Exploring AI, Power Virtual Agents, and the Power Platform: A Journey with Sarah Critchley, Microsoft's Principal Program Manager

Exploring AI, Power Virtual Agents, and the Power Platform: A Journey with Sarah Critchley, Microsoft's Principal Program Manager

Exploring AI
Sarah Critchley

FULL SHOW NOTES
https://podcast.nz365guy.com/505 

  • How would you feel if you could make your interactions with technology more human-like, more interactive, and more efficient? Hop on as we journey with Sarah Critchley, an accomplished principal program manager at Microsoft, into the fascinating world of Power Virtual Agents (PVA). This episode is a treasure trove as Sarah Critchley narrates her inspiring life journey, from her relocation to Canada to her rewarding career at Microsoft, and her labour of love – writing tech books in an era of rapid technological change.  
  • What if there is a playbook that could guide you from start to finish for a successful PVA implementation? As we dig deeper, we uncover the secrets of PVA with the PowerCat PVA playbook, discovering the PVA in a Day feature and Microsoft's Conversational UX Guide. Sarah enlightens us on the game-changing Generative Actions feature which makes the user experience more natural by inferring context from user queries. We also delve into the very foundation of PVA – the large language models and generative AI. 
  • Now, imagine integrating Azure Open AI service with PVA – the possibilities seem endless! We dissect this integration process and shed light on the deployment of adaptive cards in PVA. We also have Mark Smith, the nz365guy, who shares his valuable insights on creating with the Power Platform. He sparks a conversation on the potential role of voice technology in PVA in the near future. So tune in and join us in this exploration of AI, PVA, and the Power Platform – you'll be amazed by the things you'll learn! 

AgileXRM 
AgileXRm - The integrated BPM for Microsoft Power Platform

Register your interest for the 2024 90-Day Mentoring Challenge. ako.nz365guy.com

Support the show

If you want to get in touch with me, you can message me here on Linkedin.

Thanks for listening 🚀 - Mark Smith

Transcript

Mark Smith: Welcome to the Power 365 show. We're an interview staff at Microsoft across the Power Platform and Dynamics 365 Technology Stack. I hope you'll find this podcast educational and inspire you to do more with this great technology. Now let's get on with the show. In this episode, we'll be focusing on the amazing tool that is Power Virtual Agents. We might touch on things like co-pilot, ai and all those lovely things. My guest is from Canada. She works at Microsoft as a principal program manager. She's a writer, speaker, author. Multiple books to her name. Check them out they'll be in the show notes. She truly believes that customers and employee experience should be at the heart of business. You can find links to her bio, social media, show notes, etc. In the show notes for this episode. Welcome to the show, Sarah.

Sarah Critchley: Hi Mark, how are you?

Mark Smith: Good Good to have you back. I think last time we had you on the podcast was before you worked at Microsoft. You were being an MVP. Just as a bit of background, sarah and I have spent some time together our families in the UK when she used to live there, but so the audience can get to know you, sarah, tell me about life in Canada, how you ended up in Canada, and are you safe from only the bushfire reports that we're hearing going on in Canada at the moment?

Sarah Critchley: I am safe. It's a pretty terrible thing what's happening in Canada. It's actually happening all over Canada. There's a huge amount of wildfires, unprecedented in the scale. It's definitely good to raise awareness about what's happening in Canada Because, even though I'm in a good place in terms of we don't have any drastic bad things happening in Vancouver, you can see the smoke. You can actually see the smoke traveling so far, which for other parts of the world, that's, people don't really understand that and how far these kind of things can travel. But Canada is a beautiful place. It's so amazing to be able to just go out on a hike on a weekend rather than spend an immense amount of money to do so for a holiday. It's a true privilege, so I love it here.

Mark Smith: How did you end up moving from the UK to Canada?

