Skip to content

Personio - Reporting, Investments & Forecast Config - 2025-10-21

Metadata

  • Date: 2025-10-21
  • Company: Personio
  • External Participants: Tom Thorn (Treasury)
  • Palm Participants: Emma
  • Type: Customer Call
  • Domain Areas: Reporting, Cash forecasting, Investments & debt & IC loans, Scenario Planning

Summary

Context

Deep-dive customer call with Tom from Personio Treasury to discuss reporting automation (PDFs to Slack), investment tracking requirements, and forecast configuration options. Emma showed prototypes for self-service reporting and forecast blending capabilities.

Key Discussion Points

  • Self-service reporting with PDF export and Slack integration
  • Investment tracking: term deposits, money market funds, overnight accounts
  • Credit ratings management and counterparty risk monitoring
  • Forecast configuration: blending system vs confirmed forecasts
  • Daily granularity needs for collections forecasting
  • Maturity tracking and liquidity planning
  • Desire for assumption/driver-based modeling on forecasts

Pain Points

  • Current investment tracking in Google Sheets is manual and error-prone ("the upkeep of this sheet is quite manual")
  • Credit ratings require manual updates from multiple bank websites - each bank uses different terminology
  • AI agent for scraping credit ratings "works once and then it's not working again"
  • No automated way to track investment maturities in forecasts
  • Collections forecast is just one monthly number, need daily distribution
  • Bloomberg is expensive and not in budget
  • Currently screenshot data from G-sheets into PowerPoint for reports

Feature Requests & Needs

  • Text boxes/commentary in reports for context (explain outliers)
  • Slack channel integration for automated report publishing
  • Investment tracking with rate, maturity, and credit rating fields
  • Ability to attach credit ratings to any bank account (not just investments)
  • Forecast showing deposit maturities as future cash inflows
  • Hybrid forecast blending (conservative/aggressive options)
  • Percentage-based assumptions on forecasts (e.g., -5% on collections)
  • Single forecast entries for periodic items (tax, VAT reimbursements)
  • Alert/notification when credit ratings change

Jobs & Desired Outcomes

Job: Track and report on investments (term deposits, money market funds, overnight accounts) with associated rates, maturities, and credit ratings

Desired Outcomes: - Minimize the time required to update investment tracking data (currently manual G-sheets) - Reduce the frequency of errors in investment data entry and reporting - Increase visibility into total investment position by type, currency, and maturity date

Job: Monitor counterparty (bank) credit ratings to maintain compliance with investment policy

Desired Outcomes: - Minimize the time required to gather credit ratings from multiple bank websites - Reduce the delay between a rating change and awareness of that change - Increase confidence that credit rating data is current and accurately sourced (S&P, Fitch, Moody's)

Job: Plan liquidity by understanding when deposit maturities will return cash to available balances

Desired Outcomes: - Minimize the uncertainty around future cash availability from maturing deposits - Increase visibility into the gap between total cash and available (liquid) cash - Reduce the risk of being caught short when large maturities don't align with outflow needs

Job: Distribute monthly collections forecast across working days based on historical patterns

Desired Outcomes: - Minimize the inaccuracy of assuming equal daily collections within a month - Increase the ability to see intra-month cash flow patterns (e.g., higher collections in first two weeks) - Reduce the gap between the FP&A monthly number and treasury's daily visibility needs

Job: Apply conservative assumptions to forecasts for risk-adjusted treasury planning

Desired Outcomes: - Minimize the risk of overestimating collections or underestimating outflows - Increase the ability to model worst-case scenarios quickly (e.g., -5% on collections) - Reduce manual effort in creating multiple forecast scenarios

