So we’re gonna start off today with standing up a new Treasury DXP’s five year Treasury roadmap. And I want everyone to pull out their apps and make sure that you can make votes. Basically, we have a couple of questions during the call, or during the presentation that we want you to be able to interact with. So for this, you’ll want to go into today’s event, and then you can see the questions as we go through them. Perfect and basically follow these instructions. I’m going to wait for people to stop looking at their phone before we get started.Thanks. Okay, and with that, I think we can go through today’s agenda.
Bobbi Cadena:
Can everybody hear me? So today we are going to talk a little bit about DXP. We’re going to talk about the initial process, how DXP adopted Treasury transformation, benefits of AI enabled cash forecasting and key takeaways. So a little bit about DXP. DXP was founded in 1908. And so we’ve been around for over 100 years. For those of you that don’t know, DXP stands for Distribution Experts. We are headquartered here in Houston, Texas. We are a billion dollar annual revenue company. And our big three main subsidiaries, our service centers, innovative pump solutions, and supply chain services. So here’s kind of an overview of our Treasury landscape. Myself, I report that as our CFO, I have two Treasury analysts, one analyst’s main focus is on our m&a and bringing our new acquired companies on board, and also works on cash reporting and cash forecasting. Our second Treasury analysts main focus is going to be cash reporting and cash forecasting. And then our Treasury admin who works on all of our payments. And she does troubleshoot issues, voids, checks, things of that nature. So some of the objectives of our cash forecasting model are to have complete visibility on our cash flows. So before I started with the XP, they kind of monitored their main operating account, which DXP has almost 200 accounts now. So they definitely needed visibility to what’s happening on all of our other accounts, to understand the impacts of the acquisitions that we were making, on how they impact our free cash flow. And then to make short term and long term decisions. So just a small overview of how I came to be at the XP.
So before I was hired, and again, I’ve only been there a little over four years, there was no Treasury Department, our CFO, and our controller managed all of the Treasury. It happened right around oil and gas crashing, that we started coming really close to breaking bank covenants. And they had to hire a third party company to build out models for them they didn’t already have. So start shopping out for a different Treasury services bank, in a different Credit Services bank. So that’s when they decided they didn’t need to hire somebody to work in a treasury department and build out a treasury department. They found me after I had just come out of chapter 11 from my previous company, where I got to work hand in hand with all of the consultants on rebuilding our complete finance department, rebuilding all our reports, rebuilding our forecast rebuilding our reporting, and so they thought that it would be a good fit for the role. When I came on board, there was nothing in place. I had to build all of the models.I had to build a cash report. I had to build a forecast.
And considering the company is very acquisitive, I did a lot of data gathering and it took almost a year to put a model together. So once I had something in place, I went to management and said hey, we need something to streamline. As I need something easier to pull this data that I’m, you know, spending hours a week pulling. And that’s where we started shopping around looking for a company to help us do that. And that’s kind of where we learned about high radius. And one of the things that we like about high radius is that not only did it have a treasury solution, but we also needed work in our AR team, and it had several AR solutions that we really liked. And that’s when we went with a high radius. So what’s next is, basically to get up and running with our pro our projects, finish, see where they’re at, see where we’re at, see how they’ve helped us along, and then work on other aspects of Treasury, such as improving payment, automation, eliminating checks, things of that nature.
So a little bit about our initial cash forecasting process. And this is all stuff that I had to put together, do manually pull lots of data from our earpiece, we have multiple entities, so I had to go into multiple entities to pull historical data I tried to do about two years back, I also had to get historical payroll data, I had to get historical tax data, how to utilize a lot of our bank statements to just build out actuals, so that I can then build out a forecast that took me like I said earlier, almost a year to get done.
Once all that data is gathered, I can use trending analysis and kind of see where we’re at in our aging, how we did the month before to build out what’s going to happen in the month going forward. It took a long time, it was very time consuming. also trying to build a treasury team, it was just a lot for one person to handle. So we definitely were looking for something to automate and make our lives a little bit easier.
