 What's going on everybody welcome back to another video. Today we're going to be taking a look at Microsoft Copilot. Now I think we can all agree we've been waiting a really long time for this because Microsoft announced Microsoft Copilot back in February of 2023. Now it's February of 2024 and now I'm just getting my hands on it. There were a lot of rumors and beta testers who were using Microsoft Copilot who were basically saying it wasn't working at all and so they kept delaying and delaying and delaying the launch for all of us normal users. But now us normal people can start getting their hands on it, start using it and seeing how it works. Now in this video we're going to be doing quite a few things. The first thing we're going to do is I'm just going to show you how you can get Microsoft Copilot if you're a Microsoft user and if you want to pay for it. Yes, it is a paid product. I don't think anyone is shocked by this but I'm going to show you everything that it's integrated into in the Microsoft ecosystem and then I'll kind of do some demonstrations along the way. I'm not going to demonstrate every single thing is integrated into. I think you'll see kind of a pattern here of it's pretty similar in a lot of their products. Like it does a lot of the same things. The one that we're going to be focusing on the most is Copilot in Excel because it is probably the most different out of all the ones it's integrated into. It does a variety of different things. We're really going to dive into it. I think that one is going to be the biggest change for most people because a lot of people have been using chat to BT for emails or writing notes or summarizing things and that's what it does in a lot of the other ones. But in Excel it has a lot of different functionalities and hundreds of millions of people use Excel every single day. And so I think that is probably the one that I think most people are going to be interested in seeing. So without further ado, let's show them on my screen and get started. All right. So here we are on my Microsoft account. You have to have a Microsoft account in order to get Copilot. I think that makes sense. But all you have to do is go to your subscription and you go to the managed subscription and you can pay for Copilot. I'm not for this video going to stop paying for it and then pay for it so you can see it. It isn't crazy hard once you get to this part of your profile. You just go to your subscriptions and you pay for it. And so here it is cost $20 per month, which isn't terrible. It's the same cost as something like chat, GBT, but it's going to be integrated into everything. So I think that's a fair price. I don't think it's crazy, crazy expensive. Now let's come up here and I just want to show you what it's currently integrated into. Here we have Copilot in Teams, Copilot in Outlook, Copilot in Word, PowerPoint, Excel, OneNote in Loop and in Whiteboard. Now the ones that I personally use are Teams, Outlook, Word, PowerPoint, Excel and OneNote. I don't use Loop, Whiteboard and I don't really use this one at all either. But as you can see, I use quite a few of these and so $20 a month to have it in every single one of these products that I use almost every single day. It's not a bad deal. Now before we move into Excel and I'm going to show you kind of what they tell you you can do in Excel. I'm going to demonstrate just a few of these and I think you'll get a pretty quick feel of exactly how it works. For our first example, let's start out in Word and right when I open a Word document, this is what I get. So it's automatically giving me this option. I didn't ask for it, it just doesn't. And so we're going to ask it to do something for us and we're going to see exactly what it does. I'm going to say write a few paragraphs about the differences between Microsoft Excel, if I can spell correctly, it's nice, but Microsoft, Microsoft Excel and Google Sheets. And then I'm going to go down here and I'm going to click generate. As you can see, it's starting to put everything on the screen. This is taking maybe 15 to 20, maybe 30 seconds. As it's pulling in the information, it's putting it on the screen. And so there you go. We have a lot of different information. And then we have this thing down here. Now this allows us to edit the prompts so we can go in and change what we actually wanted it to do. We can say keep it or it'll actually keep all these changes. We can refresh it to have it repopulate this information or we can delete it, or we can make different changes to it. So if we say, okay, add this, take away this, it can do that. But if we just bring this back over here, we can also open up the co-pilot option. Now this is what you're going to see in almost everything. This little display on the right hand side of the page is going to be what you see in almost all the applications that it's integrated into. Now what we can do is we can take any type of document and we can just say summarize this doc. So I'm not going to type it out. I'm just going to click on this and it's going to summarize this entire document for us. Now for this example, we're working with just random sample data. But if we had a really large document, let's say like a hundred pages, we can have it summarize it into maybe two pages or three pages and it would do that for us. It's going to summarize the main ideas instead of four or five sentences. We now just have about one or two per topic. Next we have OneNote and OneNote is going to be really similar to basically Excel, right? You have this co-pilot on the right hand side. I'm going to say generate 10 ideas for YouTube videos around data analytics. Put them all in the list from best to worst. All right, just like that. And we're going to see what it comes up with. Now we can just copy this, paste it right over here. Now we have all these different options and maybe some videos I need to think about making. But it's really similar to ChatGBT or Bing. It's going to give a lot of similar responses. It's running off of ChatGBT4. And so all the stuff that is output in ChatGBT or Bing is going to be extremely similar here on co-pilot. The very last one that I want to show you is PowerPoint. Now all the other ones, almost all the other ones are going to be the exact same. You have this thing on the right hand side and you prompt it to do something. It's mostly just working with string and text. So even if you go into email, it's basically the exact same thing. So I'm just not going to dive into