Domain Insights

  • Investment types: Term deposits, money market funds, overnight accounts, interest-bearing balances
  • Investment criteria: Security > Liquidity > Yield (in that order) - "for a lot of people it's flipped, people think yield is most important, but actually it should be security itself"
  • Credit ratings: S&P, Fitch, Moody's - need both short-term and long-term ratings; banks use different terminology (secured vs unsecured, etc.)
  • Money market funds: Value based on daily NAV, not fixed rate - "appreciate in value" over time
  • Collections pattern: Varies by month - January has significant uptick due to yearly customers paying
  • Deposit protection: German banks have deposit protection scheme (relevant for security assessment)
  • Investment visualization: Prefer line graph with maturing deposits shown as spikes/different color blocks in forecast

Action Items

  • [ ] Explore text box/commentary feature for reports
  • [ ] Design investment tracking with rate/maturity/credit rating fields
  • [ ] Consider credit rating attachment to all bank accounts (not just investments)
  • [ ] Explore forecast assumption/driver-based modeling
  • [ ] Follow up on daily granularity for collections forecast
  • [ ] Consider deposit maturity as forecast inflow feature

Notable Quotes

"The intuitiveness in terms of how you actually set up the report and what you can add in there looks really good." - Tom

"What you're showing me now is a million times better than what we do things at the moment." - Tom

"Security is always the most important... for a lot of people it's flipped, people think yield is most important, but actually it should be security itself." - Tom

"If we have that the cash balance, then the forecasting piece linked to each bank account... if we do have a maturity, that you actually see that cash flow showing as a forecast value there as well. That would be awesome." - Tom

