Mo Khalifa: Make sure your data's clean. If not clean, at least labeled. And if you are unsure about it, do what Microsoft does. Microsoft has what we call a a zero trust policy. Always assume that you've been compromised.

Tom Arbuthnot: Welcome back to the Teams Insider Podcast this week. We have a familiar face to the show. We have Mo. Very well known in our Team space. Uh, but he's recently taken on a new role, uh, looking after the Big four, and we talk about his experiences with Copilot, with business change, with adoption. Uh, always great insights from Mo.

Thanks so much to him for joining the show and also many thanks to AVISPL who are the sponsor of this podcast. Really appreciate their support. On with the show. Hi everybody. Welcome back to the podcast. Uh, second time, maybe third time for Mo on the podcast. Everyone in this community, uh, knows Mo I think, but you know, he's left us for, for new, new, uh, new shores and new customers.

So we're gonna talk about that. Uh, Mo how's things? 

Mo Khalifa: Yeah, things are, things are great. Things are. Really busy, but learning loads. So yeah. Awesome. Super excited, super happy. 

Tom Arbuthnot: So give us the cliff notes of, uh, your, your journey in our Teams space and then what you're doing now. And we will talk about some interesting things around kind of adoption and change and AI and Copilot.

You've got some, we had a really interesting conversation in the prep, but we just recorded the prep. Really? 

Mo Khalifa: Yeah, no, absolutely. So, so, so, hey everyone, uh, for those of you that don't know me, my name is Mo. And, uh, for FY 26, so I joined Microsoft or rejoined Microsoft, call it a boomerang. So I was at Microsoft before, you know, part of the Team's rooms business with Michel Bouman and you know, all the greats that we see in the industry.

Then I Moved over to HP Poly for a little while. Um, and then rejoined Microsoft. And this time around I've joined Microsoft as what we call an AI workforce specialist. Uh, for this FY uh, last FY it was called a Modern work, SSP. So that for, uh, that, that term hopefully is quite familiar to a lot of you. Uh, and effectively I look after everything pre-sales on a Modern work perspective, which is things like Copilot, of course is SharePoint and OneDrive and M365, and of course, Copilot.

As well. And um, I'm part of, uh, the big global Team at Microsoft that look after Microsoft's four largest customers, the big four, which is, uh, you know, hopefully everyone knows who the big four are. They are professional services. You know the big name brands like KPMG and Deloitte and EY and PWC and those sort of guys.

I think everybody knows 

Tom Arbuthnot: the names, don't they? They're huge, 

Mo Khalifa: huge forces. Absolutely. And huge customers, right? Massive Microsoft users already. And these guys are super, super excited about how they can, you know, make things More efficient for their workforce. And, uh, yeah, they're an interesting bunch, really, uh, interesting to work with.

'cause you think, hey, Mo's only got four customers, but in actual fact, these four customers have. Hundreds and hundreds of member firms globally. So it ends up being hundreds of customers. 

Tom Arbuthnot: Yeah. See I'm the other way around. Like I'm, I'm surprised you can manage four. I've worked with some of them as well as in con like to consult, you know, on the, it was on the Lync and Skype days back in the day.

But like these customers have. Awesome internal Teams, like they're, like, they have FTEs for all the stuff. They obviously have consulting arms. They have great relationship with Microsoft, so you know, they, they have reach into different product bits and pieces as well. So I feel like you've got quite the challenge to cover something that broad for those types of customers.

Mo Khalifa: Absolutely. Absolutely. You know, but I'm not, I'm not gonna sugarcoat it. It is challenging. Uh, but it's rewarding. Really rewarding, right? Because these guys are looking forward to be ahead of the game, right? Yeah. So every time Microsoft launches or announces new updates, and it seems to be happening literally on a dAIly basis.

Right now with Microsoft, we, we announce something new. And especially when we think of AI and we think of Copilot. And all the new updates that are coming through, the big four are just on it, right? They're just like, how do we get our hands on it? How do we deploy in this regulated industry that we have?

