Three Ways Finance Leaders Align Cash Flow Forecasting and Strategic Planning

[Joe Fleischer] Welcome to our webinar “Three Ways Finance Leaders Align Cash Flow Forecasting and Strategic Planning”, brought to you on cfo.com by Industry Dive and sponsored by Trovata! I’m Joe Fleischer. I’ll be your moderator. Before we begin, I’d like to go over a few features of the console you’re viewing. I want to point out that in the middle of your console, you’ll be able to view slides and respond to polling questions. In a moment, I’ll go over how to respond to polling questions when they appear. But first, I would note that in the related content area of your console, you’ll be able to click on a link to a case study from Trovata, the sponsor of our webinar, as well as download a PDF document comprising slides from our webinar. Although we will plan to dedicate the latter portion of our webinar to addressing your questions, we do invite you to type in your questions and comments at any time during our webinar in the Q&A area of your console. You’ll also have the opportunity during this live webinar to earn a certificate, representing one continuing professional education or CPE credit. To be eligible, we ask you to respond to all three polling questions we’ll intersperse throughout our webinar. To respond to a polling question when it appears, we ask that you select the radio button that corresponds to your answer and then click on the “Submit” button so that we are able to record your response. After we have posed our third, our last polling question that is, during our webinar, if you’ve answered all three polling questions while attending our live webinar, you are therefore eligible to receive credits, you will be able to click on an icon to download a certificate that will be located in the certification section of your console. I would add that if you’ve answered all three polling questions but are not able to retrieve your certificate, you will be able to do so when this webinar is available on-demand later this week within the “Past Webinars” section of cfo.com.

Now, I’d like to tell you about our distinguished guests. They’ll have the opportunity to elaborate on their respective backgrounds, but I think some background is certainly in order, at least before we begin our discussion. Joining us today, I’m thrilled we have, joining us today are Brett Turner, founder and chief executive officer with Trovata, his colleague Joseph Drambarean, chief technology officer with Trovata, and Linda Copeland, associate principal with The Hackett Group, providing some high-level perspective on the various topics we’ll address during our webinar. By way of background, I would mention that and point out that Brett Turner is an experienced senior executive with more than 20 years in public and private high-growth technology companies. He’s developed a strong track record for building, financing, and growing tech startups from early stage to successful exit. He began his career in controller roles after starting as a CPA at Deloitte. He later served as an SEC reporting manager at Amazon. And more than 15 years ago, he began a transition to CFO, a role in which he served for several companies. And observing what was happening in retail banking, and with a front-row seat to the digital transformation of enterprise IT, it became clear to Brett that the next major digital transformation would take place in every company’s finance and treasury operations and how they consume banking services. So he founded and serves as the CEO of Trovata, which offers a native cloud software as a service platform to visualize cash positions and automate cash forecasting so that companies can better use the financial troves of data they generate every day. Brett will provide more background in a bit. Joining, as I mentioned, Brett is Joseph Drambarean, chief technology officer with Trovata, working with key Fortune-level brands, including Capital One, Marriott International, Microsoft, Harley-Davidson, and Allstate Insurance. Joseph has helped brands navigate the digital landscape by creating and executing innovative digital strategies, as well as enterprise product integrations that incorporate cloud architecture, analytical insights, industry-leading user interface and user experience, and technical recommendations that are designed to bring a measurable return on investment. Sharing a high-level perspective will be Linda Copeland, associate principal with The Hackett Group. Miss Copeland is a results-oriented finance executive with more than 25 years of experience in financial planning and analysis with market-leading companies. Throughout her career, Linda has improved the performance of financial planning and analysis teams by reengineering business processes, realigning resources, and strengthening partnerships with business operations. She also has a proven track record of improving profitability through revenue enhancements or cost reduction efforts. And indeed, it is now my pleasure to begin our discussion. And what I would like to do is pose our very first question to Linda and then Brett and then Joseph, as I pose this question to you and as you answer it, you’re welcome to elaborate on the background I’ve shared. I do want to start off in the meantime with our first question, which is “What are the most significant limitations finance leaders encounter as a result of relying on traditional assumptions and methods when forecasting cash flow?”, starting with Linda, then Brett, then Joseph.

[Linda Copeland] Thanks, Joe! Hi, everybody! It’s Linda Copeland with The Hackett Group. Before we dive into the question, I thought it’d be worthwhile to just give you a little bit of a context, maybe set the stage for this upcoming conversation. As many of you know, The Hackett Group provides benchmarking transformation, as well as technology services to companies like yourselves. But to highlight some of the benchmarking we’ve done, when we looked at some of the key reasons why people were not effective in planning their cash flow, there are four different, four major categories of areas that companies tend to focus on. Number one is your ability to predict and understand the changes in your receivables as it relates to your cash cycle and understanding and having the tools that will allow you to do a better job at predicting all those elements of that cycle, the cash delivery cycle, right? And that’s really part of what we’re gonna talk about today. In addition to that, your ability to plan your inventory is another major item related to managing your cash flow. I mean, all of these things you’re going to be very, very familiar with. But I thought I’d pull that out as well. And then also, when we asked our clients about the types of tools that they use to provide their cash flow forecast, and I don’t think anybody would be surprised by that, but in many cases, the number one tool used to develop cash flow forecasting is Excel. And I think what we’re gonna show you today is that there are other processes or other methods that you can use to do that. And so when we get back to this question about what are the limitations, the way I kind of look at it when I work with my clients, one of the major limitations is associated with whether the information you’re providing is actionable, right? So traditionally, what we see treasurers using are other folks within the finance team to predict your working capital as the example with is DSO, right? So it’s a great metric. But really, it doesn’t really provide you the level of detail to take action on whether to sell to a particular client, whether you need to pay more attention to a particular client. So traditionally, we see people using these metrics, right? But again, we need to get underneath that. And that’s really what this conversation is about today: “How do we get underneath the basic metrics used to produce a cash flow forecast that may be the treasury approved and provide more actionable information by understanding the data that predicts the DSO, right?” Breaking that data down into actual pieces of drivers that are going to drive your performance. And we’re gonna talk a lot about that today. And my colleagues on the phone are gonna get into detail about that as well, right? And then finally, obviously, if the majority of people are using Excel to predict cash flow, it has its limitations the same way it has its limitations on financial planning in general, right? You’re limited to what you can do from an analytical perspective to understand how to take action against driving your ability to forecast cash flow. So those are some of the comments that I thought I’d start the conversation with, and I’ll pass it off to my colleague. Brett!