Full Transcript

Meeting Info: Tue, Oct 21 2025 · 3:45 PM | 1 hr | 2 participants

Participants: tom.thorn@personio.de, emma@usepalm.com

Emma S. Here. Hello. Tom T. All right. Emma S. 0:00 Hi, don't worry. Emma S. 0:05 Now, how about now? No. Tom T. 0:07 I can hear, you know, sorry about that. You hear me? Emma S. 0:09 Yes, loud and clear. Tom T. 0:11 Good. How is it going? Emma S. 0:14 And I want to say good, but I'm coming down with something to be honest. Tom T. 0:16 Oh, I'm sorry. Emma S. 0:18 Don't worry. Tom T. 0:18 Are you not? You're like the third person I've spoken to today saying They're not not feeling 100%. I know it's this time of the year but I also know a bunch of people who kind of got covid or something recently. Like That's yeah sort of Tom T. 0:30 thing. I hope it's not too bad. Emma S. 0:33 Yeah, me too. We're traveling to Sicily on Saturday, so Tom T. 0:36 Oh yeah. Emma S. 0:38 Fingers crossed. Hey, I'm gonna record this session if yeah. Just so you're Tom T. 0:39 Yeah. Tom T. 0:41 Sure. Emma S. 0:42 aware. I'm starting it and it's take a little while but yeah. No, it's all good. Tom T. 0:44 All good. Emma S. 0:48 How are you? Tom T. 0:49 Yeah, yeah, good. Thank you. Nice. Couple of weeks off feeling nice and rested. Tom T. 0:54 It was yeah quite a busy one but then the last few days while we were away was just like pool and beach time in the sun which is nice. It feels like a long time ago. Already given different back in the Amsterdam weather but yeah very good. Thank you. Emma S. 1:10 That's nice to hear. Emma S. 1:12 Something nice. Tom T. 1:13 And you have you have to Sicily sooner? You is that holiday plans or Emma S. 1:19 Yes. Emma S. 1:21 Yes, yes, it's it's holiday plans. It's actually a celebrating my mother-in-law Tom T. 1:23 Yeah. Tom T. 1:29 Okay. Emma S. 1:30 turning Steve so we're that 13 people going, five children, three toddlers. Tom T. 1:31 Nice. Emma S. 1:36 So it's quite of an intense in that sense, but yeah, I will be cool. It's always Tom T. 1:39 Yeah, very good. Emma S. 1:44 fun to Tom T. 1:45 Yeah. Emma S. 1:46 See something else. Tom T. 1:48 Indeed. Emma S. 1:49 Yeah, for sure. Emma S. 1:52 Cool. Thank you for taking time. Emma S. 1:55 And glad you had a nice time off. Emma S. 1:58 I actually have four. Emma S. 2:01 Agenda items for us today. Emma S. 2:05 I was thinking. Emma S. 2:07 Because he in Christians spoke about reporting and PDFs sending pdfs and slack. Tom T. 2:14 Yep. Emma S. 2:16 And then he also spoke about the collections forecast. Tom T. 2:20 Yes. Yep. Emma S. 2:25 I would love to cover those two a bit more in-depth. And then we have a design prototype that we're just for Tom T. 2:33 Mm-hmm. Emma S. 2:36 Forecasts In general what we refer to as our base forecast, kind of the baseline that we set you up with and some initial like controls that we want to give to Tom T. 2:42 Yeah. Emma S. 2:46 our customers. Emma S. 2:47 In terms of timing. Emma S. 2:50 some nitty degrees in terms of the behaviors, Emma S. 2:53 I know you're probably quite interested in in daily granularity on things, Tom T. 2:53 Yeah. Emma S. 2:59 right? Emma S. 3:00 So I would love to just get some initial feedback there as well because we are thinking we currently aren't working actively a lot on it. But we are thinking a lot on how to Tom T. 3:10 Okay. Emma S. 3:13 think about the short mid long term, use cases and also, in terms of kind of Emma S. 3:19 What granularity is suitable? Tom T. 3:22 Yeah. Emma S. 3:22 We can talk a little bit more and then also investments. Emma S. 3:28 And for, I heard rates, I heard credit ratings. Also, just curious to learn how it works for your currently, what expecting from palm and how we could set Tom T. 3:37 Yeah. Emma S. 3:41 something up? Emma S. 3:43 Works nicely for you. Tom T. 3:44 Awesome. Yeah, there's a lot. Okay, that's good. Emma S. 3:46 Yeah. Emma S. 3:48 Me actually start with showing a little bit of a prototype for the reporting use case. Emma S. 3:55 If that's fine with you and you can feed back, we can have that as kind of our Tom T. 3:56 Yeah, please yeah. Emma S. 4:00 starting point and you can just Emma S. 4:03 And let us know. Emma S. 4:05 How you think about that? Let's see, that's the right one. Emma S. 4:13 I'm gonna remove us. So I won't see you now but just anything. Tom T. 4:16 Okay, no good. Emma S. 4:20 Okay. Can you see the screen? All right. Tom T. 4:22 Yes, thank you. Emma S. 4:24 Cool. So we are going to release. Emma S. 4:31 Before end of this year, a self-sub feature in poem in terms of the reporting so that you can build own. Emma S. 4:40 Reports. Emma S. 4:42 And with that comes a few redesigns. And as you can see here, this is Emma S. 4:48 More the look and feel of the dashboards and reports that we're going