So massive, massive challenges. Um, but the good thing is, is part of Modern work still involves Teams and Teams, Rooms, and phones are still in that world as well. Yeah. All that heritage 

Tom Arbuthnot: still, still pays off. You still know what's going on in 

Mo Khalifa: that space. Absolutely. Yeah. Yeah. So super excited to be, uh, here, uh, this time around, um, adding to the, uh, the career ladder, of course, and, and, and learning so much More.

And it's really good to see all the, the good stuff Microsoft is doing around AI now and, and, and Copilot and, and all the updates, especially with Ignite last week, right? So, um, all the announcements, it wasn't last week, it was the week before. Um, uh, with all the new announcements that I've, uh, have come out and really good to see what's, what's, what's happening in that space.

Tom Arbuthnot: And those, those customers, obviously they're big, big Microsoft customers, but they, you know, they, they look at other tools and other platforms, you know, that, so it's, it is, uh, like I, I imagine a lot of your role is talking about AI at the Moment because it's just a big thing in the industry and it's a big thing to Microsoft.

But, uh, uh, like I'm sure you are being asked all sorts of questions both on the. Compliance and adoption side, and on the kind of competitive option side, what's that look like? 

Mo Khalifa: Yeah, I mean, it's, it's, it's, it's crazy, right? So, so, so literally every single day conversations will, will, will talk about things like data sovereignty and how's data processed and what happens in AI.

I mean, we. I usually tell our customers to kind of take a step back and it's, it, it's pretty much the same across all of our customers, not just the big four. So take a step back, relax, hit that reset button, and then think about where we are in the world right now. So if you remember, the last big massive change that we had was smartphones, right?

Smartphones talk on average. I think it was three years to get to like a hundred million users. And, and, and AI's taken three Months to do that. Um. The one previous to that was, was, was internet. And we had to change the way we think. We had to change the way we deployed, um, you know, uh, uh, technology, et cetera.

And the same principle we're starting to see here with AI, especially going back to the, the, the, the, the last sentence I said around how Microsoft generally is, is, is sending out new updates and new features pretty much every single day. So we need to rethink, like how do we deploy, um, you know, updates?

Do we use update rings? Do we treat it the same way as. Update that land on a PC or is it different and, and the rate we see AI Moving today, uh, is crazy. I mean, me and you, uh, and many of our listeners as well, we come from that generation of watching Terminator and artificial intelligence and all of this good stuff, IRobot, where AI is known to be this like.

Dangerous thing with a free think in mind, and it's designed to kill us all. But in actual fact, when we look at AI and we look at Copilot, and the reason why Microsoft call it Copilot and not autopilot is it's there to help, right? It's there to, you know, take off those mundane tasks and those repetitive tasks that we have.

But it's much More than that now, right? We've gone beyond what Cortana and those, those, you know, great services used to do back in the day. Now it, it's agents talking to agents and creating virtual Teams and you know, kind of Moving that way. And, you know, when we, and. When we look at kind of the way the world is today, especially with, you know, the geopolitical front and whatnot, a lot of customers really do start asking those questions around where is data processed?

Yeah. I think you've got some of 

Tom Arbuthnot: the toughest customers in terms of data processing, not just because of. Their, uh, opinions on where they should or shouldn't be, or is, or isn't, but they make commitments to their end customers as well. Absolutely. And we, we were talking in the prep, a really good conversation about like, um, like it being in a particular geo versus it being in the service.

So like, and obviously the, the anthropic Models recently came in and that was a real revelation for me. That's the first time in my memory that Microsoft have offered a. First party integration does something that is actually hosted on AWS, which shows the, the commitment to bring customers what they want.

Obviously it's optional and they opt in and all the good safety stuff, but like the, the, the, the, the, the way that AI is scaling so quickly, like is this process. In region and go local? Is it focused in the the West Geo? Is it process globally? It's quite an intricate conversation. 