every single one. But now let's look at co-pilot for PowerPoint. Now I have this prompt down here. Create a presentation on Excel for data analytics. Start from basics to advanced. Have a total of 10 slides. And let's see what it comes up with. And now it just populated this. Let's take a look. It says welcome to Excel for data analytics. Let's scroll through and just see what it has. Okay. This isn't a ton of super useful information, but it is a good, maybe a good format, a starting place to start with. The pivot tables in Excel says introduction to pivot tables, creating a pivot table. So maybe I needed to give it a lot more context. This is just a quick example. But as you can see, it created a PowerPoint with 10 slides. And I need to go in and I need to do a lot of work to make this right or add a lot more context, a lot more information for it to understand and give me a lot better output. But there's just a quick example of what it can do. So that was just a quick example of some of the ways that you can use co-pilot. Of course, in Teams, you can have it summarize the conversations because it's taking all that, putting it into text, and then it just summarizes it. So pretty similar, but a really neat feature. You can write emails. And now we're going to look at Excel specifically. Now let's come up here, because I'll put all these links in the description, by the way, if you want to take a look at this. But this is co-pilot in Excel. Let's see exactly what it can do. And then let's actually test it out ourselves. So we have co-pilot in Excel. It says analyze, understand, visualize your data with ease. Let's come right down here. It says you can help explore and understand your data better. It can analyze and visualize your data. And then it can effortlessly highlight, filter and sort your data. And then it can generate formulas. So this is exactly what it can do. It doesn't go much beyond exactly what it told you. But let's actually test it out and see how it works. All right, here we are in Excel. I have my co-pilot button right up here. Now one thing I'm going to tell you is there are some things you have to do in order for co-pilot to work with your data. The first thing is, is if you pull up the co-pilot and you come right down here, it says I only work in an Excel table. If you have a table, you just need to click inside of it. So if this data wasn't in a table, I have to go in, I have to insert a table right here. I have to specify the data and put it in a table. That's the only way it works. Now once I click into this table, now we have a lot of different options. We have add formula columns, highlight, sort and filter and analyze. Now really quickly here at the bottom, I have different tabs and those are there for a purpose. For example, right now we only have about 15 rows of data. We have one, two, three, four, five, six, seven, eight columns. In the stocks data, we have about 506 rows of data that may be about 10 or 11 columns. With temperature, we have quite a bit more. So now we're working with 1500 rows of data and then we have this bus breakdown and delays NYC and this is going to have quite a bit more data. If you can't see that, let me zoom in quite a bit, 282,000 rows. So there's a lot of data. Let's go back up to the top and this data comes from my Excel for Data Analytics course on Analyst Builder. If you haven't checked it out, shameless plug right there, go and check it out. I'll leave a link in the description. But let's go back and we're just going to go to this sample data. Now this sample data was generated by Copilot itself. It said you want to generate some sample data and I said yes and this is exactly what it generated. Now I'm going to start off really, really, really simple. I'm just going to show you some of the things that it can do and then we'll get to some of the limitations that I have personally found when I was using this for the past week or so. The first thing that you can do is ask it to do some general generic things, kind of some things that you could easily do yourself, but you kind of want them to do it. So you can just say filter the budget where it's greater than and we'll say 1,000. So I'm asking it to filter this column where it's greater than 1,000 and it takes a little bit. I'm doing this in real time so we're going to see how long it takes. It looks like it doesn't really understand what we're asking. Maybe I need to use the right capitalization here. This did not happen, filter the budget column where it's greater than 1,000. When I was testing this there were some things that would work, some things that wouldn't work or it didn't understand what I was asking it to do. So right now it really doesn't understand what I wanted to do although when I was testing it out earlier it was working. So I guess that's one of the limitations. Let's do the next one. So we're going to say highlight and I'll say total users targeted. We'll say highlight the top five, total users targeted. So this should be using right over here conditional formatting. So if I come in here I do highlight sell rules. I can do greater than. I can say the top five or actually it's the top bottom rules. So I go to the top 10 items and I just say top five. So now it highlighted the top five for us. Here's one thing I will say just while we're here because again I know it's going to be a long video. The co-pilot takes a long time. It usually takes like 15 to 20 seconds whereas if I already know how to do this it takes me about like five seconds. If I could just come over here conditional formatting, top 10 I'll do top five and click okay that takes me like three or four seconds whereas this takes like 15 to 20 seconds. Here's what I will say about that and I'll get to some of my overall thoughts at the very end but for the average user who is not a data analyst or who really doesn't use Excel or know Excel well I think this will be useful because maybe they don't know how to use conditional formatting and they just want to highlight it and they're able to converse with it. It makes it more user friendly for the average person. For maybe like you and I who are probably better at Excel than the average user this is going to be really slow. It feels very slow a little bit slow and a bit clunky but that is something that it can do. The next thing we can do is ask it questions about the data. So we can say which owner has the biggest budget overall and so I want to ask it this question and we have our owner right over