[Joe Fleischer] Thank you very much, Linda! Thank you so much, Linda, and Brett, and Joseph. I’d love to hear your perspectives, not only with regards to tools, capabilities but even the times in which we’re living, how the times that we’re living in have forced us, perhaps, to rethink traditional assumptions and methods. So, Brett and then Joseph, I’d love to hear your perspectives, and you’re welcome to elaborate on your respective backgrounds as well.

[Brett Turner] Great. Thanks, Joe! Yeah, I appreciate it. Great to be here with everybody! Obviously, this is a topic that’s very near and dear to my heart. Also, it’s great to be doing this and sharing with you, alongside Joseph Drambarean as well because you’re gonna get a great one-two punch around. It’s opening question, too, just right at the heart of what’s going on in finance around limitations from Joseph and I. And really, I think a big part of this is telling this Trovata story that really is solving a lot of these limitations. So right out of the gate here, significant limitations that everybody knows is time, team resources, and IT support. Just everything has changed much. I think when you look at the dynamics of just, you know, whether your perception of the world speeding up or if you look at even just public valuations in the market right now, last fall, with all the major investment banks coming out with their reporting is that all of the companies that are data-driven or B2B SaaS or have this data automation mentality to their company, they’re valued. They’re sort of having have-nots. You’ve got companies valued at 30X forward revenues and higher. And so you can see even just from a market standpoint, just valuing the need for data in automation, and who has it. And so when you hear these terms like “Data is the next oil,” things like that, well, it depends. It’s not valuable. Data is not valuable if you don’t know how to use it. So I think at the end of the day, you know, my background is just being, over 20 years, as a CPA controller, financial reporting specialist, having done tons of cash forecasting, working in startups where… You know, if there’s ever a time where cash is king, it’s definitely in high-growth technology startups because you’re having to raise a lot of capital, make efficient use of every dollar that you have because you’re reaching these milestones and working through those phases. And so all of that has to come down to efficiency. And like everybody else, my background is working in Excel, building what many of you likely have on the phone here just… You know, models are great ways to manage your forecasting in Excel. And you kind of look at, you know, “If cash is king, why aren’t there better tools to manage this?” You think of these cash workflows around cash forecasting. They really sit between the bank and the ERP system. And folks, really, today, most 99.9% of companies, other than the large elite treasury shops, where there are 5,000 global companies that… And, you know, half of those have a TMS to be able to do at scale, but even those are sort of lagging behind in their data requirements or the speed at which you can utilize and help drive decision-making. So at the end of the day, how do you get over these resources? They’re finite. And you kind of look at how everything has sort of changed in IT, and needing IT resources to be able to take advantage, it’s all about getting access to data in real time. And then, everybody’s finding they can’t really just Excel their way out of this problem because you need to be able to drive to that decision-making faster and you need to be able to build all of your tooling faster to be able to enable that. So at the end of the day, we’re at this point where there’s a massive paradigm shift that needs to take place, and everybody’s kinda looking ahead cuz you’ve gotta future-proof your data. And this whole notion of… Like back in my days in Amazon, it was a bias for action. There was always a part of trying to six sigma everything you can in the corporate finance area. Well, now, with COVID hitting, it’s like now everybody has to because now more of that pressure is on in terms of that “What’s gonna happen next?” And it’s all about heightened risk and preparedness that you’ve gotta focus on. So more than ever, you have to be able to have your data strategy nailed down. You have to be on a path to drive automation, or otherwise, you’re gonna start to fall behind. And that’s never a good situation.

[Joe Fleischer] Thank you very much, Brett and Linda! Joseph, I’d love to hear your perspective on limitations regarding traditional assumptions and methods when forecasting cash flow.