Mo Khalifa: No, absolutely. I mean, I mean if you think of the last year with Microsoft and all the changes that have happened at Microsoft and even the year before that as well, you know, Microsoft's been really working hard in the backend to really support this for the future.

Mm-hmm. So we look at things like, uh, you know, the EU data boundary, for example. We made big announcements really early on in the year around the EU data boundary to give our customers that sit here. Um, you know, uh, to, to to, to give them that confidence to still work alongside Microsoft. And then of course, the UK leaves, uh, the EU as part of Brexit, and then all of a sudden we sit outside of the EU, uh, data boundary.

So a whole range of other questions start coming out of there. Uh, but Microsoft has done. A lot to change, uh, in terms of data sovereignty, where data sits where data's processed. We're being a lot More transparent, uh, in terms of providing guidance and, and support material to our customers to see exactly what happens with that data, but at the same time, constantly evolving as well and changing the way we do things, uh, to ensure that.

Uh, we respect those boundaries and we respect what customers, uh, ash, uh, asks are and wishes are around, around data. It's, uh, it's not a world. 12 Months ago, I thought I'd, I'd, I'd, I'd be so thoroughly involved in, right. It is quite easy to say, Hey, if you are the first person on a Team's call, your data's held there.

But when it comes to Copilot, it's, it's where's your data stored? Where's it, where's the transport route? Like watch data center and Microsoft's constantly spinning up new data centers. Mm-hmm. Uh, you know, literally every other Month, uh, to help support these in region asks, uh, that, that, that customers have as well.

So yeah, it's, uh, it's um, really interesting subject. Um, but you know, that's the one thing I love about Microsoft and, and you know, other competitors out there as well, is we're building these data centers to really help support customers for the future. And it's not, the way I see it is, you know how like sometimes you see new technologies come out the really cool, really exciting, but you don't know if they're gonna last.

This is one of those technologies where you are like, it's here to stay. Right? This is the new, new smart. 

Tom Arbuthnot: I often have this conversation about like, if the development didn't get any further than this, which seems unlikely, but like, let's, let's say the worst case, right? It's not gonna get any better than it's now.

We're still, there's so much we can leverage out of it today if the Models just stopped and, and we've just seen with, you know, the new Opus Model from Anthropic and five one from ChatGPT and like Google have done an amazing job with Gemini 3. Like it is still Moving forward that it probably will continue for a bit, but even if it didn't, there's so much value to be gained for businesses out of the existing stuff.

And that's really the challenge I think is. Adoption, I'd be interested in your perspective on like, uh, the included Copilot chat versus M365 Copilot. What, what are you seeing there at inter agents and, and all sorts of, you know, More advanced stuff? 

Mo Khalifa: Yeah, and, and, and you know, the, the thing I always tell customers at the very beginning, the hardest.

Part is really the adoption piece, right? And when you look at humans overall and human nature, on average, it takes us 28 days to create a brand new habit. And when you're using the likes of Copilot or any other, you know, kind of AI tool, it's about creating that habit to start off with, using it for a four week period, you know, on average.

And then it becomes second nature. That's step one. And when we look at kind of like the journey of AI and the journey of Copilot, especially, it starts with yourself, right? How can I use. Copilot to help me as an individual. So when I wake up in the Morning, um, because I'm, I, I sit within a global Team. All the US folk, for example, my US Team will be sending mails at night while I'm, you know, having my, my, my dinner with my kids.

Uh, and then in the Morning I'll use Copilot to kind of summarize and every, anywhere that I've been mentioned, highlight and tell me if there's any action notes behind it. That's now become second nature to me. I don't sit there for hours looking through 50 emails. Now I get Copilot, do it, and then, and then I'll pick and choose and be like, oh, okay, this is important.

Gimme the notes for that, et cetera. So you start off by personal productivity. How does it help me as an individual? Then all of a sudden I'm starting to use agents, right? So for example, you mentioned it really nicely, Tom, with the big four. Having just one of these customers is hard enough as it is, right?

