Modernization Deep Dive: Reconciliation Agent

In this Modernization Deep Dive see how the Reconciliation Agent can turn reconciliation into a continuous process instead of a month-end scramble. This session demonstrates how the Agent connects with JD Edwards and financial data to match transactions, identify exceptions, recommend solutions, and even take action based on your governance rules. 

Transcript

So we'll get started so real quick, just want to tell you all about ERP Suites.

So if you haven't heard us or of us or have been on the previous episodes, I guess they're not episodes.

What I'm looking for sessions, deep dive season sessions.

We can handle all kinds of things with your JD Edwards from hosting, managed services, consulting with functional and technical consulting and then AI within JD Edwards or any of your other enterprise applications.

And we also can help out with EPM services.

So, so, yeah.

So hey, good to see you everyone.

Thanks for having lunch with me.

I told you at the last time that you were going to get a bunch of me the next couple weeks.

So you know, what we're going to cover today is, is kind of why reconciliation weights, you know, what causes the problem?

How did, how do companies get there?

Then actually go through the agent and and look at what it's doing for you and then talk about what next, right.

So this is going to be very similar to the AP agent, if you saw that one.

And this is kind of that next agent in our financial digital worker family.

So here's me.

I had to have the big picture of myself on this one, right?

To kind of scare you away.

But again, 34 years JD Edwards.

Actually I got to update that because it's a little bit more work for JD Edwards on through the Oracle acquisition and now getting to spend my time with the RP Sweets and loving it.

So, you know, what is our approach real quickly to kind of our ages, right?

And we really have a three-step approach.

You know, the first one is to assess your needs, right?

You know, clearly we're going to talk about reconciliation agent today, but you know, quite often we'll we'll talk to somebody who thinks they they need AP or they need something else.

And as we go through this, we help them kind of prioritize their digital workers and, and their needs and really try to help companies get the most value out of this out of their agents and AI.

And then we automate, right?

Then we, you know, the automation is, is how do we deploy those agents for you, right?

How can they they execute your workflows and leverage orchestrations and how do we build your governance into the model?

But the third part is, is, is we optimize, right?

And it kind of makes us different from a lot of other companies, right?

We're not just giving you a digital worker and walking away, right?

What we're doing is, you know, we're working with you with that digital worker to help them improve, right?

To help them get better, to help them manage changes in your organization, right?

So we'll help you, we'll give you that, we'll give you the dashboards, we'll show you how they're approve, how they're providing value, their returns, you know what they're doing.

But then we'll also give you recommendations of how we can make those better for your organizations.

But you know, agents as digital workers, right?

You know, this is the part of the presentation where I stopped calling it an agent and start calling a digital worker.

And, and, and the reason is because they are like your workers, right?

You know, I was talking to one of our other marketing folks the other day and we were talking about a podcast and it's like, I look forward to the day where digital workers are on people's org charts, right?

Because they're actually doing the work.

The difference is they're always on, right?

They never take a day off.

There are 24/7, right?

They're completely goal oriented that, you know, they're they're trying to solve for that.

They continue to learn and evolve.

They work together and they can scale amazingly, right.

So, you know, that's how digital workers are different.

So this is a we're going to stage it, you know, on on what does this reconciliation?

What is it trying to do right?

And really kind of the key to it is, is, you know, we can look at it two ways, right?

The traditional manual reconciliation, you know, done by hand and Excel spreadsheets and reports and brute force or an agent that, or a reconciliation that was helped by an agent.

Wow, I've already broke my own rule.

A digital worker, right There was the same problem, right?

A $63,000 break, same company, same people involved, right, Doing it by hand.

That problem was found in day 35, right, five days after the close right?

Because it was found somewhere somehow, whereas our digital worker found it in day 16, right?

14 days before the month end close, right, provided solutions for that reconciliation problem and allowed companies to close, you know, you know, right on time without having to kind of manage to this month and close process, right.

And, you know, companies get there, right?

You know, I mean, and, you know, I've worked with so many companies over the years.

We're reconciliation.

It's important.

Everybody understands.

It's part of the, it's part of the financial business process, right?

But ultimately over time, we've seen reconciliations go from weekly to monthly, right?

It just happens, right?

They're like, Oh yeah, I can't, I can't do that.

I don't have time for this.