[Joseph Drambarean] Yeah, and thank you for having me! It’s great to be on this webinar. And just a little short snippet about my background. I have a very different background from Linda and Brett. I come from the technology sector, primarily within cloud and modern high tech, focusing on big data platforms, machine learning. And I have spent quite a bit of my career focusing on the real-time economy, whether it’s within banking or within e-commerce. And the problem set of real time is really fascinating when it collides with what is essentially here the collision with the old-school way of doing assumptions and planning and the methods by which you get the data that you will use for those assumptions. And what I have found as a fascinating discussion point, even just so far in this discussion, is the fact that we see the reality of, even in the current economy that we’re in, that real-time decision-making and reacting nimbly is of utmost importance, especially in the volatility that we experienced across every sector. And if you want to be nimble, you have to rethink how you have the access to your data, how that data is set up in a performance orientation so that you can be nimble, and then ultimately, whether or not you can depend on that methodology long term. And if you compare that with the old tools, for example, using Excel, Excel, when it was first built in continued form, was intended to be a personal database. It was intended to be a piece of software that sits on your local hard drive, that allows for you to manipulate a database to do certain things, whether it’s planning, whether it’s analysis, whatever it might be. It was never intended to be a performance and distributed centralized resource for managing data across multiple actors, as well as multiple departments. And when that happens, when you kind of extend it beyond its core capability is when you run into issues. And what we have seen here at Trovata is really this moment in time, where the call to action is around data centralization and data transformation because what it’s leading to is an infrastructure that future-proofs your ability to plan nimbly, to react nimbly, when there are changes that can happen on a daily basis and allow for you to do it in a way where you are centralized, allowing for access in an entitled way across different stakeholders within your organization. And I think that those themes are really important, and they challenge the traditional assumptions because, in the past, it might have been okay to do your modeling and your planning on your local hard drive, doing it in a way where, maybe, you generate a report or a model that will inform a forecast for a whole quarter or a whole month. But when you’re in a situation where you have to plan multiple scenarios, 10-15 different scenarios, given different lever points, whether it’s your operation span, whether it’s sales fluctuating in the volatility of what’s happening right now with COVID. All of that requires a data strategy that is built for scale and built for performance. And without that, you’re kind of attacking the problem with your hands tied behind your back. But I’ve seen this as a really interesting opportunity to dig in and roll around in the notion of “What does data centralization and data strategy mean for this topic?” because, ultimately, it’s at the heart and it’s the most elemental form of planning and forecasting. Bad data in, bad data out. And that’s really what I have found is a fascinating component of this.

[Joe Fleischer] Thank you very much, Joseph! And we’re certainly gonna elaborate, especially on some of the technological advances that enable organizations to adapt and evolve beyond traditional assumptions and methods when forecasting cash flow. Before we continue our discussion, I’m gonna pose our first of three polling questions to attendees. What attendees are welcome to do is select the radio button that corresponds to your answer, then click on the “Submit” button. Our first of three polling questions ask attendees to indicate the extent to which you agree with the statement that your finance team has full visibility into underlying drivers that have the greatest effect on your company’s cash flow. You are welcome to select the radio button that corresponds to your answer and then click on the “Submit” button so that we can record your response from top to bottom. The choices are “Strongly agree”, “Agree”, “Disagree”, “Strongly disagree”, or “I don’t know”. And after you select the radio button that corresponds to your answer, we do ask that you click on the “Submit” button so that we can record your response. What we’ll do is give attendees sufficient time to respond. Then, what we will do is very briefly reveal how you have responded. And I think what we’ll wanna do then is continue our discussion. In particular, we’re going to continue our discussion after displaying how you’ve responded in aggregate, “How finance leaders should collaborate with colleagues outside finance?” But before we do that, again, I wanna make sure attendees have sufficient time to respond to our very first of three polling questions, which once again ask attendees to indicate the extent to which you agree or disagree with the statement that your finance team has full visibility into underlying drivers that have the greatest effect on your company’s cash flow. I think what we’ll do is give attendees just a few more seconds to respond. And then, we will briefly reveal and summarize how you have responded. And then, I look forward to continuing our discussion with Linda, and Brett, and Joseph. And what we can see from among respondents to our very first of three polling questions is the following. It is clear that there is agreement, so close to two-thirds agree. But I would be remiss if I did not acknowledge that more than a quarter of respondents disagree, and nearly 8% don’t know. In this context, I wanna pose our next question. And I wanna pose this question first to you, Brett, then Linda, and then Joseph, asking how you would advise that finance leaders collaborate with colleagues outside finance to gain visibility into underlying drivers that have the greatest effect on their company’s cash flow. Again, starting with Brett, then Linda, then Joseph.