for. Emma S. 4:54 As so, imagine this would be a Emma S. 4:57 Operating cash flow, kind of report and now you would like to share this with Tom T. 4:59 Yeah. Emma S. 5:02 some stakeholders. We imagine something like a preview function. Emma S. 5:09 Where you could see? Hey, if this was now a PDF, what would that look like? Emma S. 5:18 and at a high level for VCR of this, we're thinking about stuff like Emma S. 5:25 Maybe you'd like to include or exclude filters. Emma S. 5:29 Given that. Tom T. 5:29 So they can see, yeah. Emma S. 5:31 Given it could provide some additional context. Emma S. 5:36 um, and maybe you'd like to Emma S. 5:41 Alter the stakeholder facing in description of the report or the title. Tom T. 5:47 Mmm. Emma S. 5:49 What are your thoughts so far? Tom T. 5:53 Very, I actually took Christian like, quickly touched on this previously. I think too. It's one of the other reports we're working on. And so I, I saw kind Tom T. 6:03 of first-hand how he started building out a report and the way that you can, I don't know if we'll get the full access like the users would get full access to how that looks straight away. I imagine not, but in terms of the ability to a essentially create a bespoke report and then have it as like a Tom T. 6:22 Quick Quick, Download Quick. Export PDF, it. It's great. The Tom T. 6:28 Free flow kind of like almost table base. I think it was where you can add things in and add separate them out and in terms of how you want to see them, it looked looked really good. I guess, question, from my side when if we were to build out a report Is there options to add in like a text box or something as a. So, if you have a graph like that potentially you can actually put in some Tom T. 6:51 commentary next to it. Or Besides some of these, I think that that would probably be there one thing to stand out or just one clarification, I guess from my side. Emma S. 7:01 That's really good feedback and I will definitely look into it. I've heard it before so I will. Tom T. 7:07 Mmm. Emma S. 7:09 Take that. And I will ask our engineers. Emma S. 7:12 Because that I agree, that would be amazing if you could add your commentary and Tom T. 7:12 Okay. Emma S. 7:16 just to clarify. Would you like Emma S. 7:20 Be okay with it being the same internally as if this is your dashboard that you look at and even if you could, it be like the same type of commentary. Tom T. 7:27 Yeah. Tom T. 7:33 Sorry. Sorry I'm not sure I followed. Emma S. 7:36 so, if I were to add a text box over this pie chart, Tom T. 7:39 Yep. Emma S. 7:40 Or under it. Could that commentary be the same here as here? Tom T. 7:47 yes, yeah, yeah, yeah, exactly that's that's what I was thinking in even if I Emma S. 7:49 Nice. Tom T. 7:52 mean Tom T. 7:54 I this is probably getting a bit. Very I can imagine how difficult something like this would be set up, but even if there's an opportunity to say, if there's like an outlier or a certain data point that looks odd to like flight highlight that in a way. So like obviously here, there's nothing really it's all quite within a trend but if there's one One, specific cash flow, one big item that really shifts, everything up to flag that in, like a different color or highlight that in some way in the charts themselves, that sounds super difficult to do for me. Even though even I'm not Tom T. 8:27 familiar with the technical setup side, but it probably more than a night more of a nice to have. But yeah, just just a interesting one, right? I don't know. Emma S. 8:36 Gotcha, I would love to cycle back to that once we've gotten a bit further. Tom T. 8:40 Okay, sounds good. But no I I guess overall the the reports look good. I know Emma S. 8:46 oh, Tom T. 8:50 this is relatively a straight like a relatively straightforward. One is an example, but I think from what we were looking at what Christian was kind of showing me in the background of what's being worked on the intuit intuitiveness Emma S. 9:00 Yeah. Tom T. 9:01 in terms of how you actually set up the report and what you can add in there, looks looks really good from my side. Emma S. 9:08 Super cool here. That's really nice feedback. Emma S. 9:11 Okay, moving on. So yes, you could download it immediately. We're also, we do Tom T. 9:16 Yep. Emma S. 9:18 have already this scheduled email capability. Sorry, I think I think I have to Tom T. 9:22 Mmm. Okay. Emma S. 9:24 do this. Now, I can click Emma S. 9:27 and then you can sort of Tom T. 9:30 Send it out. Emma S. 9:31 Add anything to your email subject, email body. And then add a frequency of send out. So on now, I'm going to see if my managed to. Okay. No, I can't much but Tom T. 9:37 Mmm. Okay. Emma S. 9:45 essentially we'll do the same thing. Emma S. 9:49 We're thinking for slack. Tom T. 9:50 Yeah. Okay. Emma