AI VS Excel

Hey, everyone. Welcome to Finance AI Academy powered by OneStream. My name is Tiffany Ma and today we'll be talking to Chris Cordovari about AI versus Excel. Chris, thanks so much for being here today with us. Yeah, thanks, Tiff. It's great to be here. My name is Chris Cordovari and here at OneStream, I'm a global sales director as a part of our AI and operational analytics team. So Chris, we all know finance users love Excel. Excel definitely has its strengths, but there are also some weaknesses.

What are your thoughts on Excel within the office of the CFO? Yeah, I think the first thing, Tiff, is that the capability of Excel is always going to be around in the office as CFO, although Excel does have its limits. It's great for personal productivity, productivity across a small team. It's not something that is connected to the broader set of models or reports that exist across an enterprise that is often needed to drive strategic decisions.

I think the other thing that we know about Excel is that it's very much error prone, right? A lot of studies suggest that 80 to 90% of Excel models, especially the ones that are complicated, have errors within it. And some of those errors, when you're making decisions off of these models, can become very problematic. We talked a lot about Excel's capabilities. Let's look beyond Excel and talk about capabilities that OneStream brings to the table that aren't necessarily possible in Excel alone. In order to get the value of a time series machine learning forecast, you need two things.

You need the data and you need lots of it. And the data becomes the fuel. And that fuel is what needs to power machine learning models.

And machine learning models, as you know, Tiff, require lots of compute, right? And both the data as well as the compute and the scalability, are things that OneStream brings to the table where our customers can finally tap into the vast amounts of information that they have collected within their company and start to use that as an asset where they can create forecasts that are data-driven, they're objective, and they're leveraging the power of the machine learning models that we bring to the table inside of OneStream.

And the result is that they can produce forecasts that are much more accurate than doing so in a traditional way with a driver-based model, for example.

These forecasts can be produced in a much more automated fashion where rather than sending out forms for people in finance to fill out that you have to collect and aggregate.

Sensible AI forecasts can simply take you from zero to about 90% of the way there in an automated fashion.

And then that final 10% is where finance tends to bring their business intuition to the table where there's always going to be something that they know that the model doesn't know. It's this combination of the power of AI machine learning with the intuition of the humans and all the intelligence that the finance community has together that helps our customers get a great outcome.

When it comes to forecasting with sensible AI forecast, you don't get this automation and this type of scale within a productivity tool like Excel. So what's your recommendation to finance teams on how to get started with AI and ML?

What we have seen to be very successful with our customers is find that division or find that business unit or find that part of the business where you have that problem, that if you can solve it, there's a significant business outcome on the other side.

And focus there, get a win, because that win and that demonstration of value that we can deliver in a matter of weeks, it's going to give the company the confidence as well as the motivation to take this new way of working further.

So that first win is really important. And typically that win is what funds project number two and number three and number four. And at the end of the day, when customers, especially large enterprises are taking on these initiatives, they're going to pay a dollar to get a capability.

They have to get two, three, four, five dollars in return. And I think starting small, starting focused, demonstrating value in a matter of weeks, is the fastest way that you can build momentum. You can get people to start to work in a different way. And then you can take that momentum further to start to scale it out across the enterprise. So in short, find that problem that's worth solving. Attack that problem with a lot of intention.

Solve that in a matter of weeks and then start to scale that out more broadly across the enterprise. So start with a small, quick wins, deliver value, and then proliferate out to the rest of the organization.

Exactly right. And I would say the biggest thing that I see as a part of these projects is the technology part. More often than not is the easy part of the project.

The hard part is how do we get individuals in finance that have been working in a very specific way for a very long time to work in a new way, right? And start to use solutions and trust solutions that are powered by machine learning and that have generative AI built into it. That's a tough barrier to break down. And one of the things that we've done a lot with our customers is we've taken them on this journey where at first we deploy the solution and perhaps the company uses it as a challenger model.

They run their current process and they run this new AI powered process in parallel. And very quickly over a forecast cycle or two, they can start to see the efficiency gains. They can start to see the accuracy improvements of this new way of working and the individuals who are responsible for these plans and forecasts. They start to lean in. They start to see the proof of this new way of working. They start to gravitate towards it more. And then where companies start to evolve to is they no longer do their old process. They let the machine learning capabilities take over to an extent.

where these capabilities can get them from zero to 90% of the way through the process. And then the finance teams can spend time doing the things that they like to do, which is analyze the business, allow different scenarios, bait different strategies to make decisions to drive the business forward. Hey, Chris, thanks so much for joining us today on Finance AI Academy powered by OneStream. Yeah, thanks for having me, Tiff. Enjoyed the discussion.