[Brett Turner] Yeah, that’s interesting on the polling question because, you know, with 23% or so, 24% saying “Disagree”, you know, I think that’s… And then “Agree”, I think… “Agree” and “Disagree”, but that’s not a big resounding endorsement for having access to all this because the big thing that I found and sort of another why Trovata is really that… You know, forecasting, as we all know, has been sort of this… If there’s a cloud-based forecasting tool, it’d be a bit of a holy grail in the industry. And it’s the problem. Why it hasn’t been built earlier? The clouds have been really around for 10 years. Salesforce started roughly 20 years ago. It’s not an application problem. It’s a data problem. And the biggest question is, “Can I get access to all the different data sets that I’m gonna need in order to build a really robust forecast and then to do scenario planning and stuff?” So definitely, the application is a part of it. But if you can get access to all your data in a way that you can control it and manage it, then the biggest problem, every time you build a forecast, you’re having to go back and refresh your data sets because your forecast is obsolete the next day or the next week. And so if you constantly have to refresh and regather all of the data from a lot of complex sources that aren’t feeding nicely or playing nicely with what you’re working on, it’s Excel, then you’re doing all of your, in the tech world, it’s ETL, all your transformation, all your normalization of all the data, you’re having to do all that work in Excel yourself. And that’s all this prep just to make it usable in the model or the reporting that you got in there. So I think that’s at heart of why you really need access to all this underlying data. And having that centralized for you is paramount to all this. But when you look at a big part of just really underlying the… You know, getting access to the data, but then also knowing really what to do with it to be able to really build good assumptions. And that’s so much about just managing risk and then knowing your business. So I think, like anything for any forecasting is that, especially now, when you’ve got to really stress test more of your assumptions, you gotta build out more scenarios because of risks that are heightened with all that’s going on is that it’s just critical that you know your business. And so, everybody in finance wants to be more strategic. I think that’s the aspect of it. But in order to be more strategic, you really have to understand the business, know the business, and a big part of… You know, what I’ve kind of learned in my career, and it’s sort of, I would say, a big testimonial that’s helped me a ton, is just early on, it would always be if I could get the data out of the ERP system and I could democratize that data with reporting to the different business heads, and in doing that, then it helps them, and then they start investing and teaching me more about the business. And so I’m giving them information. And then, all of a sudden, I’m learning about the business. And that’s how I learned so many aspects like at this company in telecom called World Wide Packets, where we had a really difficult close-to-cash process, and I needed to understand these drivers by really working closely with our head of ops, who was really one of the, I would say, sort of a legend in just logistics and offshore manufacturing and the telecom sector. And just learning all those aspects about how tooling is gonna affect margins, all these different aspects to it, it helped me understand the risks. It helped me to understand the risks from his perspective as well. And I could start making my modeling and all of the risk assessments or various assumptions or scenario planning much more adept and more fine-tuned to all those risks, and it made a huge difference. But I think that’s being able to then… It also has the forcing mechanism. It’s then by default you’re learning more about the business, and you become more just relied upon and more indispensable from your company and the different people who know that you have that information that can help them out in terms of kind of frontline decision-making. So that’s what I would just say is key about the data but then paramount of really engaging with different business folks leading the various aspects of the business.

[Joe Fleischer] Thank you very much, Brett! Linda, I’d love to hear your perspective as well.

[Linda Copeland] Those are great comments, Brett. And I’m gonna follow up on that and take the conversation a bit in a different direction in that. What we have seen over the last five years or more, particularly, is that the tools that are available to the finance team have changed dramatically, right? Well, we once thought, when I was an FP&A for 25 years, what we thought was unachievable has now become a real possibility because these tools are becoming more and more complex and providing us with more and more capacity to do what we do in finance, in which you should be more of a consultant, right? And what that does is it places the responsibility on the finance team to become more of a reviser. And what we’re working with our clients to do now is transition the finance team for more data accumulators and data managers, as was mentioned earlier with Excel, and, basically, database managers and working with them to transition the team to provide more consulting opportunities, meaning that the team you need in place to help support these advanced analytics is different than the team you have today, right? And these tools have great effect capacity. So we’re working with teams to realign their teams and their capabilities of the finance team to have people on staff that can then become consultants and look at things less about trends and more about drivers. We want them to understand the business more than just the finance. Joe?

[Joe Fleischer] Thank you very much, Linda and Brett! And Joseph, what perspective would you bring to this question?

[Joseph Drambarean] You know, I think both of the comments so far have been really interesting. Just as I’ve been thinking about the problem holistically. But as I was thinking about it, one of our customers came to mind. So we work with Square on one of our products called The Treasury Cloud. And I was reminded of how they framed the issue when thinking through the long-term viability of their data strategy. And it was that it’s the wrong perspective to think about your needs, from a finance perspective, just solely through the window of the apps that you’re used to or the processes that you’re used to. Instead, what you need to think about is the consistency of the data, the availability of the data, and the resilience of the data. And those three topics, when you attack them kind of head-on, you’ll realize that a lot of that is an IT problem. And so I think to answer the question directly, “How should a finance leader collaborate with colleagues outside of finance?”, well, I think a primary stakeholder that needs to be a part of the conversation is IT because ultimately, to Linda’s point, the transition needs to happen naturally, where all of the database management tasks, the integration tasks, the data cleansing tasks, the reliability of that data, the extensibility of that data, which is becoming a very powerful driver in the decision-making… When I refer to extensibility, I mean, “Can that common data set, banking transactions, balances from all of your different accounts across different currencies, can that data be harmonized and made available to many different applications within your ecosystem, whether it’s your ERP, whether it’s your payment platforms, whether it’s your analysis platforms, whether it’s even Excel?” Having a common model makes the job of analysis and the job of advising more predictable because you have a common playing field. And I think that IT needs to be involved in that conversation because, ultimately, they have to drive the resilience and the commonality of that data. What that does, though, is it frees up the folks in finance to focus on what they’re best at, which is strategy and analysis, and taking that data that is common and resilient and providing insights that can drive the business. And I think that that’s an interesting transition that’s taking place, that because there’s a double click on data transformation and having reliable data, it’s creating this duality between IT and finance, working together as partners to deliver value that will be long term. It’s no longer just an isolated task that sits just within finance. It’s a task that needs to be attacked across different organizations if you want to be successful, especially if you want to accomplish things like real-time analysis or real-time forecasting, or in the upper echelons of using AI and machine learning to extend your abilities from an analytics perspective. You can’t do any of those things if the data platform that you have isn’t standardized. So I think that it’s been interesting to kinda hear both perspectives because both are true and both are accurate. And they lead to the same answer, which is you need a foundation that you can move forward with.