Sarah Critchley: I was lucky enough to get a job offer to move actually, so I moved for work and truly grateful for that opportunity. It's amazing actually and, mark, I mean you've done it yourself right in terms of lived in another country and definitely there's things you take for granted. There's definitely lots of life lessons learned when you move and so, yeah, it's been truly life-changing and amazing.

Mark Smith: Nice. Tell me about your journey into Microsoft, because you're working for Microsoft now. How did that come about? Is it amazing? Is it the you know? You hear? The term that you've landed a job at the mothership is a common term out in the community. You've become a blue badge. You lost your MVP not because of an NDA violation, but you have joined Microsoft and Microsoft employees cannot be MVPs. What was that journey for you?

Sarah Critchley: Yeah, it's kind of funny, isn't it, ray? It's almost like people see it as like kind of like a career path, like especially as an MVP right, it's almost like okay. So your next step is going to be Microsoft. My experience has been truly amazing, Like it has been again super fortunate to work where I work and in the team that I work in. You did interview my colleague Dwayne recently as well, so it's an amazing kind of experience, because one of the things I truly value about working here is the collaboration. It's definitely a massive difference from where most of my career actually was, which is professional services, right. So coming from that kind of world into primary tech firm is definitely a change, One that I'm actually grateful for, right. So it's definitely really really good and it's quite nice where I've been able to leverage my background in dynamics and actually kind of bring that over to the conversational side, to the conversational AI side and everything we do in dynamics, right, and in business applications and all of that, and that's been truly useful for the role.

Mark Smith: Awesome we touched on it at the start there on your books that you've created, and always find it intriguing the life of an author and just how wealthy they become from writing books.

Sarah Critchley: You don't do it for wealth.

Mark Smith: You've created a tech book. Would you do it again? What were your thoughts?

Sarah Critchley: Oh, that's a hard question, you know, because, like this day is, I wake up and I'm like I've got so much to say. I love writing. It's how I learn myself and I don't know, to be honest, it depends what kind of day you. It would depend on what. I don't think I've write a purely technology book anymore. Like kind of the book it was really a guide on dynamics. It was around the time dynamics suddenly became very app focused Right, kind of back in the day it used to be fairly flat, right. You know, you would have your modules, you could learn the entire like kind of dynamic stack right and be an expert on dynamics 365, even though it was called something different, right Whereas it was around the time when it was really more about the depth of knowledge in each of the apps. So the book was meant to be something that you could hold in your hand and be able to at least be able to get started with it. But I don't think I would do that anymore. From a tech perspective, if I was going to write another book, it would probably be something a bit more kind of human centred or something a bit more long lasting.

Mark Smith: Yeah, so how to survive an office romance or lifestyles of the rich and famous at home?

Sarah Critchley: Lifestyles of the rich and famous. That was a good Charlotte song, wasn't it?

Mark Smith: Yeah, I just think it's a labor of love really writing when it's you know you've got a publisher on your case. And that tech sector the lifespan of a tech book right, is that the technology is changing and back when you know it used to be every three years, would get a product update or release back in the early days of dynamics. Fine, at least you got three years of runway. Yeah, Now you get what? Two wave releases a year. It changes a lot. It's hard to maintain that a book in a physical form when it changes so much.

Sarah Critchley: That's the problem, right? It's funny because I remember when, back in the dynamics days, I remember when and you're spot on right you would have such a massive amount of runway, and I remember just like you would be waiting for the day that the feature that was promised would be dropping because all your customers would want it. I remember that distinct to area was business process flows when they came into the dynamic. Yeah, everyone was like, oh wow, these are finally going to be in the product, the new like kind of UI, right, we're going to do so much stuff and you know it was that excitement that can be seen. You know, it's almost a bit different now in relation to that. It's a continuous update cycle. It's definitely something that people need to kind of like switch their thinking around when they're thinking about like upskilling, a learning.