S. 9:53 my question here is, Emma S. 9:55 Do you have multiple slack channels that you'd like to share things to, for example? Tom T. 10:02 Um probably to make things a little bit challenging. Yeah, we probably do but we could also what I could pay potentially in physics is that we set up like a dedicated channel just for updates or reports that we get from Palm because it would likely only hit a certain group of individuals, it wouldn't necessarily at the moment, we have different reports that are going to different teams. We have different individuals who are interested in things, but if we come with a, if we have a system, we're looking to automate automate, this like the buzzword of everything. At the moment. There would be a very strong argument where we can Tom T. 10:39 push back to the business and say, Can we refine this a little bit and create a dedicated channel for Tom T. 10:47 For Treasury updates, Treasury reporting where it's automated. And then the reports themselves. We would just tag though. You know, flag those this is our Tom T. 10:56 daily cash report. This is our day what I like our month-end cashflow categorization whatever. So the interested people can can always reference. Those just if we have consistency throughout the updates, that that would be it, Tom T. 11:07 but I think they'll be a good argument to go to go ahead with that as opposed to Tom T. 11:13 The other alternative I guess gets a bit more complicated, right? If you have Tom T. 11:17 lots of different slack channels that you're looking to to send out to. Yeah. Okay. Emma S. 11:22 Okay, cool. Emma S. 11:23 so yeah, this is Emma S. 11:27 Basically what we're envisioning for now for the first iteration of reporting and automated voting. So and Tom T. 11:34 As is really a really interesting to see like in terms of the, the flexibility, like, my just comparing it to some of the reporting functionality that I've seen Tom T. 11:43 in other systems. It's been so limited and rigid really and what you can create. So, I'm excited to see this and play around with a little bit. When it's when Tom T. 11:52 it's when it's ready to see what we can build out. Because actually, I mentioned to Kristen yesterday, I was on a call with him that what we have is kind of legacy. At this point. It's just been built out over a necessity of a particular Tom T. 12:03 request that are particular point in time. What we're going through at the moment is like a trying to replicate that within the system, but I'm also thinking outside of that and I'll challenge things a little bit to say over time. What else should we be? Doing is a reporting. Like you mentioned the Tom T. 12:20 investment side. We do. We do have cash and investment reporting at the moment, like our deposits, and the rates and credit ratings, all this kind of stuff. We Tom T. 12:28 we have all of that and essential space at the moment but if there's a potential over time to shift that Tom T. 12:34 To a palm report as well and automate that as an overview. It's just a huge time saver on an ongoing basis, not just for me, but for the rest of the team. So yeah. Emma S. 12:43 But maybe we should dive into the investment and do collections for car. We can Tom T. 12:46 Sure. Emma S. 12:47 do any order but maybe that's makes more sense. Where so where does it live Tom T. 12:49 Yeah, sure. Emma S. 12:54 today? Where do you have the data? How does it? Tom T. 12:56 Yeah, it the data is in a, in A, in a g sheet, essentially, we have various different investments, but we have what I've tried to create. It's just like a very standardized table of all of the investments that we have. And as they mature, we add a new lines and very well. Essentially, we have investments Tom T. 13:14 across term deposits, money market, funds, and then Emma S. 13:17 Sorry. Sorry for interrupting. What would would it be possible to screenshot Tom T. 13:19 Okay. Tom T. 13:21 Yeah. Yeah. Emma S. 13:22 show me? Yeah. Tom T. 13:23 I can do that one second. Tom T. 13:27 Let me just get this up. Tom T. 13:32 So it looks like this is like the base data. So everything in gray is like a Tom T. 13:40 past deposit or a mature deposit and going back all the way to where we 2022 and then, everything in white is like, a live investment. So we have certain Tom T. 13:52 investments which are turned the posits. We have a big shift towards, like, overnight cash. These are like, interest-bearing balances that we have with Tom T. 14:01 banking banks, but the liquidity is not. Tom T. 14:04 It, you generally need to request and have maturity and then you have, yeah, investments in money market funds as well. So you have various different types of investments. What? We, then layer into that is like, balances that we Tom T. 14:15 actually do hold on