[Joe Fleischer] Well, thank you very much, Joseph, and Brett, and Linda! I wanna get to the heart of our discussion now. And I wanna pose this next question, primarily the first part of it anyway, to Brett and Linda. Joseph, I’d like to hear your perspective on the second part of the question, and that is, “How would you define scenario planning?” This is primarily a question for Brett and Linda. And then, Joseph, I’d love to hear your perspective on “How might technology automation, as well as artificial intelligence and machine learning, be used to improve current processes?” So, again, starting with Brett and then Linda regarding scenario planning, and then Joseph regarding automation, as well as artificial intelligence and machine learning.

[Brett Turner] Yeah, thanks, Joe! So, yeah, just having done a lot of forecasting and, you know, like a lot of you, Excel being the way to do that, it’s… My experience has always been, you know, always having done scenario plans, but then you’re sort of limited to how many you can do typically. And so if you look at some sort of high-low on revenue or different aspects to it, you need build out some of those, it’s in your plan, but you’re limited to some degree because you know how hard it is to kind of crank out some of those pieces within Excel as your system of record, essentially. And so I think that’s one of the big things, especially now, when you look at all the heightened risks and all the uncertainty and what’s around the bend and who would have ever… You know, the COVID, obviously, caught all of us flat-footed. Wouldn’t it have been nice to have some other scenarios modeled out? It’s always a matter of… It’s coming down to managing risk and being prepared. And you wanna be able to have that roadmap, so you can understand because if things start to come up that are really similar, you’ve already thought through how you’re gonna respond, and you understand some of those contingencies, and a big part, obviously, with liquidity, making sure you have the right liquidity. You understand where those bounds are. You’re kind of understanding, like, where the ditches and the potholes are on the road as you’re moving forward to making sure that you have an adequate line of credit or you have the right credit facility to be able to help you, if and when you hit those times. So I think that’s a big part of it. When you look at now automation and really what we’ve built with Trovata to be able to help you now, letting the machine now start to advance that and break through that ceiling, and that finite of two or three different scenarios, letting you have an infinite number of scenarios essentially, it’ll make it really easy to build out lots of scenarios. So now you’re letting the machine do things that are just hard to do in that finite world of Excel or even just human capital to just drive all of that activity, from a planning perspective. And letting the machine really understand and flush out all these issues and help you do that at scale makes a huge difference and really helps you kind of fill out that whole risk playbook and roadmap that you know that you want. And I think that’s a… You know, just really quickly, too, in terms of what we focus on, too, it’s… We focus on companies that are 5 million or 50 million in revenue to 5 billion in revenue. And so that’s the other big advantage. Now, we can able to democratize a lot of these capabilities to a lot of mid-market and even smaller companies that wouldn’t normally have this capability and make it very affordable for them to do these kinds of things, too, and so… And then, the other aspect is that, for a lot of those companies, you don’t… Again, we talked about it, you don’t have IT resources. And so we’re able to be your IT partner. We’re able to be your data and automation partner and step into that scenario. So that way, you can focus on doing what you do and not let that to block you of not getting those resources internally to be able to do that. And now, if you’re a much larger company, we released this product called The Treasury Cloud. It’s really more focused on IT as well. We’re having a very robust back-end developer portal with lots of APIs to access the data directly. And so that’s actually becoming a lot, as Joseph was talking about, a lot of the centralization and use of the data and really unlocking it and even going beyond some of those aspects. So, I think, when you look at the big part of how we can do that, we’re connected directly to banks with APIs, and we’re pioneering those direct connections, those deep integrations with banks to be able to drive all that data in real time. It’s really breaking the whole paradigm. It’s sort of moving from the analog telephone to 5G or fiber in terms of that transport of data from the banks into a cloud-based solution like Trovata. So that whole aspect is allowing us to do these kinds of things very quickly, nimbly, and have the built-in IT sort of at your back to help you drive. And that means really covering all these other areas of risk that you may not have had otherwise the ability to do.

[Joe Fleischer] Thank you, Brett! Linda, I’d love to hear your perspective on scenario planning. And then, Joseph, I’d love to hear your perspective on automation, as well as artificial intelligence and the subset of artificial intelligence known as machine learning. I wanna move it over to you, Linda, regarding scenario planning and then Joseph regarding technology.

[Linda Copeland] Thanks, Joe! And Brett really highlighted a lot of what I was gonna say and did a great job at doing that. But what I will say is that, as it was mentioned earlier, COVID really highlighted how important it is to have an organization and the tools to do scenario modeling, right? So as it was mentioned by Brett, we work with our clients, and typically what we see are Excel spreadsheets that the treasurer would have, and they were very complicated, right? So many, many sheets, macros, links, so all that kind of great stuff. And what happened with this unusual event is that it increased the volatility in the marketplace, right? And what it did is it required that finance teams be in a position to do more modeling and more scenario modeling. And what we’ve been working with our clients on is helping them leverage tools in the financial planning function that allows them to do more aggressive scenario modeling. And in the end, what’s gonna happen is that once we get beyond the limitations of Excel and provide tools or leverage tools that allow you to create a scenario, as an example, in 30 seconds, that has all the financial logic already built into it and all the data associated with a base-case scenario, and then you can model off that automatically, that’s really where we’re seeing clients move towards, right? Everything we’re talking about now is a movement toward the leveraging of technology so that we can be more nimble, as it was mentioned by Joseph earlier, in our financial planning. And so we’ve been spending a lot of time with that on our clients. And the tools that are in place now allow you to do the kind of 20 scenarios to understand what the revenue curve is gonna be like in Q4 or Q1 or Q3 because, in the end, when COVID hit, no one understood that revenue curve, and that volatility forced our clients to adopt technology that we have them implement, so… Thanks, Joe!