Mark Smith: Yeah, interesting those business process flows, I just think, you know, in their day they were absolutely phenomenally amazing and even just to hit a project recently that used them and I still think they're amazing. I think they could be extended on massively from a BPM, you know, business process management perspective and more, you know, visually enriching, but, yeah, cool feature. A lot of people found it hard to get their head moving away, remember, from the old dialogues into this kind of metaphor as a way of grabbing multiple step. You know engagements. Anyhow we digress, let's talk about the amazing technology of power virtual agents and why every person in the power platform ecosystem needs to gain experience with PVA and think about interface design in a different way. Like there's been a massive index on power apps, for example, let's build an app for that and that we're wanting a screen based experience. And with PVA, it opens up different dimensions of engagement which, yes, you know, you've still kind of got a micro app experience in that you can put adaptive cards and stuff, let's say, in a Teams context, but it's a different way of communicating with data, extracting data in a really timely fashion. When you explain PVA to people, how do you explain it? What do you suggest they do with the tech?

Sarah Critchley: Oh, wow, how long we got Like kind of I would. I can explain it in a few different ways, but I think probably the best way to think about it. And actually, mark, you talked about something very important and that is that the difference between an app right so you know, we're so used and this isn't just a power wrap thing, right, this is something that, when you would make WinForms apps and even a website right, is technically an app. Right, you've got an interface. You use essentially a keyboard, a mouse, a screen reader or whatever to essentially use it. Right, and the very important difference, when we're talking about conversational AI and as much as that word can be super intimidating for people who are not in the industry it shouldn't be right, like, definitely recommend to kind of get an educated on it. You're watching this, you listen to this, so that's step one. Right, it's you're using your voice to interact with it, right, so this is about not using your hands. Right, this is about being able to almost like, just use language to get what you're looking for. And if you think about that from an enterprise perspective, there's something just truly amazing and accessible about that. That is so, so, very important and that should really always stay at the heart of everything we talk about. When we talk about conversational AI, because it's about enablement that. You said it kind of like micro app experience, for sure, because most people will probably be used to like chatbots on a website, right, but the idea, and especially as you start to, you know everyone would have heard Satcha talking about co-pilots, right, and you know this idea of this co-pilot experience and how AI is embedded in so many different Microsoft applications today and it's the same thing, right. Well, what we're trying to do is we're trying to make the fact that you don't have to spend a year trying to learn an app right, like, kind of. I mean, everyone surely can empathize with this as their own consumer, right, and you know, as a consumer, you're going to have, like, an internet service provider. You're going to have, you know, other service providers where they're probably going to have an app. Right, you know your bank will have an app and you need to learn that interface to figure out oh, how can I ask for help on this app? Where do I check my balance on this app? And the whole idea is a fact that you can just say what's my balance and you can just say, hey, I need to raise a support ticket, right. And the truth is like Mark, you and I will say those things slightly differently, maybe completely differently, but the whole idea is that the from a conversational AI perspective, that should not matter, right? Like you know, you're not hard coding slightly different differences in the fact that, oh, sarah said it this way and Mark said it this way. The fact is that the technology, the AI behind it, knows and can understand and derive the intent from what's being asked and then can then go ahead and perform the action right, and that's a truly fundamental difference as opposed to what should I put on this UI to make this kind of understandable for a user.

Mark Smith: How do people get started using PVA?