Amber Thompson:
So now we have the first question that we would like you guys to fill out. So the first one, the first poll we have is what is the number one issue that you face in your company today for cash forecasting? Is it data gathering, variance analysis, your forecasting approach or the modeling?
And while we’re filling this out, I want to ask Bobbi as well. We’re, what do you think the answer would be here for DXP?
Bobbi Cadena:
Let’s see, I would definitely go with data gathering, considering how many entities that we have to pull data for. So the biggest one.
Amber Thompson:
And it looks like about 80% of y’all who’ve answered so far 83 now agree with her. So it does look like data gathering is one of those harder, harder pillars here to accomplish, especially when you’re working with a manual forecast process. So these four options that you have to choose from here actually make up the four pillars that you can use to make a highly accurate forecast. They start with your approach. So the question you want to ask yourself on this is when you’re thinking of how you’re going to approach forecasting, are you going to be making?
Are you going to be building a model that is more of a direct cash flow model where you’re building your forecasts from the bottom up? And then consolidating at the end? Or will you be building something that’s more like an fp&a model?, where you’re looking at the top down and trying to estimate your more granular information? for data gathering, you want to ask, how much automation do you have in play? And are you actually capable of getting all of the data that you need to create a proper forecast? In the third one modeling Really, the question here is, are you using the most optimized model for every cash flow category that you’re forecasting? And finally, with variance analysis, you want to ask yourself, to what degree are you completing a variance analysis. And I’m going to go through these in a little bit more detail. And we’re also going to add in a couple of examples with Bobbi here today. And I want you all to think about where you sit on the maturity model for each of these four pillars. So to start us off, we have the approach, and it’s often said in the Treasury that forecasting is more of an art than a science, but I will say that you can actually get more to a science driven approach when you’re capable of including technology into the forecasting. So, Bobbi, where would you have placed DXP when we first started the high radius implementation?
Bobbi Cadena:
So we placed DXP on a laggard approach. Basically, a lot of it was intuition type things. And again, like you said, art versus science, where I’ve been doing forecasting for a while now. And it was more of a gut feeling and an intuition thing, and oh, I already know this is going to happen, I know that’s going to happen. So that’s why I chose that laggard.
Amber Thompson:
perfect. So you can mature your model, as you improve your forecasting by adding a bunch of manual man hours to your forecast. So you can bring in more people to collectively build a forecast, you can even bring them in to look at information at a more granular level. But to get to the best in class, most objective approach, you will need to build a bottom up forecast for every cash flow category, basically, across the field. And remember, I want everyone to think about where they’re sitting through each of these pillars, so you can know where you stand today. So for data gathering, your maturity will increase the more that you have automation. So in these first two stages of the maturity of data gathering, it’s going to be a very manual process. And to get from laggards to proactive, you’re just adding on something like a little bit more data. So you’re going to include something like forward looking adjustments, which is right about where DXP is today, right?
Bobbi Cadena:
Correct. So we do a lot of trend analysis with our forecasting. But we also work super close with our CFO. So we’re kind of in the know about maybe an m&a project happening, or maybe a bonus payout that’s happening. I also work closely with our AR group on trends and how quickly we’re getting money in so I can make those forward looking adjustments versus just using the trending analysis.