And I'm constantly, when I'm jumping from calls, I have to remind myself, which customer I'm talking to or else I don't wanna say the wrong customer there. And it's 

Tom Arbuthnot: happened to me once, 

Mo Khalifa: once, and I learned my lesson never again. Right? But what I've done is we, I have a, like a central repository of literally all the information around my customers, right?

Everything from EA agreements to, you know, conversations, et cetera, et cetera. I built a fir, uh, an agent myself using Studio Alert, a Copilot Studio Alert, which was known as Agent Builder first to basically reference each one. So if I say, Hey, this particular customer you know is asking this question, can you remind me what it's all about?

It will go into that repository and bring it out. So it's making my life easier without me having to kind of like search through it. The stage that then comes after is how can it help my Team, because all of a sudden I'm like, this agent is great. It's helping me out a lot, and it's saving minutes. I then share that to my Team.

It's now become a, a Team agent. So my Team's now using it and then we think, um, when we start Moving ahead. We start looking at, well, how can agents then help business processes? So not just as an in, as as an individual. Uh, how can we use it as as business processes? For example, the first ever time that I joined Microsoft first time around, um, when I was onboarding, I had to actually go to the office.

They gave me my laptop. I had to sign a whole bunch of things, and they talked me through it. I have to run down to MSIT, and then they set help set up my VPN and multifactor authentication. This time around. I didn't even step into the office. Literally, it came with a barcode. I scanned the barcode. It took me to an agent.

Uh, it's like an employee agent, and the agent was, did everything for me. It ordered the laptops for me. It, it basically set me up, it set me More defactor authentication. It told me about my holidays that, 

Tom Arbuthnot: that Microsoft Onboarding Pro is non-trivial on the backend as well. The aMount of different systems integration.

So if they've got AI around that, that's impressive. 

Mo Khalifa: Yeah. Yeah. But this is the thing, right? Because when you start Moving into the More advanced stages of agents. It's what we call you. You might have heard the terminology agentic or agentic conversations. That's what we refer to. So stage one is always about yourself.

How do I make Copilot and AI work better for me when we do that day to day with, with with, with chat, GPT and Gemini and all these other good ones. Of course, Copilot is the best one out of it. Of course. Um, of course. And then we Move into, like, the business process, and we get agents now talking to agents, right?

So that's kind of like what we call phase two at Microsoft, where we, I, I have an agent that would, uh, you know, help me set up my pc, but if something's wrong with my pc, that agent then speaks to another agent. A procurement agent to be able to re uh, order me a brand new pc and then that agent can then go ahead and talk to, I dunno, um, uh, ServiceNow, for example, and raise a ticket on my behalf.

So we have agents talking to agents. In fact, um, one of my customers, uh, actually gave a really, really good quote and. It resonated with me and stuck to me really well When we Move into this phase three of the agent conversation where we have Teams of agents actually doing, you know, whole swaths of work.

Hmm. IT Teams that we have here today then become the HR Teams. For agents. 

Tom Arbuthnot: Fantastic. Yeah, it's an interesting, it, it's an interesting parallel, isn't it? 'cause um, uh, we were talking about the, uh, Manchester use group last week about this, like the, what Microsoft are really good at in enterprise, I mean, various things, but like the, the interest slash ad, the control, the purview, the intune, like all the stuff we do around managing people.

And you can see how mentally that. Cleverness jumps to, well now we need to manage these agents. So the, the, the, not saying that agents are people, but like the, the same controls, like it needs our back. It needs logging, it needs access, it needs control. So it is interesting to see how Microsoft are well placed.

And that's kind of the agent 365 story that's coming through now is, is to be a, a management control layer for the agents. 

Mo Khalifa: Yeah, absolutely. And, and it's, it's so funny to see and how crazy it is to see like the, the pace of innovation really happening in this space where, you know, it's not, people are still, we're still having this debate, uh, with, with, with a lot of customers at the Moment.

Where is it designed to like replace people? Absolutely not. No. You know, is it gonna cause redundancies and that sort of stuff, you know, hopefully it doesn't. Um, and the way we talk to our customers, especially, you know, our, the exec leaders, et cetera, is people will always be an element as part of Copilot as part of.