Sarah Critchley: Oh, wow, that's like I mean. There's so many different ways, so one of the biggest, the easiest ways is our PowerCat team did a playbook. So if I think about this from the consulting element and also from a customer perspective, to me honest this is so useful for everyone, not just implementing organizations that want to use it internally, and if you're like a Microsoft partner or something like that, the PowerCat PVA playbook will be the best way and we'll put those in the show notes for your audience and that actually takes you from start to finish of a PVA implementation. So, and there's actually a massive part of that playbook that covers the actual use case, like the actual business analysis. I can't again stress that enough. This is the make or break of any software development like project, and it's when people do not spend enough time understanding what business need is being worked upon right, and so the playbook actually talks about that quite significantly and we actually embedded that playbook into our PVA in a day our latest update for PVA in a day. So that's also another way that if people are a bit more hands on, if people like Sarah, I don't really want to read like a 10 to 20 page PDF document please go ahead and take a look at PVA in a day. We actually kind of really squished that down, we actually summarized that and made that really succinct. It's not dependent on each of the modules anymore either, so you can like kind of just pick up a module and kind of go. If you really are limited in time, go ahead and start with module three, right, go ahead and start with module four, whichever one will get you the biggest ROI, and then the final one will probably be. There is actually a conversational UX guide to the Microsoft have made, which also has, like again, we actually talk about this in a few of our materials, but again it goes back to the human element, right, because ultimately, you're still designing an experience for a human being, right? And so as much as we're trying to take away the kind of almost the technology interface and make this very kind of you know, familiar with language, there's still lots of differences that people can bring, and so the actual CUX guide is really good as well. Yeah.

Mark Smith: Soundbite that I just took away from that is the conversational UX which is I had to even picture that out of my mind. You know, and I think it's such a great way of looking at this, that it's not just a, you know, it's not just a very programmatic. You know, ask a question, get an answer. It's the nuances around answers and it's a nuances probably more around the question, especially when people ask multi-part questions. Yes, and understanding that, and one that comes to mind if I would want to validate an AI is I'd like to book flights to Vancouver on the 20th of the month and I want to find first class. You know, there I actually asked a whole range of things. Yeah, I asked scheduling things, I asked destination things, I asked service level of product things and that was just one question, and so many of the historic bot technology out of there needed question answer, question answer, question answer. Rather than understanding there were multi-part question and answer is a multi-part answer. How is that changed now in the context of large language models where it can understand that nuance of conversation? It has that context to know that that was what you were asking. Are you seeing a bump in experience from that?

Sarah Critchley: Yeah, for sure. It's really interesting because the example that you just provided is very similar to what we actually demonstrated at Build. We actually launched one of our latest features in PVA. It's in limited preview right now. That's progressing. It's called Generative Actions, so that is essentially really what you just described, which is like chaining technology. This is around how we're using this technology to really create again an even more natural experience for the user, because the truth is, you're spot on right. You're not going to and you shouldn't have to pass your own questions to a bot, because then again you're having to be trained to talk and it's just the same as using an app. We have to manage these natural differences in ways we talk. I'm the same. I would tend to ask 20 questions in one big paragraph on Teams People. My colleagues probably hate me for that, but it is what it is. That chaining technology is really important because they almost infer context from one thing to the next. So the other thing that's really important is the things that you didn't say in that statement, so you didn't actually say where you were flying from. There's actually missing information for the actual technology to complete the query, so there's lots of things that really need to be factored in. One of the amazing things and again we'll put it into show notes to our build video because it's actually amazing it's near the end and it was booking a cruise. This is where it was missing the information the departure port and it actually used generative AI to actually generate the question of the missing API information from a required component in the API that was needed to make the cruise booking, and it's that again, it takes the effort of the user. That is the biggest thing here. That's so important is you as the person making that query. You shouldn't have to do the heavy lifting. You shouldn't have to bridge the gap between technologies. You shouldn't have to make do with what you have. You should actually be served and you should be able to just be asked that question, but then, at the same time, as an enterprise, you can't hard code that information to every single request.

Mark Smith: Yeah, tell me about the building blocks of PVA, because you mentioned their actions, was it I think you said?

Sarah Critchley: actions.

Mark Smith: Yep, what are the other building blocks when you're thinking about using PVA? And I wanted to go out and I want to do gain a level of understanding around anything from large language models to generative AI, to what are those building blocks in your mind? What do you need to go out, understand what it does and then you'll have, kind of like, the tool set, the understanding to start creating.

Sarah Critchley: Well, you come in at this from like a perspective of like a consultant or an architect, mark, yeah, yeah, sue.