We're in our budgetary cycle and I've got to do this.

And we've got an inventory valuation that we have to do and all this kind of stuff.

And it's like so that that process of reconciliation that was supposed to be managed kind of on a weekly basis.

So that month end close was a non issue becomes a month end close issue, right?

And who pays for it?

It's that it's those extra days when you're trying to close the books that you're also trying to do these reconciliations, right?

And problems just sit there until they're found, right?

And and that's the problem, right?

So none of it was a hard problem problem, which is why wait, that's the problem.

So let's go through the solution today.

So I'm going to pause the presentation and I want to get into our agent dashboard.

So if you guys are on for the AP agent, you saw, you know, you're, you're not unfamiliar with the way this looks.

We start, we typically start out with kind of a KPI dashboard about what's going on.

The cool part about this is that, you know, this is telling you right up here that we're connected to JD Edwards in real time, but we're also connected to your bank feed.

It tells you what period we're working on.

Right now we're working on August, excuse me, April 2026 and that so far during that time frame, we've analyzed over almost 16150 transactions, right?

You remember I talked about digital workers being goal oriented, right?

So you know, some of the goals are, you know, how do I, what is my match rate and what is my auto match rate, right?

You know, so I'm, I've matched 99% of the accounts, right?

I've auto matched and resolved the problems on 90, almost 94% of those, right?

You know, it's looking at, I've already reconciled over almost $32 million of accounts right now.

And I have open exceptions and you know, days to reconcile or, you know, or, you know, 2.4 down.

So it's giving me a warning there.

But these are the goals that that reconciliation is trying to solve for, right?

They're the ones that they're trying to improve constantly, right?

So you can see KPI's around match rate and whatnot, and then you can actually see some of the reconciliations, whether it's your bank or your AP or your AR, your payment processing or inter companies, how many have been reconciled, how many have been, how many exceptions there are?

What's the value and what's the status of them, right?

So every one of these, right, you know, every one of these, you know, are real important information.

And if you look down and say, God, we got 62,000,000, you know, $62,000 problem here, you know, if I click on it, it actually shows me those accounts, right?

That these are the problems that make up that $62,000, right?

These are just numbers, right?

They're doors and insights into your, into your business processes, into your problems, right?

So now you can see that, yeah, I've got a variance around my operating, my cash account, my AP trade account, some my inventory account, some unapplied cash, you know, things like that.

These are the things that we need to work on, right.

All right, let's talk about the reconciliations, you know, upfront.

So here we again, you know, we're connected, but we have, you know, a bank reconciliation apar payment processing and intercompany.

AR and intercompany look good, right?

You can see these little numbers next to them.

There's no issues.

It it hasn't found any issues that they're clear.

That's a good thing, right?

People don't have to waste time trying to manage to something that is that is tying out that don't that doesn't need reconciliation, right?

But you know, so we've gone through that.

So if if we go down to let me see where is it, we can actually pull in here and you can see the transactions from the bank detail, right?

So it actually pulls in that detail file, pulls it in and you can see all of them.

You can see the beginning balance and the activities and you can see, you know, the different kind of transactions that have gone on with them and the dates.

And you can see these four down here are are, you know, this one here maybe is misclassified.

So from a reconciliation perspective, we can give you suggestions of how to fix that account.

But these four down here are missing in the GL, right?

It's posted to the bank, but it hasn't cleared yet, right?

So you could actually see, you know that so, so the, so you can look here and you can actually see the exceptions in this set that, you know, hey, we're missing this, you know, this, this $87150, we're missing that, right?

It, it looks like an office lease payment, right?

And it can even then go in and propose a journal entry and we're going to get to that in a second.

So it, it's, it's not just giving you an more exceptions to deal with, it's actually trying to figure out what the problems are and help you solve for those.

OK, if I go back up to the top, I'm going to hide those and go to the exceptions.

You can actually see a little bit more detail behind it, right?

Here's that same kind of exception that we had 8700 comp post through the bank.

But for some reason the GL, the GL side of it hadn't posted yet.

The system can look at it and say, you know, what do you you know, what should I do here, right?

And it can actually learn, right?

It can look back on previous months and say that yeah, this $8700 you know there's a problem we need to fix here, right?

I've seen that that has posted as a office lease payment for 11 consecutive months, but somehow the GL always flags the, the bank account.