Uh, the agentic conversations, you still need somebody to go in and verify and make sure things are correct, et cetera. Um, but the mundane tasks you can leave through agents to do, whereas humans being so innovative that we are, we can concentrate on the More value driven tasks, right? It's not about I'm gonna replace Mo because an agent can do his job.

It's no, actually, I'm gonna put Mo into another position. Where he can actually do things that an agent can't do at the Moment. And I let the agent do all the boring stuff that Mo was doing previously. Yeah, we've, we've all 

Tom Arbuthnot: got things we'd happily hand off, which are just like, all I'm doing really is, is clicking mice and Moving things around like that, that, you know, proactive summary of content or me Morning briefing, like my favorite at the Moment is just having.

Copilot reason over my, my day and my week on the way in, on voice Mode. It'd be like, what's important? What have I gotta catch? And it's amazing, like how useful that is. And it's just, it, like I, I haven't got a person that could do that, but now I feel like I've got my own, you know, own virtual ea that's helping me out with my day.

Mo Khalifa: It's funny, Tom, that is probably one of my favorite features at the Moment with, with, with Copilot, um, voice right now where I do that. And the funny thing is, it's not just reading it out like. Like a robot, right? Excuse me. It's, it's having like a podcast. It's like two people talking to each other, right?

So you can even do that with the next time. If you have Copilot already and you click on the recap button in the top right hand corner, you'll see like an audio recap if you hit the audio recap button. You can choose between different tonalities, like executive style, et cetera. But what it will do is it will create like two people talking to each other, uh, about that particular meeting, and they're having a conversation and explaining to you.

And it's great for like, you know, if you are, if you're going for a walk or you are walking your dog, or you are, you are starting to run. I've started running, by the way, that's a complete different story and, uh, I'll tell you another day, but, uh, like I, I, I will listen to podcasts and things. Sometimes, you know, because time becomes so precious right now, especially if I'm like running to the train station, I'll actually use those audio recaps as a podcast.

Yeah. Um, and those two people talking to each other, that a, that the AI has created, it spins it in a different way. It doesn't feel like, like Siri or Google Assistant or you know, what it used to do, where it just kind of just narrates and just reads. Like, yeah, 

Tom Arbuthnot: there's, there's something it tweaks in your head, which is just a little bit More engaging, isn't it?

It's like, oh, okay. There is something there. 

Mo Khalifa: See, I, I need to pitch to engineering. I need to get them to create a Tom voice in there. So that podcast would be, I'm not sure people are ready 

Tom Arbuthnot: for that.

Mo Khalifa: They're definitely one of my favorite features. Yeah. 

Tom Arbuthnot: Yeah, and, and I think talk, talking of Ignite, as you mentioned, I think Microsoft nailed the Work IQ messaging, kind of the IQ platform. But Work IQ specifically about, you know, we talked about, again, in, in your customers, they're obviously looking at different options.

They have, you know, relationships with multiple vendors. I think that work IQ conversation about helping people understand. Why it's different to have like, uh, ChatGPT and a connector to graph versus native integration to the whole M365 platform and your, your docs, your data, your email. That's where our Team's kind of background comes back full round is the meeting transcripts, the call transcripts, the summaries, the chats, the channels.

I'm so excited now that that's kind of coming together and it has been journey, you know, these have been different. Co-pilot's doing different things now it's all converging. 

Mo Khalifa: Yeah. No, work IQ is definitely something that's been in the works for, for, for, for the past year. In fact, you know, a lot of, uh, the big four customers, they have a lot of, um, I wouldn't say influence, but a lot of say in how our products.

Shape up because they're such big customers and value customers of ours, and they have so many different use cases that we don't even think about during the, the, the, the engineering design phase, you know, and a lot of them were like, Hey, you know, we as the Big Four for example, have multiple member firms across.