[Joe Fleischer] Well, thank you very much, Linda and Brett! And then Joseph, I wanna hear your perspective regarding automation, artificial intelligence and, once again, the subset of artificial intelligence known as machine learning.

[Joseph Drambarean] Yeah, well, interestingly enough, when thinking through the problem of modeling and forecasting in this volatile time, it does begin with “Can you guarantee that the data that you are using to model is, first of all, clean and accurate and that you can guarantee that you get it often?” And “often” is starting to become a very blurry term because, in the past, “often” might have meant “Maybe, I can get it every day on a prior day basis,” or “Maybe, I can get it every week.” But “often” now is starting to become “Can I get it every hour? Can I get it every second?” And as banks have continued to evolve and drive more and more value through their technology platforms, providing, in some cases, real-time banking, providing insight into intraday on a real-time basis, or whether it be your balance positions on a real-time basis. What it’s forced is a technology stack that needs to be built for real time and also built for compute and analytics that can be done at scale. And obviously, that can’t be done on your local laptop, unless, of course, you have a crazy rig that is used for modeling hurricanes. And so the only way to address that is to have cloud computing and to have access to a computing environment that can first of all process that amount of data in real time and, then secondly, model it out given the requirements and levers that you’re pulling. And that is really the crux of automation in this space: it’s can you use that linear sequence of events in a way that takes advantage of each link in the chain, whether it’s the data coming in from the bank on a real-time basis, whether it’s the standardizing of that data in a real-time format, or the processing of that data for multiple scenarios on a real-time basis. Now, the transition between that, which is a common technological problem that can be solved using today’s tools, the transition from that to using hybrid, if not fully autonomous, models that can be driven by an algorithm, that’s where the jump comes in. And that’s where… You know, we’ve been pioneering that at Trovata and using artificial intelligence to drive insights in forecasting that just cannot be accomplished through traditional means. And it’s really built on the foundation of performance and proprietary models that take advantage of various formats of forecasting, whether it’s from conservative viewpoints or more aggressive viewpoints, and then allowing for the models to grow and evolve, based on the DNA of your data. And that’s another key factor. It’s that, in the past, when modeling was applied to big datasets, it was kind of a one-and-done situation, where you would define your model and you would then use that model on a go-forward basis. But models evolve, given different factors that are existing in the marketplace, different factors that are true of your company, different factors that are true of the operator. And if you don’t take that into consideration, you actually miss out on a dramatic amount of insights. And so really, the bleeding edge of machine learning in the forecasting space is around the mutation of models and being able to guarantee that run over run, over run as new data comes in and that data has been cleansed, and it’s being processed, can your model react to inconsistencies and anomalies in that data, learned over time, and then start to predict more reliably given input that you put into it as a user? And that’s really where the future of forecasting lies. It’s, “Can the human work together with artificial intelligence to continue to inform and craft and evolve a model that is able to process vastly more information than the human could and do it in a way that respects the insights that the human is driving?” And we firmly believe in that future, and we’ve been advocating for it and innovating within that space here at Trovata.

[Joe Fleischer] Thank you very much, Joseph! I appreciate the detailed explanation of the role of machine learning in particular. What we’re gonna do now is pose our second of three polling questions. Our second polling question ask attendees to indicate “the extent to which you agree with the statement that your finance team has the resources and capabilities your company requires to forecast cash flow based on multiple scenarios.” What you’re welcome to do is select the radio button that corresponds to your answer and then click on the “Submit” button so that we are able to record your response from top to bottom. The choices are “Strongly agree”, “Agree”, “Disagree”, “Strongly disagree”, or “I don’t know”. And after you select the radio button that corresponds to your answer, we do ask that you click on the “Submit” button so that we are able to record the extent to which you either agree or perhaps disagree with the statement that your finance team has the resources and capabilities that your company requires to forecast cash flow based on multiple scenarios, applying the approach to scenario planning that we have described: rather than planning for one scenario, planning for multiple scenarios. It’s certainly applicable now. What we’ll do is give attendees just a few more seconds to respond, and then we will briefly reveal and summarize how you have responded, and then we will continue our discussion. And I would note that what we can see from among respondents now to our second of three polling questions is the following. I think what’s readily apparent is that a very slim majority of respondents perceive that their finance teams have the resources and capabilities their companies require to forecast cash flow based on multiple scenarios. It is indeed about 40%, so 2 out of 5 disagree with that assessment of their companies. Percentage, under 10% don’t know. About 7% don’t know, but I would highlight that 2 out of 5 respondents disagree with the statement that their finance teams have resources and capabilities their companies require to forecast cash flow based on multiple scenarios. With that in mind, we wanna pose our next question, starting with Linda and then Brett, and then Joseph, asking “How you would recommend finance leaders identify 1) which scenarios to plan for, and also considerations that finance leaders… So the first question we have, “In what ways can finance leaders identify which scenarios to plan for?” And number two, “What are the most important considerations finance leaders should keep in mind when forecasting cash flow based on multiple scenarios?” I’d love to hear your thoughts starting with Linda, then Brett, then Joseph.