Mark Smith: They're going to build a PVA solution for their business.

Sarah Critchley: Yeah, okay. So the first thing I'll definitely recommend is go read those resources that we talked about a moment ago, right?

Mark Smith: Yeah.

Sarah Critchley: If you was to begin, one of the things to be mindful of in today's technology landscape from a business application perspective is there is a lot of interoperability between the technology Right, let's just be upfront about that Like power vector agents, has a dependency for our generative AI features on Azure Open AI service, right. And so one of the things to be very mindful of is there is this kind of element of people thinking they have to stay in their lane, and that's not really how I would recommend people thinking about it. Right, you know you don't need to become an Azure expert, right, but you do need to know what it is. Right, you do need to understand, because, ultimately, once you understand even just the some of the fundamentals, you can do a better job explaining about how that service is used in power virtual agents and also why and we kind of cover this in one of the blogs that we just launched, which kind of has a little bit of a matrix as to well, you know, these are some of our major services from a generative AI perspective. You know what do you get with one of them and you know what don't you get with them. Right, like, how can you compare and contrast? Because that is one of the challenges, especially when you come at this from a concern and an architect perspective is you're so worried about what? If I, like you know, lock myself out of doing something and stuff like that You're kind of trying to future prove something already You've not even built the basic thing. So that's one of the things I would say to understand is to try to just understand the kind of actual application landscape. First. I am you know that's power virtual agents has power automate right that some of the Azure open service right to is literally like one docs article I'd probably recommend, which we can also put in the show notes. And then it's really about understanding what power virtual agents does out of the box, right, like what are the fundamentals? You know what's the author and canvas, what can I do and what are our generative AI features? The power virtual agents team have worked so hard to deliver. You know generative actions is limited preview, generative answers and our copilot feature, right. So those are our kind of like flagship kind of components when it comes to generative technology, genitive AI technology. But ultimately it's like how can you actually create your own topic as well? Because what we gotta be seeing kind of what we're seeing now is where, sure, you can use generative AI technology, like generative answers point and a website, but some customers are still gonna want to, you know, go ahead and actually author their own topics that they have control over and they can manage the messaging of. So understanding the fundamental authoring canvas will also be key.

Mark Smith: Yeah, we've touched on Azure and the open AI service. Tell me about how you'd get that lit up and working inside PVA.

Sarah Critchley: Yeah, so there's almost two ways that we can look at that. There's actually like a complete hands free way where you'll never gonna have to go to the Azure portal so you can leverage Azure open AI service, the service, directly from PVA. So again, just crystal clear, right that you don't need to go to the portal and Azure. You don't need to go ahead and provision Azure open service. The generative answers feature, for example, utilizes that service and you can even see this when you start using. If you start test, if you hook this up and if you turn this on, you're gonna be able to start, you know, pointing out a website and ask a question. Once you get an answer, you're gonna see like a little bit text says you know, provided by Azure open service, and so that is one of the key things is the fact that I've heard from a few people it's like all to connect it. You know you gotta have to go, do a manual connection. No, you don't have to. You can certainly go ahead and just literally use our own service. Now the other thing you can do, which is what we just launched, is that you, if you're developing in the Azure open studio From your service, from your Azure open service. There is actually a deploy to PVA button now, so it's specifically on the Azure open AI, on your data service, and you can literally, just literally deploy, and that's another way you can get started.

Mark Smith: So I don't have to worry about things like GPT 3.5 turbo or 16 K turbo. No, because you know, when you jump into a zoo you've got those options. You can go. Okay, which model do I want? It just handles that for you, right.

Sarah Critchley: Yes, so if you're going from the PVA side, yeah, absolutely right, and it's that's the kind of irony, is that, like when people have tried to do this on the Azure open AI side, they're just like I have no idea what I'm doing and it's like you probably shouldn't be in there, right, like kind of that's. You know you, this is why we've developed these things, and the fact that you know we want to make it easy. We want to make it. We want to still provide business value, we still want to be able to make this so simple, so easy and cost effective that you know you can absolutely just provisioned power virtual agents and you're going to be able to get the generative features.

