Before you hand off your sheet to AI, here's a few things you need to do and a few suggestions that I would say will help your AI help you, and they will not take very long. So here we have a table . It's just some sales data . We have some months. And I just wanna say the very first thing you should do, which I think actually helps me with Sheets as well, is if I have any sort of helper information, summarization, um, dashboards , anything else other than this table , just move it to another tab. Let's say I did move this to another tab. This is just like a SUMIF . I can move January, February, and it'll tell me the sums here of that month. Just move that to another tab, m-make sure it works, and then go ahead and

delete that row So get a clean row as your header row in row one And then make sure your header names are absolutely representative of the data inside. And don't name it something that you personally said, something cute, something clever, and also try to use real words. So for example, this percentage sign might have been because you wanna keep this column very narrow. If I write percent, it's filling up the whole thing here. For using AI, it's okay. Be more verbose than you normally would. The reason is, is that when we as humans look at this header , we are gonna make some assumptions, and we actually make a lot of assumptions as we read it.

For example, this column, column D, has no header . I would expand this a little bit and write what this is. This is an indicator, you know, green for something. This is finding the maximum. So if you say maximum, it's okay, but I wanna be more clear that this is an emoji and that this emoji means Equals maximum Clear here is better than clever. And just know that if you don't have information here, the AI is going to infer. Like, the AI doesn't really think, but there are times when it's going to make some assumptions. And so what I would do Is just really be exact here. Instead of sales, sales count or sales amount, don't use something that is very

vernacular Or something that is very particular to your-- or very jargony, very particular to your work business. It is okay if it's generally the right thing to say in your line of work, but if it's something very specific to your own company that only you would know, change it to that industry-wide thing. Rep, for example, it's okay, but I think it's going to help a lot if you say something like sales representative or even more clear of like who sold or name of sales rep Now this might look bad, and it's okay for this particular case. It's just taking a few moments, maybe not even minutes, to just look at your header and really make sure that 90% of people who are gonna read this spreadsheet ,

if they don't work at your company, are gonna understand what these columns mean. And I really do mean 90%. So try a couple of iterations and just go for the most general thing. In the moments where you have some kind of internal calculation , something maybe that is industry-wide, like CPM is okay, CRM, CRO. But CPM, M- MVP, CPV, those are okay because 90% of the people in your industry are gonna know it and you have a wide industry. But if it's like a particular indicator, say what it is. Don't just let it assume. And again, this is not for you to present this and send it to someone el- a human in your company. If you're sending this to a human in your company and, and they've seen this before and they know the data , you don't have to do these changes.

This is very specific to how do I prepare this data so that AI can help me better? For example, I would put in parentheses the currency, especially if I don't have the currency in the data . But also sometimes our formatting in Google Sheets will make the data , underlying data different than what it s- shows and what it sees, like if you add the currency as a format . So I like to add little bits of information. Instead of net here, I would say profit, and again, USD. Something where it's very, very, very clear that this column of data … R- remember, we're giving the AI data, not this sheet. Now, here's another thing that I would do, and I think it's actually really

important to know that in the future, this distinction is going to be very different. What I'm about to say here is my own personal choice, and I'm gonna try to give you the best recommendation possible at this moment in time. However, in the future, file format may not be of importance. But I'm gonna share with you some ideas around file format . So we go to File and download either a CSV or a PDF . Now, in my little experience with this, I have tried to give CSVs as much as possible. My theory behind that was, was that data is good. Now it's getting a little bit grayer. It's not actually becoming even easier. So I would say actually in some cases, a PDF is going to be better or the same.

And why I would say 50/50 right now, because previously I, I said, "Okay, let's just download a CSV ", especially because the download process is faster here. So if I am on the sheet that I'm on, when I go to File, download a CSV , it's going to download exactly that tab. and only that tab. The file name is going to be the tab name and the sheet name When I do file, download PDF , I'm gonna have some kind of thing to go through, and sometimes I don't really wanna do this. If I'm getting AI to help me, I, I wanna get to that pr- process faster. So in the past, 100% of the time I was doing download CSV . Now, I'm doing 50/50 CSV and PDF , and I do believe there is a benefit to

PDF , and in the future we'll actually be doing PDF much more And, and that's because AI is knowing more and more. The more AI develops and the further we get into more and more different models, it is actually understanding these kinds of indicators. It actually is understanding formats much more. And so a CSV is gonna give you the underlying data . Let's look at that CSV And it does give us this emoji. It gives us the percentage, dollar signs over here. Great And a PDF is gonna have a couple more things. It's gonna have the actual shading and color of all of the rows and columns. It's going to have color indication of the months and this

profit over on the right side. And yes, actually, so one of the key things here, actually, I think I did this in a previous video, is yes, the color means something in profit, but also the data is different. So it has literally shape of the data is different when it's negative. It's, uh, has these parentheses. Th- that thing is going to help you with AI. So if I haven't mentioned it yet, I've mentioned tons of previous videos, don't ever rely o- only on On the color or shading of something. Make sure your data has some data there, some data difference. If it's, uh, the maximum is colored, like this row is colored for the maximum, this indicator that it's different than all these, this is data . This is real data that shows, okay, this emoji is different than this m- emoji. The character is different.

And even say maximum. Even… You could write in maximum here. In, in the video that I showed you how to do this, I did some cute emojis. Great. At least those emojis are different characters. But I do believe that PDF over time f- is going to take more hold and also help us discuss with AI these kinds of things and have real good output from AI if you just send a PDF . And the cute thing about that is that you don't have to do so much work in preparing this data Which I think is actually beneficial to us as work. We, we may not wanna change the shape of data . We might not want to prepare data for AI. We wanna say, "Okay, this is what I'm working on now. This is how it is, and this is what it looks like. Help me, please." Talking to AI So I hope this was fun for you and il-illuminating.

I don't think this takes more than two or three minutes, honestly, to review a few things, change a few headers . If you do take time, I would say take the time in really preparing your data for als- other humans too. Don't just rely on color, don't rely on shading, rely on the data structure itself so that you can always talk to AI well. This kind of thing, the, the green emoji equals maximum, is also helpful to h- other humans. Other humans include the people you work with, people outside your company who might be, like, trying to help you, consultants might be AI, but also, to be honest, it's you in six months. Help yourself in six months when you, when you're not checking this every day and you're like, "Okay, I need to check the sheet now. It's been six months. Oh, what does this mean? What does this green dot mean?

Oh, here it is." Six months from now, I'm not sure if I'm gonna remember that yellow and green, what they mean Also, the last thing I would say is that please understand that preparing data for AI doesn't really mean preparing it for only a machine. It means preparing it for 90% of the world. Like, people who generally can discuss these things and understand them, think of them. Think of that person who's like, "Okay, I, I know a bit-- I know a little bit of this thing that you're doing. Talk to me like I don't, I don't know deep down exactly what you're doing." So I think that's very important to think. We're not preparing this just for a machine You are watching better sheets here on YouTube. Make sure you check out this video or this video and subscribe right now to get more tips, tricks, how tos, get more out of your Google

sheets than you ever have before. I'm excited to be making a ton more videos here. Ask me questions down in the comments and I will answer them in future videos. But for right now, right here, one of these videos is gonna be your next Google sheet.