(0:00) all right, well, I think we'll go ahead and get started. so thank you everybody who joined us today, welcome to the chat. today's webinar we're going to discuss AI automation and deep learning in finance and see how this plays out in real-world AI applications and finally, we'll conclude with questions. if you have questions along the way, just please go ahead and put them in the Q&A window at any time, and we'll address those at the end. so to start us off, let's just go ahead and introduce ourselves. AJ, would you like to go first?

(0:30) yes, Lee, thank you, good morning everybody. thank you for joining this webinar. it's another AI webinar, hopefully we add value and we create something that you can take back and apply to your businesses. so yep, I'm AJ, I'm the co-founder and CEO of ezyCollect. I'm dialing in today from sunny Sydney, so it's early morning here. my background is I started ezyCollect because I ran a wholesale and manufacturing business myself before starting ezyCollect. and in that previous business, I was very much growing a business and experiencing the pain of not getting paid on time. I hired lots of people to solve the problem, I did a lot of Band-Aid solutions like finance, and I really kind of struggled to get paid on time. and when I sold that business, it got me thinking, how do we fix this problem for other businesses? ezyCollect is really a labor of love and it's focused on helping businesses to get paid on time. and talking today is really about sharing that journey over the last 10 years, how that landscape has changed, and in a very practical sense, how we can apply some of the more recent changes to getting paid faster. Lee, I guess you're the other side of the coin, did you want to talk about the accounts payable side of things?

(2:01) yeah, absolutely. I appreciate it, thanks everybody for joining us. I'm Lee Jacobs, I am the head of product at SquareWorks. SquareWorks makes the mirror image of what ezyCollect handles, and so we're on the accounts payable side. my personal background has been in the payments and finance space, so I've led product teams both on the consumer payments side as well as now into the depths of the accounts payable B2B payments, in the broader AP automation suite. SquareWorks got started in 2016, we actually started off in a consultancy manner, working with finance and accounting teams implementing NetSuite, doing implementations and optimizations, and we saw similar pain points around the payables process from invoice capture to the actual completion of payments. and as we were building out our book of clients for implementations, we began developing widgets, and those widgets turned into products, and now SquareWorks is an end-to-end AP automation solution. so I joined the company just about a year ago and really focused on where we take our strategic development when it comes to the end-to-end AP automation solution.

(3:26) I think everyone would agree that AI is making a huge impact in all of our daily lives, both professional and personal, and we'd like to explain a little bit about the history of AI as it pertains to accounts payable and accounts receivable. so AJ, can you tell us a little about the history of data generation as it pertains to AR teams before AI, the present where we are today, and then some of the ideas of what the future may hold for us?

(3:58) yeah, I think just for everyone's benefit, I want to talk very practically with lots of examples because AI is such a commonly used term, and I don't want to just talk about AI that you can hear in the news or read. it's really going deeper into the domain of accounts receivable and accounts payable. and for me, my personal experience, I'm not an AI expert, but I understand and can see how technology is changing. the biggest emerging trend we're seeing is the insights that businesses can get from the raw data in your AR or ERP systems. it's moving from a reactive to a proactive collection state, and there's huge amounts of data being generated by your ERP or accounts receivable system when you chase a customer, when someone pays you. in the past or even now, it's been harder to take that data and surface it for you as a business, but what we're seeing with the emergence of language models, a lot of businesses are adapting their user interfaces to query large amounts of data and surface that in a very proactive way.

(5:14) so from an ezyCollect or AR context, it's not just telling you that you've got a late debtor, and they're 30 days late, and you’ve got this aging report, but now you're able to get insight into not just that someone's late, but why are they late? what's the root cause of their behavior? and then actually giving you some tips on what you can do about it. so I think we're moving in terms of this timeline from just chasing a customer who's late and sending them an email or text to going deeper beyond that and driving better conversations with your customer about why they're late. and a practical example of how we apply that at least from an ezyCollect context — and this is not limited to ezyCollect — is we try and look at segmenting customers based on how late they’re paying you. so we have that data that, hey, Lee is paying you 25 days late, but now what we can do using language models and tools available to synthesize data more effectively is we can say, “Hey, Lee's paying me late as the business,” but let's do more analysis on Lee. for example, we can see that Lee is paying other suppliers on time. we can quickly query this now in a more cost-effective, efficient way. we can use data sources from third parties like credit reporting bureaus or trade bureaus. we can start to query publicly available information through Google and Reddit. and then we can start saying, “Hey, you know what, Lee is actually a relationship issue customer because he's paying us late but paying the rest of the market on time.” right? and if I know that, I see you're smiling because I can see that you're not a bad payer. you have the money but you just pay me late. why is that? when I call you to chase that payment, I can say, “Hey, Lee, I see you're paying us late but you have a really good credit rating, what's the reason?” so you're going deeper into conversations.

(8:07) I think we're in this emerging phase where we can get more predictive on the data. and in the future, which I think is still a fair way away, it's really about saying, can we hand that off to a bot? but I have a different view on bots — we'll discuss that in the next section. so that's what we're seeing at a high level with a real practical example we're actually implementing in our system today. Lee, are you seeing this? I'd love to hear your thoughts from an AP point of view.