So let's create a journal entry that supports that from the bank, you know, and I'm not smart enough to figure out the debits or credits, but debit in the bank account and crediting that, that that office lease account, right?

So we, we can actually go in there and, and look at that as well, you know, and the other cool part about this is that you can see over here, there's each one of these transactions, each one of these exceptions that the agent will give you a confidence score, right?

It says, I've seen this same problem 11 times.

I am supremely, I am 94% confident that that I know what this problem is and I can solve for this problem, right.

If you go down a little bit further, I thought I saw one down here that was that was a little worse.

Oh yeah, there's one down here that there's, this was a possible chargeback or a processing fee that it's not quite understanding what's going on.

Maybe it didn't have anything there, but I've only got a 68% confidence score in this one.

So that's not something that what's going to recommend, it's not going to create a journal entry for you, but it'll allow you to start investigating that issue, right.

But, but if I do go to the journal entries, you know, based upon, you know, your governance and your rules, it can actually go out and say, hey, you remember that $87150 account there that, you know, for the, for the lease here, here's the journal entry that will fix it.

I've looked back and I've seen that this same journal entry has been posted every single month.

You know, we should fix this.

But, but the rent account is 626240.

And if we debit that and credit our cash, we should be in good shape.

Do you want to approve and post this to JD Edwards?

Awesome, right?

You could do that right here.

It's going to reach back and create that journal entry for you and JD Edwards, right?

We can go in and look at at, you know, the wire transfer stuff.

You know, you can see one down here that was rejected.

You know, it was rejected by Jim Smith confirmed with AP that that this it thought it was a problem, but it was actually coded correctly.

So we want to keep it here, right?

So it can tell you.

And this one, for instance, the second one is another interesting one because we created a rule that said, hey, if it's under this amount then or if it's for this type of payroll processing fee, automatically post that to JD Edwards.

And again, no one's suggesting that, you know, you let these things automatically do stuff without your, you know, complete confidence in that.

This is just an example of how you can create rules your governance model that allows the reconciliation to perform work for you.

All these first ones were proposing journal entries right and and pending approvals.

But this one for payroll processing fees, we've said no.

We had we had 97% confidence that this was right.

I'm going to, I want to let that one do that every single month.

If I then go, I'm going to go back real quick and say, you know, the proof in the pudding is back here, right?

Is, you know, it's going to look at, you know, by company, what are the accounts, right?

What are your accounts and what are the ones that are out of balance, right?

It's looking at 148 different accounts for you.

I mean, it could be thousands, right?

It's telling you that 147 of those are in balance.

You know, 7 are out.

You know, there's a gross variance of OF90K with a net of 6 or 62 to and we've compared almost, you know, 187,000 records and it can actually tell you, you know, what's going on by account by account basis.

We can actually then follow this through and go down to certifications and this is kind of how it connects to kind of your monthly close, right?

You know, this is this reconciliations, you know, not working at the end of the month.

It's working every single day of the month, but it's constantly looking forward for you and saying how prepared are you to close that company, right?

And it can actually go out and certify or close or not.

You can see ones that are pending approval.

You know, this one's, this AP trade account is, is 2000 over.

Somebody changed the, you know, the posting edit code and put it, put a manual entry in there.

I could actually open the package for that account.

You can see the account itself.

You can see any kind of materiality threshold and you can see what's going on with it, right?

So you can actually go out and, and it will create supporting evidence for you.

You can assign a reconciliation isolation to certain people.

So we can go through that.

So it is prepared by somebody, reviewed by somebody, approved by somebody, right?

So it's, it's pretty awesome how it kind of goes through this and looks at this by account by account basis, right.

And I could actually then obviously go back to the audit, right and say, you know, you know, at the end of the month is building itself, right?

It's going you can see the ledger's, you can see the company bank statements, you can see all the settlement statements, you can see how they've been analyzed.

And it'll actually create reports for you, the summary report, the detailed report, the proposed journal reports for you, so that your auditors that you have that package saved every single month for your, your audit team.

So I'm going to go back to the beginning.

You know that the, the, the point here is that this isn't happening on the 30th of the month.

This is happening every single day of the month, right?

And as we go through that process, this agents going to work towards that end of the month so that reconciliations are happening all the time, helping you through this process.