Mark Smith: Wow, pretty compelling. Tell me about the relationship between PVA and adaptive cards.

Sarah Critchley: Oh, yeah, for sure, I mean they like each other. It's funny, adaptive cards and one of our was actually one of our most anticipated features when we just dropped the latest release that build this year. So Our previous version did not support adaptive cards. You would have to go ahead and, like you know, install composer and stuff like that so you can now literally, you know, make some very, very rich adaptive cards as well as it's also the behavior around them. So I just want to be super clear. You know, especially especially for maybe those that haven't played around with adaptive cards in power, virtual agents, when you add the conversational layer on top, you know, it's not just you have to be really kind of like, kind of mindful of the channel, of what that's being kind of deployed in right, because the actual kind of how adaptive cards are constrained by channel is quite critical because of the interactivity involved and when you've got a conversational interface. It's important.

Mark Smith: So you're talking about channel being? Is this going out in an email?

Sarah Critchley: exactly inside teams chat. Yeah, I love it. I love it.

Mark Smith: Okay, good, so you're saying you need to know that context of where that's going to be surfaced.

Sarah Critchley: Yeah, yeah, for sure, because that you're going to have like different kind of things available to you and right. And if you've got like a web chat with adaptive card, are you making an interactive adaptive card or is it an informational type adaptive card? Right, because, again, you've kind of gone switch to an app type experience where you might be expecting someone to like click something. And that's also really important when you're thinking about voice technology as well, right. So, like, if you've got kind of if you've got your bot enabled for voice, things like adaptive cards are not going to be relevant in that area. Yes, yeah, interesting, interesting. You touched on voice. I wasn't going to go there, but let's do it.

Mark Smith: What's your familiarity personally with things like Alexa, google Home, those type of products just checking my one didn't light up and do you see, and maybe under the co pilot moniker that we're seeing, it's going to come into windows and AI on windows, you know, is in preview, I think at the moment Do you see that the potential interface out will even move maybe further away from the keyboard? No-transcript. As we go forward, as in so on my Windows device, I assume as soon as it goes public, my Windows 11 computer will update and it will have co-pilot for Windows on it. Do you think in the near future we will get to that point where we will become much more tactile in voice engagement rather than using the bot from it? I'm still typing a question into a keyboard and so therefore all my bad spelling, my bad grammar, my dyslexia still comes out in that, but it doesn't when I talk.

Sarah Critchley: There's a really interesting question, actually. Even if you just look today, people will use Siri 24-7. Some people won't use Siri at all. They'll just turn it completely off, and it's a similar experience. You get some people that love Alexa being in their home. Other people don't like it. I'm a big fan and I talk about this fairly a lot on various platforms about I do truly believe in the ubiquitous nature of conversational AI technology, because I think that I mean, even if you just watch films, you don't see someone really using a chatbot. You see someone using their voice. You see the actor or the main person using their voice and asking the computer what to do. I think how much the world has now become aware of conversational AI technology from chat, gpt. I think that that's just going to continue, the people being more comfortable with talking to things, because the other thing to be mindful of that is it's not particularly great. On customer calls. You have to take into consideration noisy environments. There's new things to consider when you change the interaction model of the technology.

Mark Smith: Yeah, babies, that definitely affects things.

Sarah Critchley: Yeah, for sure. If you've got a screaming child or something, yeah.

Mark Smith: I'll be halfway through a prompt and then my daughter will talk to me and of course it's now picked up on. That part of the conversation is trying to respond. I just want to turn the lights off and turn the lights on or something like that, not set a new timer or play.

Sarah Critchley: Because then you're backtracking right. You're just like no, no, no, no, don't do that, don't do whatever you're going to do.