Different continents in different countries, et cetera, but we, we have different languages that we use for certain bits of documents or certain abbreviations. You know, anyone that's worked at Microsoft or worked with Microsoft knows Microsoft loves their acronyms. You know, we've, we've actually got a whole glossary full of them as a dedicated website, um, and.

You know, using that kind of analogy and using what our customers have been telling us about, you know, I want Copilot to be a bit More personalized for me. We've got Copilot memory, which is personalized for you as a person, but then how do I have it where it's personalized for one of our customers?

Right? Yeah. Like the old. Yeah, so KPMG as an example, if KPMG have certain abbreviations that they use or certain project names that they use can work IQ. Be able to understand what that person is talking about regardless of where you are in the world. And then how do we go ahead and regionalize that across different regions as well.

So, you know, this is just the beginning that we're seeing with work IQ. I think it's. Super excited. I already love Copilot memory where I'm like, don't call me Mohamed. Call me Mo or, you know, end my emails in this particular way. Or, or, or, or, you know, use my personality when I'm, when I'm talking to emails.

Um, we're now gonna see that within work IQ as well. And, and, and like you said, I guess that's where Microsoft has the strength is, you know, in, in dataverse and fabric and, you know, kind of all of these data points that Microsoft has, that customers use day in, day out. We have our fingers in there already where Copilot can go in and reason, but understand More importantly.

What their users, what the customers are actually talking about, especially when they're talking abbreviations. 

Tom Arbuthnot: Makes sense. Promise will become so 

Mo Khalifa: much More easier if you think about it. Right Tom? You know, you don't have to like re-say a prompt and you know, trAIn that particular prompt and say you are such and such and you need to do this now.

Help me reason now, you know, with work IQ it should just fig, automatically figure it out 'cause it has that as part of its memory. 

Tom Arbuthnot: Yeah, and it's things like memories that are awesome component of that 'cause it just learns, but also like, like the level of data Microsoft have got to help you. If you are an M365 customer is like, it knows who your boss is, it knows your org chart, it knows the documents you previously worked on, therefore it knows your style.

So I, I still think we're early on this journey where it can just get better and better and better. So even if you're a net a new user, you are, you know, you are, you, you've got the org history on your side already from day one and the org structure as you said, that's, that's really interesting. 

Mo Khalifa: And I think, I think, you know, if I was to give any advice to any customers that are looking to get onto like the the agentic Copilot journey, uh, the first thing I would say is take a look at your data, make sure your data's clean, right?

So use the likes of purview, you know, start classifying your data. Use data classification labels on your data. What is highly confidential? What is public knowledge? What is okay? Because whenever you go ahead and then start deploying the likes of M365 Copilot. It knows which data to use and it knows which data not to touch.

Right. And that's, I think I would say that's probably been the biggest blocker for a lot of customers today, where you wanna go ahead and deploy. They're scared of oversharing, they're scared of data that's gonna be accessed by the agent or, or, or by the AI because they haven't done a due diligence in the past of, yeah.

It's nice that this has become 

Tom Arbuthnot: a bit of a forcing function. We always all knew this was the right thing to do and needed to do it, and it was one of those IT projects that would always get pushed, pushed off, and now there's a real compelling reason to get it right, not just a like negative risk reason of.

Things might go out that shouldn't go out. But a real positive one to helping scope the, the data to help the AI perform as best as it can. 

Mo Khalifa: Yeah. Yeah. It, it kind of forces IT Teams to do a, a bit of spring cleaning on, on, on the data. Um. Because, you know, we've become very complacent over the years, right?

We've got, so, I mean, just take, take, take M365 as an example, right? But prior to that it was perpetual licenses, you know, things locally stored in servers, et cetera. It took us a little while to Move to this new subscription Model with M365. Everyone was uproar like, oh, I'm not paying per Month. But now it.

You just set it and forget it, right? Yeah. Now you don't have to worry about updates. You don't have to worry about new features that just work. It just comes through and we can't think otherwise. If you've ever used office 2009 or something, like if you found the CD back in the day and you plugged it in and you tried it, you are like, oh my God, this is so basic.