OK, I'm going to go back to my slideshow real quickly.

So what happened, right?

We, you know, we match 27 of 31 bank lines.

We identified a $63,500 break.

We were able to auto match, you know, almost 94%.

So we reconciled it without anybody having to do that work, right?

Seven accounts out of balance.

You know, we, we were able to create automatic postings verse manual and the agent actually was able to post 1 based upon your governance rules, right?

So as we get through that, you know, the the point here is around, you know, and you've heard me talk about it, particularly in the AP agent side that you know, that that we built a digital family, right?

I mean it's not just it's not just a reconciliation.

We have a monthly close agent, we have a financial planning agent.

And if you guys were on my call, obviously we have an AP agent, right?

All of these working together.

The key here is that when you start the journey with to AI for JD Edwards with us.

We help you build that foundation, right?

The JD Edwards agentic platform, your control tower, your, your agentic dashboards that you want, and then every other agent that comes along after to be able to help support your business just plugs into that, right?

So every agent after your first one reuses the same agentic platform.

So the time to value shrinks and more importantly, the cost shrinks, right?

We know we've identified that second and 3rd 4th agents, you know, whether they're in the same family or not, are generally up to 60 to 70% less expensive than the first one because they all work upon that same AI foundation and that AI agentic platform for JD Edwards.

You know our digital workforce is, is growing day by day, right?

You've already seen the AP agent.

I showed you that before.

Today's the reconciliation agent.

After in Focus, I'll show you the monthly close agent.

But we also have financial planning, sales order agent, We have an inventory allocation agent.

We've even set up some very industry specific agents for the home building community that focus on specific problems there.

But we're building out our portfolio every single day.

So what goes next?

You know, we'd love to work with you.

We'd love to talk to you about design.

We'd love to talk to you about how to do a process assessment for you and then, you know, ultimately be able to take, you know, whatever financial agent that's going to add the most value for you, whether it's reconciliation or monthly close or the AP agent to be able to help your organization and build from there.

So Scott, I'm going to turn it back over to you and, and you can talk about what some of the next ones are and and then how we how we finish up.

Yep, I like how you highlight the the one in two weeks with the closing agent because that's me, right?

You know what I mean?

You know, that's the one I got to shamelessly plug myself, right?

Yeah.

So thanks for joining us so far.

Have you been able to attend a few?

We've got another one next week about the benefits of connecting your ERP and CRM.

And then the week after that is the trial balance or or close agent.

So that'll well, I'm sorry, that's not the next week, the week after, the next one week after.

OK, so we have a little break and then on the 29th Kevin's back for the trial balance agent 1.

And even if you're not, you know, in the home builder industry, I'll be doing a a presentation at in focus on Monday about some of those industry specific agents.

So I think, I think it would even be useful for anybody or any industry to kind of see that because you know, the idea is we can work with you to build purpose built verticalized agents that really support whatever industry that you're in.

Right.

And that's really kind of the the message and the goal around that presentation.

Yeah, yeah.

So if you can join us next Tuesday and then if you or anybody from your team is it in focus, come see us, we can talk through more of this.

Does anybody have any questions?

I haven't seen any yet.

So just got to thank you.

So other than that, I'll give it a few minutes and see if anybody posts anything.

Yeah, we'll be doing demos, add in focus, talking through.

Like Kevin said, he's got a presentation.

We've got a few other presentations.

You can all find that on our website where we're going to be at our party on Monday night.

All right, well, all right.

You're off the hook for questions.

I don't see any coming through.

Well, I don't know how to feel about that.

Right.

Good.

You know, did I do such a good job that you just answered everything or nobody wants to say anything?

But you know, I, I, you know, you can always reach out to me or Scott, You know, our, our information has been on here has been here on the presentation.

You'll get that if you want to reach out to us directly afterwards, we're here as well.

Love to chat with you about this.

You know, obviously me being AI did AI did a user group presentation two weeks ago and, and from the first time I was a customer of JD Edwards.

I'm at 36 years right now and this is one of the most exciting times that I can remember, right?

I mean, this is like the move from world to our one world and, and, and, and the advent of orchestrations.

But, you know, AI for JD Edwards is just so incredibly exciting.

And I can't wait to chat with all of you guys about that.

All right, thanks, Kevin.

See you all next week.

All right, bye, everyone.

Bye.