[Linda Copeland] Well, it’s like that. Starting off, it’s been an unusual year from a cash flow forecasting perspective. This black swan event, like COVID, is a very unusual thing. And it’s been difficult for our clients. So the volatility that we’ve seen across all our clients as it relates to the effects of COVID is something we haven’t experienced in the past, right? And what we see our clients focusing on is really understanding what the resulting revenue curve will look like as a result of the impacts of COVID. So we have a lot of our clients focusing on the revenue, predicting revenue basically, which is a very difficult thing at the moment. But I would like to highlight a couple of things that Joseph said earlier. And some of the technology that’s coming out will help identify which scenarios are appropriate, right? So we see our clients leveraging technologies that we implement to identify those characteristics of, let’s say, accounts receivable: how a client pays the bills, basically, in a much different way, right? So we’re seeing that with the use of advanced technology, we’re seeing patterns that we would not be able to see, a human wouldn’t necessarily be able to see of characteristics of customers: those customers that are constantly paying late, right? Or those customers that are buying a smaller volume of products, right? So what we see is that the use of this technology allows us to uncover a lot of these characteristics that wouldn’t necessarily be able to be uncovered. And so with that information, leveraging that information, you’re in a better position to pick which scenario you’re going to use from a cash flow perspective. Back to you, Joe!

[Joe Fleischer] Thank you very much, Linda! Brett, I’d love to hear your perspective.

[Brett Turner] Yes. So if you think of it… As I was mentioning before, in terms of just having to go with maybe two or three different scenarios and how technology is, you know, like it tries really breaking the barriers, so they’re kinda not limiting you. I mean, you kinda might… I remember one thing, and part of it is just knowing your business. I mean, one of the things that, at this company, I mentioned World Wide Packets as an example, you know, we had a lack of visibility into our sales or our sales forecast. And so, therefore, we were having to do a lot of risk buys. We had offshore manufacturing. We gotta have a lot of heavy logistics issues. And so clearly, there’s lots of risk throughout that quote-to-cash process, especially on the fulfillment side. And so it’d be one of these things where depending on if everything works flawlessly from a fulfillment standpoint, you know, then if we’re shipping, then we can build, then we can collect. But if there’s a bottleneck there and something goes wrong and we can’t ship, we can’t ship, we can’t recognize revenue. This is a hardware and software product, and therefore, we can’t bill. And therefore, we can’t collect. If we can’t collect, that is gonna have a pretty significant impact on cash, cash burn, liquidity, etc. And so, knowing your business, as I mentioned, is just so critical. It’s just an example of that. But I think the other thing, too, that’s helping to further unlock this is that… And I’ll maybe have… Joseph can kinda explain this a little better what this means from a technology perspective, but we have this notion of no penalty. So you can either focus on like, “How do you do more scenarios? How do you really, again, build out that risk coverage?” So you can now do more scenarios using tech to do that, but also, what about things you might not even know a scenario to really hone on and then sort of test the bounds of it. That’s the other thing, too. It’s getting help from the machine to help you see certain things that maybe you didn’t see otherwise. And then, you identify something. And then, you can start to dig deeper, double-click, and start to build out some scenarios around that thing. So this whole notion of no penalty is just discovering a lot of aspects. I think that’s the other thing that’s just today, especially with Excel, using Excel… You just don’t have that luxury to be able to do that. But when you can see your data, manage your data, you will have a natural language search to be able to access data in a way that you haven’t had access before. It’s like finding something on the Internet using Google. That’s what you can do now within Trovata in terms of… And because we’re tied in with the banks, we’re getting all your bank data very quickly. So to have that sort of aspect of discovery, it really breaks it wide open in terms of the things that you can do. And that’s why I think even scenario planning is taking a whole nother slant where you don’t have to just focus on like “I get to pick the two primary risks and then build out a couple of high low scenarios on that.” Now, you can not only do that, more of that, but you can also enter in this discovery mode, and then even fill out more of that playbook about what some of those risk aspects are.

[Joe Fleischer] Thank you very much, Brett! Joseph, I’d love to hear your thoughts regarding scenario planning. And then also, we’ll wanna shortly hear your thoughts about artificial intelligence and machine learning. But first, let’s hear your thoughts about scenario planning, Joseph.

[Joseph Drambarean] Yeah, I love the fact that Brett brought up, I think, one of our favorite terms at Trovata, which is the no-penalty concept, and when we refer to that, what we refer to is the human penalty. It’s the… I’m sitting at midnight on my laptop, and I have a thought that crosses my mind around “Maybe, I should take a look at our last week sales and see if there was a trend there. And maybe, if I find something, maybe that’s an assumption that I should build into this next revenue forecast that I have to deliver tomorrow.” Well, in the past, collecting that data and analyzing it might have taken if not a few hours, a few days. And so the human penalty there is… Long is not gonna do that because I think my model is good enough. And we’ll at least have a good idea. The goal behind a Trovata’s approach to modeling and kind of creating many scenarios is this idea that because a computer and a cloud is crunching all of those numbers for you, from an analytics perspective, and driving insights, you can create as many scenarios as you want. And that might explore various edge cases or niches in your forecast that could then inform a more comprehensive forecast, if you will. And I think that that is actually a game-changer. Because if you reduce the human penalty, you know the “Ann, I just don’t have the time to collect all the data and figure it all out and make sure that it’s accurate,” which is probably the biggest concern that if you’re rushing and trying to do all of that just because you had a thought pop into your mind around an area that you wanna explore, that human penalty is significant. And it actually can be very meaningful if you’re talking about a highly volatile environment, where those insights could mean the difference between making payroll and not making payroll, figuring out that “Hey, if we can spread out our invoices over the next few weeks, by just a day, it would mean this” or “If we notice that our vendors aren’t paying us on time over the last month, and we need to figure out a quick solution, maybe draw a line of credit to get us past the next few weeks,” that kind of insight, that kind of exploration, if it’s not done in a meaningfully real-time way, you miss on it. And you drive using a model that, maybe, is completely inaccurate, given the volatility. And so that’s why we really firmly believe in this no-penalty approach because it allows for the operator to be winsome, to be flexible, to explore the corner ideas that pop into their mind in a way that is rapid and in a way that gives them immediate results. And that’s just never been possible up until now.