Versus what M 3 6 5, um, uh, you know, uh, can do for you today. We see obviously the same thing happening around AI as well. Right? You know, we're constantly evolving, cons, constantly changing, but we need to do that spring clean on our data because we've got so complacent that. Things like Office 365 or M365 have done so well.

Um, now we need to look at what is, what can be reasoned over? How do we make sure that if I type in what's the salary that the CCEO gets, that it doesn't actually bring back that information? So data classification becomes quite, quite key with the likes of purview. And there's many of our Microsoft partners out there as well that can provide these services, uh, that sit on top of things like purview to make life a little easier as well.

So, you know, definitely that's the number one rule. Look at your data. Make sure your data's clean. Um, not if not clean, at least labeled. Um, and if you are unsure about it. Do what Microsoft does. Microsoft has what we call a a zero trust policy. Uh, and that's one of the first things they, they teach you as part of your onboarding is always assume that you've been compromised, right?

So things like how we lock the PC down to emails, everything is always gonna be labeled as highly confidential. Uh, the many times that people receive emails from, from, from me or anyone at Microsoft, I'm sure Tom, you've been on the receiving end of this a lot. It'll always say you don't have access and then you have to request access and Yeah.

You know, so it's having this zero trust policy that just label everything as highly confidential and then ask for permissions after, but organically help you grow, you know, kind of what clean data looks like. So that's piece of advice I would give. Nice. 

Tom Arbuthnot: And what were you seeing on the adoption side, mate?

Any tips? Like we, we've been having a lot of conversations on the pod around, like, like again, the, the included Copilot tons of value there, and it's right in the box. And then obviously the, the M365 Copilot. With the premium license, how have you seen, because these are huge global organizations, I imagine adoption is, uh, non-trivial.

Mo Khalifa: Yeah, so, so. So we are in a friendly space right now. Microsoft has probably done itself no favors in terms of our, uh, labeling, labeling, uh, marketing, right? So we've got Copilot chat, which is effectively free Copilot, and then we've got M365 Copilot, which is the paid version. A lot of customers don't know the difference between the two.

Um, one of the things, one of the changes we made over the summer was, uh, giving Copilot chat, which is the free version. Free to pretty much everybody. So if you have M365, if you're using a Windows pc, or if you're using um, edge as a browser, you'll now see Copilot pop up on the side. So a lot of customers automatically assumed, Hey, I've got Copilot, but not M365 Copilot.

And the key difference to remember is. Copilot chat, AKA free Copilot reasons over what's on your screen and kind of goes out to the internet and is web grounded. So we'll bring information back. M365 Copilot AKA paid for Copilot or licensed Copilot reasons over enterprise data. Plus the web and everything else.

So you get More with that one. So we started seeing adoption rise when people were using Copilot. As you, you, you see it pop up and you're like, oh, let me ask about this page. Or you know, Google does it really well if you've got, if anyone's got an Android phone and you do a quick, you know, search using the bar, the first thing you uses Gemini to to, to kind.

Kind of bring up your search results, Copilots doing the same. Um, so when it comes to adoption, you know, we see Copilot chat as being like the entry point. So people are starting to use it, uh, across there. Uh, and then when we start looking at M365 Copilot, um, paid for version, then all of a sudden we're asking questions about.

Deeper data, and then we're starting to see that. But you know, the question that you asked me around adoption and any tips around adoption? Always remember what I said around, uh, it takes 28 days to create a habit, right? So it's a slower process where we see Copilot being used the Most. That's within Teams.

So definitely after meetings, that's the number one area. Yeah. It's a 

Tom Arbuthnot: pain point that everybody gets, right? Like everybody has lots of meetings, everybody wants great notes, and, and I, it's a kind of automatic or one click solution as well. So it just kind of brings that value. It's such a lifesaver, 

Mo Khalifa: right?