here so that's a completely different tab. I didn't say campaign owner so let's see if it understands that and right here it says here's a pivot table based on your prompt and so then it says we have the top campaign owner by budget. Now let's add this to a new sheet because this is something that it does a lot so we added it to this new sheet and you can see it created this pivot table it has Andre Lawson and then right over here we have the sum of the budget. That's what it likes to do with almost everything. It loves to create a pivot table and then do something with that pivot table either create a visualization or just summarize the data in some way. That's what it does with almost everything which is pretty neat because I myself use pivot tables quite a bit. Now let's see if it can create a visualization. Let's come over here and we're going to go to this temperature data. Now this has 1500 rows of data. Let's see if it can handle it and let's see what it does. So let's say create a time series. Let me spell things right. A graph showing the average humidity per month and let's see if it's able to do that. Now it's working with more data so we have 1500 rows of data with about five columns which again isn't a lot of data. Let's see how it does. So it looks like we have our graph here. Let's add this to a new sheet. It's going to create a completely separate sheet and right over here it has the pivot table that it's creating this visualization on. I personally really like that because I want to see how they're actually creating the visualization and you can always go in and change this. So if I want to get rid of months I can do that or I can you know just add different information that I want to add. It's all here. They kind of created it for me but I'm able to edit it myself. The next thing that I want to show you is right over here on the stocks. Now with this stock data I want to ask it to create a formula for me and I'm going to be a little bit tricky. I'm not going to give it exactly what I want it to do. I wanted to try to view some logic here but what I'm going to ask it is we have this 52 week high. We have a 52 week low and I can zoom in on this a little bit. So I want to see the percentage difference between the lowest point and the highest point. The lowest point in the year if you bought at this point and you sold at the very top basically what is the difference or how much money what I have made on my investment. So I'm going to say create a column that calculates the percentage change from the 52 week low to the 52 week high. Now this is difficult a little bit because I'm not giving it the formula and there is a formula behind this and so I want to see if it's able to calculate this and this actually is something that I really like about co-pilot is its ability to write the formulas and functions and so there we go. So we have this 52 week change column right now. I'm going to insert this as a column and now we have that information right here and so from 175. Let's look at a really simple example. Here's 52. So we have 42 all the way up to 64 and so half of this which is about 50% is about 20 so 42 plus 20 is about 64.6. So you can see that this is really accurate. We're going from this to this and that's a 52.8 percent change that goes all the way down and now this is another column in our table that then we can use. So now we can come in here and we can sort this we can go from largest to smallest and now we can see these ones went from about 200 all the way down to about 100 percent. These are really really really big changes from its smallest to its largest point. So this is really useful and you can do a lot with the formulas and functions. I will say for me personally I think this is one of the best things about co-pilot because I always am researching different functions and formulas and calculations and then I'm putting them in here from the internet. But if I can just ask it and it knows it natively because it's all built into co-pilot that saves you a lot of time. Let's start getting into some of the negatives of co-pilot. I really have run through the bases of what this can do and what it can't do. Here's what I found that it can do. It can do pretty simple tasks. If it's in here like it said filtering creating pivot tables creating visualizations you can ask it to highlight certain values. These are all what I would consider quite basic. It isn't crazy advanced. But very very helpful especially with the formulas that's what I do like. But if you really ask it anything context-based or you ask it to do things like clean data or even separate out let's say this sector. So let's say with this this is just an example. Take the sector column and break it out into two columns using a space as the delimiter. This is going to be too advanced for it. I've tried it. It just doesn't work and I just want to break something out. Break out a column based off of a delimiter. It can't do those things. It even says right here they just can't do it. So there's a lot a lot a lot of limitations to what it can do and probably one of the biggest downsides is it does not work with large data sets. For example we have this really large data set over here. There's 260,000 rows and that's a lot but honestly it's not a crazy amount. This is probably what a lot of people have used and I'm just going to ask it to do anything. I'm just going to say I'm going to say filter on the reason column to be I'll say heavy traffic and this is something that in smaller data sets it's absolutely going to be able to do but in this large data set it's not going to work and so here it says sorry I'm having trouble right now try opening the chat or sending a prompt a bit later but typically what it's saying is I can't do anything more than two million rows so that's typically what it's doing. The other thing I wanted to note is it only works if your data is in the one drive so if I have something stored locally on my computer it does not work. It has to be stored here because it needs to be able to connect to that data process it through their Azure AI systems that they have set up and then spit back out what it needs which is why it takes so long to do but you have to have it you can't just store it locally so it's not like this software is on your computer all these new features and all the co-pilot is stored in Azure and works in Azure and you have to have that internet connection so if you don't have an internet connection and this isn't in one