Mark Smith: Honestly, as in my almost three year old, she is pretty close to saying all the commands. Really, I'm pretty sure that once she every now and again, she'll definitely get Google to respond, because we're mainly a Google Home house, Although I do play in all ecosystems. I've got a couple of extra devices as well. But yeah, at realizing, oh my gosh, I can actually control the projector turning on, I can get poor patrol by saying something rather than knowing how to control buttons and stuff. It's only a matter of time. Even her bedroom lights won't be long until she works out. Do you know what? She doesn't need to reach the light switch. She can actually just say.

Sarah Critchley: It's natural for her, though, right as well, which is like the truly magical thing. She's growing up with this technology. Can you imagine when she's older, like putting like an old interface or an app in front of her, Like she's probably going to be? Like what are you like, dad? Like what are you doing?

Mark Smith: Yeah, yeah, no, I'm going to. I'm going to get a little morse code key and teach you how to start Like Blip, blip, blip, blip, blip, blip, blip. Let's just change directions. We've talked about PVA for a bit and we're almost on time. Tell me about side projects that you are into. I see you often post under your favorite feline around careers and things like that. Tell me what tickling your interest outside of PVA.

Sarah Critchley: Oh yeah, for sure. I mean, I like making cat quilts. That's something that I do now. I kind of yeah, I've gone down that route of making.

Mark Smith: So what you kill cats, you get the skin.

Sarah Critchley: Don't say that Don't say stuff like that. You said a cat quilt.

Mark Smith: I was like I'm so mad You're using cat fur to make. No, oh no.

Sarah Critchley: I'm giving Mark so many evils right now? No, no, I'm giving evils. They are super happy. I've got good customer feedback from my cat quilts, so that's something that sometimes it's nice to just get away from the laptop, you know, and actually do stuff. Like you know this with your garden, right. But from a career perspective, like I am super passionate about, like digital literacy, right, like it's and also career planning right, like it's something that we just do not talk about and I don't know why. It's one of those things that I'm just like I have no idea why people just are so afraid to talk about their career with, whether they're manager, their peers. I've got lots of theories of why, right, but I'm hoping, by talking about it and like I've got the sub stack on it, I've just started, like Motivation Mondays, to try to give people a little bit of, like healthy dose of optimism on a Monday, monday morning, right, to hopefully normalize the fact that people should be having career conversations and being open and given them the tools to do so.

Mark Smith: I like it. For all of those in the Southern Hemisphere. She means Tuesday morning. She just yeah, ok. And so I love the stuff you're doing around resume, cvs and stuff. I just was on a podcast recently, ecosystem podcast, which will sometime be published on the show, and I am shocked with the number of CVs that come across my desk as I am One of the people that decide on people's employment inside my organization at the lack of substance, the lack of personal development, the lack of direction. I'm shocked that most people CVs represent they were asked on their last project to do something, so they learned about that something and did it. It was not so. In other words, anything that's on their CV is because somebody else, you were doing somebody to meet somebody else's agenda, not your own agenda, not your life goals, not where you're wanting to go, not this. You know, and it concerns me, and I use an example. You know I don't want my, if I had brain surgery, I don't want my surgeon to have looked up on YouTube 20 minutes ago how to do it. All right, yet in our industry, where we're touching phenomenal amounts of data and implications of that, I feel there are so many consultants that are literally three YouTube's ahead of the customer videos. Right, and that's concerning because it's not a holistic view. Right, it's a very much one person's lens in that YouTube and I think YouTube's a great micro learning type of platform and experience. But If your career is in the space, act like it Like learn it all. I love that from Satya Be learn it all. People Like be the best in your category, not just the best at the last item on the project you did.