Like taking notes, um, and, uh. Like if you, if for those of you that know me in real life, like I'm like the smallest person in the world and I've got one big work, really bad habit where I'll ask for your name and I'll completely forget remembering your name, right? I'm so bad at it. Whereas, you know, if you think of like co if I had Copilot sitting on my shoulder, for example, it would automatically take those, those notes.

Every time I ask for people, I, I'm 

Tom Arbuthnot: signing up when it's available. I would, same thing. I, you're like, yeah, you get to meet a lot of That would be amazing. A lot of people at events and stuff, don't you? So keeping up is a nightmare. 

Mo Khalifa: Yeah. Yeah. So, so that would be, that would be amazing. You know, kind of, kind of doing that as well.

So, so we see Copilot being used the Most when it comes to Teams, right? So how do we do meeting recaps, et cetera. That should be a starting point because we see that people using it. We, we see Copilot now within meeting rooms as well. So when we think of MTR and we think of, uh, Teams itself, that's where it's used the Most.

Then we wanna start asking questions. The next biggest one is Outlook. So then how do I reason over emails? How do I ask about, Hey, I've got a meeting coming up. Can you help me prepare and it will then go over to Outlook, et cetera. Um, we, because Copilot now has entry points across all apps as well.

You'll now see like when it opens up in Excel or PowerPoint or Word, Microsoft tries to give like little hints or little prompts to say, ask it this, ask it that. So generally getting people to do it, but I always recommend, um, the number one thing we talk about is. Prompts, right? So how do you, how, how do you actually do prompts?

How does AI respond to prompts and building, building out prompts? So, uh, generational prompts, talk about Teams, talk about Outlook, and then slowly starting to introduce things like analyst or researcher and showing the power of what it can do, um, but make it relevant to the business that you're in.

That's the key thing. So Microsoft has. A lot of really good learning material, but it's very generic and especially like my customers, the big four customers that have like different lines of services like auditing and consulting and tax and legal and those sort of stuff. Whenever I talk about Copilot to any one of these lines of services.

I have to kind of personalize it, that what's it like a day in the life of a tax person, a day in the life of an auditor? Um, and, and actually customize those prompts to help 'em with their work that they have for that day. So I always use myself as an example and I'll be like, this is what I do when I wake up in the Morning.

I've got kids, right? I'm a typical guy with f uh, you know, typical father with kids that drive me nuts and everything else. I have very minimal time throughout the day. So how do I use Copilot to help me across this? How do I create a script? For this podcast that I'm doing as an example. So make it relevant to the users that you're using.

Utilize the learn.microsoft.com pages as well. Um, but make sure you give yourself enough time. No one's gonna learn, learn it in two weeks. It's a 28 day human nature. Create a habit and then it becomes second nature. 

Tom Arbuthnot: Awesome. Well, Mary, thanks for dropping so much knowledge. Really interesting to hear that new role and, uh, yeah, I, I'm, I'm not envious of the challenge.

You have to keeping up with some, I know some great people in those organizations, so it is, uh, you, your, your up against really strong, strong internal Teams. I'm sure they keep you on your toes. 

Mo Khalifa: Absolutely. They're a great bo, they're a great bunch. And, you know, it's, it's still really good to be working with, uh, all of you guys, uh, all the time as well.

So, uh, yeah, I, um, Microsoft's a, a huge organization and copilots just bringing everything together and we're seeing it. We are seeing everything merge, right? So no doubt, um, you know, whatever I talk about within Core Buy, that falls into the AV world and vice versa. And yeah, yeahinteresting, it's, it's all coming together.

Tom Arbuthnot: It's one of our biggest conversations in the Team space and it is nice Teams is such an, as you said, one of the, the fastest adopting platforms for people to get AI and understand AI and drive business value out of it. So yeah, it's all, uh, all interlinked one way or another. 

Mo Khalifa: Absolutely. 

Tom Arbuthnot: Awesome. Absolute cheers Mo, nice to catch up and, uh, yeah, I was probably appreciate you one of the events soon.

Mo Khalifa: Cool. Thank you, Tom. Take care.