drive you can't use co-pilot at all and so that's just kind of unfortunate but I'm going to leave on at least this on a solid note because what I'm going to show you is one really cool last thing we have this launch date it is a date custom date but I just want it to be I'm going to say change the launch date to month month day day year year year format and so this is just one last thing I want to leave on a positive note before I go into kind of exactly what I think about everything as a whole so I think that is really cool you can change kind of the formatting or the data type of the actual cells so that is Microsoft co-pilot I feel like I showed you the vast majority of what it can do and it can do quite a few things and a lot of different applications that Microsoft has which is really great now let's talk a little bit about just the positives and the negatives the positives is that if you're in the Microsoft ecosystem this is gonna be really great because you basically have chat gpt or bing right but chat gpt 4 integrated into all of your products and so generally I think that that will increase your productivity if you have lists to make if you have things that you want to summarize you know it's really really helpful or responding to emails these are all things that all of us do on a daily basis in the real world and so I really do believe that a lot of people use this on a daily basis for a lot of those kind of menial or mundane tasks that they don't want to do I also think it's fairly user friendly so the average person out there maybe not someone who works in tech but just the average person out there is going to be able to use it and ask questions and kind of understand it and kind of have those conversations with it to change things and do things but I do think that the average person is going to get pretty frustrated at using co-pilot if they haven't used any type of AI before because I ran into a hundred different problems with internet connection with my files not being in the right place with just not being specific enough and asking it exactly what I needed and it went and did something that I didn't want it to do so then I had to undo it or I had to go back and change it which at times was genuinely pretty frustrating because with chat gbt I always had to copy and paste it but when I'm asking it to do something directly it would just go and do it which didn't always do exactly what I wanted and so then I had to go in and I had to like delete a column or reverse something and that was a little bit frustrating even for me for someone who's used AI for the past year and a half ever since it started really getting popular even I was getting frustrated so I just know that the average user out there will probably have some frustration with using AI until it becomes more user friendly and it doesn't have as many issues which I think it definitely does right now but overall for a lot of the applications that I've tried it in it does exactly what it's trying to do now it does it take a long time sometimes yes sometimes upwards of 30 seconds to do something that I think you know chat gbt if I was online in chat gbt it would take it like five seconds natively in co-pilot it's taking 25 30 seconds for some of these tasks but with that being said it is giving me the output that I'm asking for for example when we were trying to create a list of youtube videos it is giving me a list of youtube videos and it is pretty similar to talking with chat gbt but again it just takes a long time to get those responses and I'm kind of getting used to faster responses and so waiting 25 seconds or so it really does feel like a long time when you're waiting there I know that might just be me nitpicking it but I think for the average person that's gonna be a long wait I genuinely think that wait time is like too long people are so used to just instant gratification on their phones they swipe up there's a next video they post something it's there in two seconds like people want things to be really fast and co-pilot is definitely not fast with that being said overall my experience with co-pilot was about a seven out of ten I think it is vastly going to improve over the coming years it hopefully will get faster I think the features will expand I think this is a really good start I think the average person who's using it is going to enjoy some of these features and I think that's especially true for a lot of these texts summarization or creating of text which a lot of people use chat gbt for in word doc in emails and in one note these are the types of things I think people really use a lot in excel I actually found it to be quite a bit slow I could do a lot of those things a lot faster and the limitations without having it be connected to the internet how to save your file in one drive and it can't handle a ton of data so there is a lot of limitations within excel I think the one thing that I probably liked more than anything was just the functions and the formulas being able to create those it understood what columns I was working on and then it spit it out for almost everything else I felt like I could do it a lot faster and it just had limited functionality in terms of anything breaking data out any data cleaning really a lot of the stuff that we as data analysts would do and you know use with data it really couldn't do except for I would consider that a lot of the basics again I am very pro AI so I've been really excited about co-pilot but again 7 out of 10 is pretty good that's a really good start especially since this was just released I think in a year two years three years they'll polish out a lot of these things they'll make it a lot faster which I think will make the user satisfaction rating go up quite a bit so that is my full initial review of Microsoft co-pilot I will make other videos in the future about it because I really really really think that this has a lot of potential especially because it's such a ginormous ecosystem of people who use it so I will be keeping up to date on it if I see any new features or anything I think is really really interesting I'll be sure to make a new video about it so with that being said thank you guys so much for watching if you haven't already check out analystbuilder.com that is my data analytics learning platform where you can practice real technical problems as well as take all of my full courses and if you really enjoyed this video be sure to like and subscribe and I will see you in the next video