What to expect of CoPilot in Power BI using Xero

I want to talk to you about what you can expect of Copilot when it launches in Power BI. I want to show you how I see it's going to be used. Rather than it being like a gimmick, I want to talk to you about how a small business owner using Xero, how it could actually be used in a practical way.

So what is Copilot? Before I go too far, I just want to make sure everyone's on the same page. OpenAI is the owner of ChatGPT, and Microsoft have got a commercial relationship with OpenAI where Copilot is using ChatGPT, but it's also using its product. So Copilot's going into Excel. It's going into Word. It's going into Outlook. It's going into Windows. It's going into Power BI, which is what we're going to talk about right now.

So the closest way I can show you how Copilot, I think, is going to work in Power BI, because it's not live right now, is by showing you the code interpreter tool with ChatGPT. So for anyone who's not a paid subscriber, the way you get Chat is just in this dropdown, Code Interpreter. If it's not turned on, you've got to go into your settings to turn it on. You've got to have a paid subscription to have Code Interpreter. Let's get into it.

My example is going to be with one of our demo companies, Old Man Online Clothing. So they're a Xero file. They have just a heap of sales and costs and things over a five-year period that we've flushed into the system with tracking categories. Lots of different things that we do with it for examples like this. So this is the background. What I'm going to do is I'm going to put this data into ChatGPT directly. Please don't do this with your private data. Speak to us if you want to have a chat about security concerns.

So a quick way to show you what we're going to do is to go to the Executive Summary. It's one of the reports in Xero that's available. I'm going to set it up to compare with the last 12 months. Okay, so here's the Executive Summary. It's got Cash, Profitability, Balance sheet. It's just a whole lot of random demo data and we're going to export that to Excel. Okay, so I'm writing a question. "I've uploaded my monthly executive summary of my business for the past 12 months. I need an analysis of my business. Where are the issues where I can improve? What am I doing well? What am I doing not so well? Use graphs to analysis." And then I'm going to use this little upload button and upload that Excel file that we just downloaded from the Executive Summary.

So let's have a look what it's returned. So first off, it readjusts the data so that it can read it properly and it's come up with some ideas. It's saying, "Look, we're going to do an analysis on cashflow, profitability, performance, position." I mean, these are just the headings of the Executive Summary, so that's not really that special. But if we scroll down, it starts with cash flow, by the looks. So it's graphed some of those cash flows for cash received versus cash spent. It's picking up this huge gross profit in December. Because remember, it's an online shoe company, so they have huge sales in December. And then look, the net profit falls off a cliff after the sales in December. So it's actually picked this up. If you notice this comment, it says, "There's seasonal variations."

And so let's scroll down. Then it says, "Look, okay, we've examined cash." There's some examples there, some pretty generic examples. And then it said, "Let's have a look at business performance." So it's looking at return on investment, ratios. Look at this about debtor days. "Generally, debtor days have been around 40-60 days, which suggests you have the opportunity to improve cash flow by shortening this cycle." I mean, these are actually, they're generic comments, but they're really strong citings and observations about the business that you can actually use with your client to continue the conversation. Remember, with Copilot, you're the pilot. This is the Copilot. They are actually giving you some very interesting graphs, some observations that you can actually use to explore further.

So then it just gives a little summary. It says, "Look, strength in some areas. There are clear opportunities for improvement, particularly in cashflow management." This has made some concluding comments. I love it that it can actually do this and you can continue to interact with it. So now I'm going to go back to it as well and say... So I've just said, "Look, I have large sales in December for Christmas trade and I order a lot of stock to ensure I can meet demand, but sometimes I order too much and I put strain on my cashflow because I don't know how much to order." And let's just see what it comes up with.

So it's understanding the seasonality. So look at this. I mean, demand forecasting. So you've got to get better at your demand forecasting. It's talking about inventory forecasting, inventory management systems to handle this. Have cash reserves and consult with your financial advisor. I mean, I really like, I just think that this is practical and useful.

So we circled this back, because we want to picture this being in Power BI reports, and we've got this little helper Copilot here. Here's a classic return on investment. A small business owner is a lot of the times quite financially illiterate. What does return on investment even mean? Why is that useful to me? I love it that you can have these graphs and management reports put together by the accountant that the small business owner can say, "Why does return on investment even mean something to me?"

Look, it's important because you can use it to make decisions on new hires, on marketing campaigns, on buying equipment for your business. It's a good way to goal set and progress tracking. I mean, I think this is a reasonable answer. You could actually re-ask it and say, "Look, explain it to me as a seven-year-old." It's actually just giving you confidence that you can make smarter decisions moving forward.

So we haven't even touched on the other aspects of Power BI with Copilot, like actually writing formulas, creating different visuals. I mean, I'm just showing you how it could be used as an analysis tool. But I think that the use case here with Copilot coming is extreme. I think my gut is that accountants are going to start using Power BI a lot more than previously, just because now there's been that skill gap with writing DAX formulas, with modeling data, with relationships with tables. There's been a skill gap there for smaller accounting firms. I think this does substantially bridge a gap here and we're going to see a much bigger uptake of Power BI users at the smaller end of town.

Anyway, that's all from me. I look forward to your comments. I think Copilot's a big tick. I think Code Interpreter is a good example of how it can be used, and look forward to hearing from you. Cheers.