Amber Thompson:
Perfect, thank you. And then to get to the strategic investing class maturity, the best way to do that is by adding automation. So robotic Process Automation can bring in more data and give you a wider pool to be able to model off of.The next one and modeling is going to be directly affected by what data you have access to. So when you have very little data model you’re very likely to be error prone. So if you’re looking at just last year’s data, that’s not necessarily going to reflect with what this year’s forecasts will be. If you want to have a high accuracy. As you shift more into a mature model with a best in class forecasting, you’ll see that you’re using different kinds of modeling more, more complicated and a little bit more information that you need to have to make those models and best in class will actually bring in something like artificial intelligence that can optimize every cash flow category, and optimize the models for each of those. So Bobbi, can you explain to us why you’re sitting between two shirts,
Bobbi Cadena:
So I put DXP kind of struggling, proactive and strategic, mainly because I took a lot of the knowledge I learned from my lap prior to this job. And I do proactively reach out to each department that affects our cash forecast. And when I’m re forecasting, okay, do you know of anything big that’s coming up are taxes going to be slightly different this month, you know, is this month a payroll month that we have commission bonuses, things of that nature. So it’s not just using the trending analysis, it’s also kind of getting a little bit more in depth information from departments that weren’t
Amber Thompson:
Makes sense. So you’re able to use and optimize your models based on what information you have access to, but not quite at the point of being able to use artificial intelligence yet. The last pillar we have on here is for variance analysis. And you’ll notice that the first one on here is actually no variance analysis, if that’s where you’re at, you’re not alone. But you do want to try to mature that option a little bit by performing that, at least at a global level. But as you mature more, you will update that more often. And also perform that variance analysis to a degree that actually involves different cash flow categories. It looks at your forecasting by different regions and different company codes. So for DXP, we have it logged as a proactive maturity. Can you walk us through what you do today?
Bobbi Cadena:
Sure.So our variants now This is more high level, okay, we didn’t get as many receipts in as we have projected, or we had a bonus go out for payroll that was interred last minute, very high level, but we aren’t able to kind of drill down into really the root cause of why the variance was off.
Amber Thompson:
Thank you. And one of the things I want to make note of of this variance analysis, like Bobbi mentioned, is getting to the best in class where you have information drill down to that detail will allow you to understand where the variance is coming from, which may help you understand why and how you might improve it. So, now that we’ve gone through each of these pillars, I want everyone to take a minute to note down and log into their phones where their current process sits. So far, it looks like a majority of you guys are similar to DXP, and that you’re sitting somewhere around the proactive. And this may not be exactly where you are like DXP, you might be shifting between both of those stages of maturity depending on what pillar we’re looking at. So with this, I want to hand it over to Bobbi to walk through some of the challenges that she’s had with her initial state that might be putting her near her current phase.
Bobbi Cadena:
So challenges that DXP faced we’ll go through this pretty quickly here. Number one gathering historical data. As I said, we are a big m&a company. And even gathering data from the companies that we’re acquiring is not always really easy. It’s a very time consuming, very manual process. My borrowings could have been a little bit better where we’re not spending so much money and interest in fees to the bank. And then at the end of the day, we’re just not giving our C suite, really accurate data.
Amber Thompson:
Now we’re gonna do a quick summary of how DXP is currently implementing high radius to adopt a treasury transformation. So since the number one issue that we’re facing is automating that data gathering, that’s really step one of how we’re improving her current forecasting process. So we are automating data gathering from both her erp system and her bank, so that we can use the larger pool of data that we can access to be able to improve on her current models and even optimize those cash flow categories. And also setting up automation on the variance analysis.
Bobbi Cadena:
So the desired outcome now has to go through this quickly, central repository for all of our data gathering, saving time and resources, accurate forecasting, to better give information to our C suite on m&a and forecasting cadence.
Amber Thompson:
And in summary, we’re bringing DXP’s initial forecasting state from about proactive to best in class for each of those pillars. And to wrap us up, I want to go. I’m gonna let Bobbi take us through those key takeaways for you guys, today.
Bobbi Cadena:
So a couple of takeaways that I’d like for you to all think about. Number one, don’t fear AI, we are moving into a technology world. Technology is the wave of the future, embrace it, learn it further, educate yourself, don’t be afraid of it. And to take a minute to see where you’re at and your Treasury team or your finance team. And how can it be improved upon and what can you do to learn what needs to be done to take it to the next level?
Amber Thompson:
So if you know where you’re at today, and you know where you want to go, you can help you better determine how to get there. And with that, does anyone have any questions?