[Joe Fleischer] Thank you very much, Joseph! I think just in the interest of time, we do wanna make sure that we pose our third polling question to attendees, and then I think we may have time to just briefly summarize our discussion regarding not only scenario planning but also artificial intelligence and machine learning. But first, we do wanna pose now our third of three polling questions. And our third of three polling questions ask attendees to indicate the timeframe, if any, in which your company intends to improve its visibility into and control over cash flow. Now, what you’re welcome to do is select the radio button that corresponds to your answer from top to bottom. The choices are “My company has already done so”, “Within fewer than six months”, “Within at least six and fewer than nine months”, “Within at least nine months from now”, “Your company ought to do so from your point of view, but does not intend to do so,” “If your company from your point of view does not need to do so”, “If you just don’t know”. You’re welcome to select any of the choices that correspond to your answer. We do want you to make sure to not only respond but also click on the “Submit” button so that we can record your response to the question within what timeframe, if any, your company intends to improve its visibility into and control over cash flow. And then, I think, Joseph will have time for just a quick question I’ll quickly pose to you regarding opportunities to extend finance teams’ capabilities by applying advances in artificial intelligence and machine learning. So I just wanna reveal, first of all, how attendees have responded. Kudos to those of you who are already seeking opportunities to improve visibility into and control over cash flow. And then, I just wanna very, very briefly turn it over to you, Joseph, to quickly answer this question about how finance teams can apply advances in artificial intelligence and machine learning to augment their existing analytical acumen when forecasting cash flow. Joseph, I’d love to hear your perspective before we wrap up.

[Joseph Drambarean] Yeah, I think that I’ll probably address this just using classic machine learning terminology because it really informs how you can take advantage of machine learning. So within the machine learning space, the notion of supervised and unsupervised training is really important then. And it drives a lot of what you see in terms of outcomes. And we see a huge opportunity within both. And I’ll start with the supervised. When you think about supervised training, what you think about is providing a model, providing an algorithm, direction, and notionality so that it can drive the insights that you would want to see, whether within a topic or a time series. And that’s an area that we see a lot of opportunity in because if there are specific cohorts of your finance operations that you suspect may have insights that could drive decision making, providing feedback, whether on a regular basis or a one-time basis, can direct that model to understand specifically what you’re looking for. What I think is the more exciting part of machine learning’s applicability to finance analysis and forecasting is really the unsupervised space because machine-learning algorithms shine most when they are allowed to explore freeform in a greenfield way because what they tend to find are anomalies and insights that a human could never figure out without having access to the same amount of data and the time to be processing that same amount of data. And what a machine does is it measures directionality across a variety of different vectors, whether it’s time, whether it’s relationships, whether it’s text, all of those can be measured statistically, and provide insights that can then drive a decision. And an insight could be, for example, discovering that there is a relationship between certain transactions that relate to invoices not being paid on time and whether or not that could be something that could be explored and modeled into a forecast. And so when I think about machine learning being leveraged within the finance space, it’s almost like taking the human and adding a robotics kind of uniform on and giving that human the ability to jump higher, punch harder, be better in every way because of the extension of that compute capacity. And that’s how we’ve been approaching that problem at Trovata.

[Joe Fleischer] Thank you so much, Joseph! I really appreciate your perspective as well as perspectives from your colleague, Brett, of course, and Linda from The Hackett Group. Before we wrap up, I once again wanna thank you, Brett and Joseph and Linda. I also wanna let you know what will follow from our webinar. So first of all, this webinar will be available on-demand later this week within the “Past Webinars” section of cfo.com. At that point, you’ll be able to view and listen to a streaming archive of our webinar, click on a link to a case study from Trovata, the sponsor of our webinar, download a PDF document comprising slides from our webinar, and if you’re eligible, if you’ve answered all three polling questions during our live webinar, retrieve your CPE certificate. Also, at the end of our webinar, we will display an online feedback survey. You will be able to view it in a separate browser window if you’ve turned off your pop-up blocker. As always, we appreciate your feedback. Once again, I want to thank Brett and Joseph of Trovata and Linda from The Hackett Group for sharing your perspective. And I want to thank attendees for joining us for our webinar “Three Ways Finance Leaders Align Cash Flow Forecasting and Strategic Planning”, brought to you on cfo.com by Industry Dive and sponsored by Trovata. We thank you for your time, and we hope you’ll enjoy the rest of your day.