't think. I think everyone's kind of taking sides a little bit. Like I can see from Chris Ross. I can see from Chris Rock side. I think Chris Rock is probably wise to just keep his mouth shut for a little bit just to see what happens and let Will Smith like talk? I think I can see it from Chris Rock side. He was just taken tell it a joke. Can see from Will Smith side. It just wasn't a particularly funny joke. And he was he was within the vicinity to do something about it. And I can I can totally say I can totally see it from Jada Pinkett Smith's eyes... where where it's like what have I done to deserve deserve that I'm on the Academy Awards night. And I've been so much trouble if I reacted the way Will Smith did not not off everybody else but I'd be more bothered about my wife. I probably not go back to my seat for having done it. I've the only time. The only time when I've had to to probably get slightly physical is when like when Lauren got into an altercation because somebody was be being an idiot. And I've had to like step in. I'm probably more likely to get some friendly phone from something like that. And it doesn't seem like anybody's overly bothered about Chris Rock apologising, or like, like I've not seen any calls in like, not any major calls and like talk about all the news articles and stuff that I saw. It's like oh, when's Will Smith gonna apologise? How's he going to apologise when he stepped down from the Oscars? Why is nobody bothered about Chris Rock? Like if like, again, like this all this is all based on the proviso that this isn't some sort of staged event and like I've seen, I've seen stuff to see otherwise as well. Like, you know, if you look at the still animations of just before the slap happens, you can does look staged if I'm fair, and I'm sick to death of the body means it's just like, it's just slap, slap, slap name, we'll get Chris Rock being like, I just can't wait for someone to shush it up or something. Something different to happen. Now it's just a it's just insane. I think the I think Will Smith was particularly impressive, though. Like with his acceptance speech and what he said at the time. Like, I think Will Smith's a gregarious guy who I've seen him do interviews and is able to articulate himself like perfectly well. I don't know if he I don't know if the emotions got away from him or people are really sad. But I think a simple he said something that was completely out of order. And I reacted to it that would have sufficed I think yeah, saying that he was what channelling something shadow channelling love of a child a vessel of love awesome shit like that. Like, I don't, I don't know about that. But you've just hit somebody. And I thought, and I don't I don't quite understand how the Academy of now have now managed it. But one thing I will say though, like the Academy Awards, as an event has been declining in terms of popularity. And so well, is it any, any any? Any advertising is good advertising? maybe, potentially. And I guess with that, we'll we'll draw a line under today's episode, Jimmy. It's been a it's been another interesting episode. Like I've probably enjoyed this one the most out of the ones we've done so far. Been good to chat. Yeah, and again, please make sure you subscribe if you've enjoyed the conversation, whether you enjoyed the first half second half of both. We'd love to hear your thoughts. You want to reach out our LinkedIn profiles or on the podcast bio, feel free to connect with us. We'd love to hear from anybody who's watched the show and hear what your thoughts are