Sarah Critchley: And what a time to do it, right, mark? Like kind of, we have such an abundance. Going back to what we were just chatting about, right, the interoperability between this platform is amazing. Do you have an interest in AI? Go, take a look at AI, right, the amount of materials is just phenomenal, sometimes overwhelming, for sure, but overwhelm is better than nothing, and you can get the tools to navigate overwhelm right. You've got a place to begin, rather than the you know it being kind of like, you know, behind locking key. So it makes me sad, I'll be honest with you. It makes me sad when I see a poor CV, because I almost feel like one of two things have happened Because that person could actually have passion right, they could actually have passion, they could actually have experience and they've done a poorer job of showing you the person who can potentially give them employment of that passion, right. Or you know, this is the second thing, which is what you said, which is they've let someone else really give them the route to where they're at today, like they've been directed to. Oh, you're on this project, go learn it. You're on this project, go learn it. And you find this kind of haphazard piece and it is truly heartbreaking as an interviewer when you ask, oh, why did you do that? And I'll be like, oh, because it was on my projects. And it's like, so you're not actually passionate about it. I mean the challenge there is, the resource inside of the house can be, you know, especially in consulting and professional services. It does direct, where employment is a lot of the time and projects for sure. But I think people you almost see that as the fact that, oh, I don't have autonomy, I don't have conscious choice to go. Well, okay, well, I'm on this project, but I actually have a passion in data, so I'm going to go learn this side of things and that actually opens up so many doors. I feel like, for those listening, if you have a passion like that, you might not think it'll open doors, but it actually does, because you'll just get chatting to a random person one day about that topic or an interviewer and they'll be like, wow, this person has taken the initiative to go do something on their own and that's worth a lot.

Mark Smith: Yeah, totally agree. Final question I like to ask my guests more and more these days how are you personally using AI? So forget about Microsoft, the tech, all that type of stuff. How do you use it in your life?

Sarah Critchley: Yeah, so I use it a lot on LinkedIn. So, because I don't want to be putting my own hashtags into my LinkedIn post, I've got bad things to do. That's something that I think is super, super good. The other thing that I've noticed is that AI has been embedded in existing apps today, so you'll be using note taken apps or whatever Microsoft aside just any other application there's AI being used. You'll see a little AI logo or whatever it is, and you'll be like, oh wow, you've embedded that. Like I've been learning French, so G-Wo Lingo is a really good app to do that and they embedded AI in the app. I mean, it's a different SKU. You have to pay my money for stuff like that sometimes, but I think that that's kind of cool. So I almost feel like it's almost not a conscious decision mark anymore either. Is it like some people may not even realize they're using AI in some of these applications because existing applications have been updated to use it?

Mark Smith: Yeah, agreed, Sarah. Thank you for coming on the show.

Sarah Critchley: Thanks so much, it was amazing.

Mark Smith: Hey, thanks for listening. I'm your host business application MVP, mark Smith, otherwise known as the NZ365 guy. If there's a guest you'd like to see on the show from Microsoft, please message me on LinkedIn. If you want to be a supporter of the show, please check out buymeacoffeecom. Point slash. Nz365 guy, how will you create on the Power Platform today? Ciao.

Sarah CritchleyProfile Photo

Sarah Critchley

Sarah Critchley is Principal Program Manager at Microsoft and Published Author where she specializes in customer growth and success in Conversational AI and Customer Experience, developing strategic programs for both Customers and Partners to drive forward and add value in CX & EX initiatives.

She views it as critical we build technology that is designed for the all-channel world we live in, where conversational technology should aim to be ubiquitous with the day to day lives of users, meeting their needs as central to the core user experience regardless of what database, storage or contact center it is connected to.

Sarah truly believes customer and employee experience should be at the heart of business, and she brings this passion together with operational reality to inspire and show companies what is possible through her architectural and service delivery experience. The drive to help others through strategy and planning has directed her to create tools that help give others the ability to promote change including through content in several communities, including Forbes Business Council, Harvard Business Review Advisory Board and the Microsoft MVP community for 6 years, with the aim to educate and empower others on topics including technology, tech skills and tech literacy.