(8:18) yeah, and I'll try to stick with the example too, with Lee being the accounts payable team over here paying AJ on the receivable side. I think that's actually a really good example, and I also like to keep things more tangible. so on the AP side, let's say, you know, I am the AP team for a company, and we tend to pay all of our other APs on time but pay you late. well, turns out, the reality is, you're sending your invoice to a department head rather than directly to our AP team. and really within the AP organization, their main KPI is just being the percentage of vendors they're paying on time but not a whole lot of ability to apply business intelligence or artificial intelligence on top of that data. and so the evolution you talked about, from being reactive to proactive, is the same exact thing we're focusing on in AP. for most AP teams, at least in the United States, the process hasn't changed much in the last 30 years. the biggest evolution in accounts payable has really just been the movement from paper invoices to PDFs. paper invoices now are PDFs coming in, being copied and pasted into an ERP, then approvals go out, batches of payments are approved, and checks are printed. but what's cool with AI on the AP side is now we can take more of a proactive approach, ensuring not only that we're paying on time but optimizing our cash flow, paying attention to more impactful KPIs like invoice cycle times and days payable outstanding. so we can focus on cash flow, which matches the flip side with receivables as well.

(10:19) when we talk about AI-enabled automation, we're really talking about making the AP job a little more tolerable, taking bad jobs and making them easier. but what we see emerging is not just the extraction of data and intelligent mapping of data but more predictive and descriptive analytics to apply artificial intelligence to map that data to the correct fields based on historical use cases in the ERP. and as we look into the future, agentic AI is about replacing a task entirely as part of the workflow so that humans can focus on more strategic aspects of their job while machines perform tasks like analyzing historical data and predicting job performance. one interesting thing with agentic AI is how it can be either a standalone solution or more impactful when embedded in existing software user experiences. this is one of the cool things about ezyCollect or SquareWorks being native NetSuite SuiteApps, as the experience is embedded in the system you're already using, revolutionizing that workflow instead of adding another piece into your tech stack.

(11:58) what do you think, AJ? what's the future for agents and the user experience?

(12:04) yeah, it's a tough one, Lee, because agents, I think, have their place as an operational efficiency tool for a business. they can help a team internally, but I'm a bit worried about exposing agents to the end customer. so if you're an accounts payable officer, agents have a place to help you get insights, but would you let an agent communicate with your customer? I don't think it's quite there yet, and I don't know if you want it to be in the next 10 years either. we'll talk about that in the next section.

(12:48) absolutely. you know, I think it's a good segue to our next section. first, to our audience, we want to get your feedback with our first poll question: what is your biggest challenge with your AR or AP processes?

(13:12) 10 more seconds before we close the poll.

(13:29) looks like 33% said managing overdue payments on the AR side, and 67% said processing invoices manually on the AP side. no surprises there. I think the manual nature, like I mentioned on the AP side, where things haven't changed for most businesses in the last 30 years, makes the job tedious and time-consuming, something machines could do more efficiently.

(14:15) let's jump into some real-world examples of how AI is revolutionizing business workflows. AJ, can you give us some examples of how AI has changed the business workflow for finance professionals?

(14:31) I think it's important to give you practical examples. I don't want to talk too theoretically. we're seeing two big trends in AR workflows: hyper-personalization and human-in-the-loop automation.

(15:06) hyper-personalization is driven by the data stored in your systems and the ability to expose it through user-friendly interfaces. for example, ezyCollect captures a lot of data on how a debtor pays. we know when Lee pays, what payment method he uses, and more. AI can infer patterns, like Lee paying at night because he has kids, allowing us to personalize workflows, such as sending emails at the right time or offering discounts based on payment habits. we can automate these tasks and improve collection rates by personalizing the experience.

(17:11) the other trend is human-in-the-loop automation. collecting money isn't just about chasing customers, it's about building a relationship. even in the future, when bots are prevalent, the businesses that succeed will keep humans involved in the loop, scheduling human calls with debtors for a more personal touch.

(20:14) as businesses scale, manual processes become rigid and inefficient. at SquareWorks, we're working towards touchless workflows, where AI automates tasks like invoice processing without removing humans from the loop. it enables exception management and strategic vendor relationships.

(24:05) the poll shows an even split between challenges in chasing overdue payments in AR and manually entering invoices in AP. AI plays a big role in both areas, helping with data extraction, intelligent mapping, and optimizing follow-up tasks.

(25:40) deep learning, a subset of machine learning, is revolutionizing AR by providing actionable insights and predicting cash flow. it enables a more efficient process and allows account professionals to act on data without adding more work.

(29:01) deep learning also helps in AR by identifying patterns, such as a debtor who always part-pays. with this insight, we can offer payment plans more effectively.

(30:42) AI has evolved from entertainment to a tool that drives real business insights. solutions are now scalable and available to all businesses, without the need for a dedicated data science team.

(31:10) thank you all for connecting with us today. if there are any remaining questions, we can follow up via email. a participant giveaway will be announced, and thank you again for joining.