account. So we have all different. We have like four different types of balances that we're holding. Some have Tom T. 14:23 Interest bearing like an interest rate attached to them, others have? Tom T. 14:28 Our immediately available like money, markets funds. We can we can get access to those balances immediately. If we want to same with the overnight accounts, we have a, We need to request those from the banks. We can get access to them. Tom T. 14:38 Others, like, the term deposits, have a fixed maturity, which we can't access. And this is kind of semi-restricted At that point. What we try to do is through this table is just build this out into a really standardized way of kind of referencing the data, so you'll see. Yeah. The the bank reference information on Tom T. 14:55 the side, you'll see what the type of instrument that we're referring to in Column, e currency, obviously what the principal value of the actual investments are and then euro equivalent duration of the investment is important. But actually probably more important is when we start and end, you have the maturity, the top, and then other information, which I guess would be automated in a system. This is an active investment or not an active investment. Tom T. 15:22 Basically, what I use all this information for is to then put it into a series of pivot table. Tom T. 15:28 Like this where we can, then view the information. Tom T. 15:32 In the consolidated way, in terms of total balances, the types of balances that we're holding like money market, funds, bank, balances overnight, the council, this kind of thing. Tom T. 15:42 I screenshot all of this and put it into a report, like A Emma S. 15:42 Hmm. Tom T. 15:49 Yeah, PowerPoint type report then get shared with the management team but the Tom T. 15:55 It's just a lot of like you can imagine that the upkeep of this sheet is quite manual. It's not not the best in this prone to error as well. So Tom T. 16:11 the deposit only did this file for kind of the base data is is Tom T. 16:15 Yeah, takes a bit of upkeep to make sure that it's, it's working correctly. We have access like that the big differentiator and I think you you rightly mentioned at the beginning is that we have bank balances. That's fine. All of Tom T. 16:27 those balances are already already at the moment for a uni credit and our HSBC account. So already pulling into palm. So there's like, those updates going on. That would include the majority of the balances that you're seeing here. But Tom T. 16:40 what's not hitting, or not showing updating directly into Palmer, like, our money market fund, investments, or are termed deposit investments here. So what would be Tom T. 16:52 and like a super helpful thing would be able to like when say we have a new investment, Tom T. 16:57 If we have maturity, we actually input the details. We can do this manually or Tom T. 17:02 however, but input the details of that investment. Whether it's like, a, a new investment or it's we're increasing investment. Maybe on the money market fund side or however, but just a place where we have saying money market line and we can increase that value over time. Or we have a deposit with a fixed emma Tom T. 17:20 maturity, we can actually input that and that then shows just like a Tom T. 17:26 Shows us as part of our total cash balance but then you also see the individual you attach kind of the individual details of that investment. So the the rate, the maturity, all this kind of stuff because that then would help I guess to get us to sorry jumping a little bit but get us to a situation where we could probably start reporting on it and separating out the different. So we have cash balances. This is how much we're holding at this particular bank, cash balances Tom T. 17:53 plus investments. You know, this kind of thing that that's where would be great to get to. Emma S. 17:58 I get that and also I guess all the maturities and seeing how look good you are at different point in time. Tom T. 18:03 Yep. Tom T. 18:04 Yeah, that exactly. That's that. That's like the last bit. Again, lots of lots Emma S. 18:07 Yeah. Tom T. 18:08 of tables that we have in here. Sorry, I'll skimmed across that but that's to your point maturity is like feeding to our liquidity planning, at that point. So, we're looking at how much GBP and USD we're holding it any one time. We have Tom T. 18:21 certain criteria that we need to hit in terms of a policy. So all it's not yes, it's the balances, and the investments themselves, but it's also then liquidity planning and decision making when it comes to FX trades or all this kind of stuff, it all comes from that base data, basically just above that, I kind of skimmed over a little bit but we do have a investment policy that's the credit rating piece. So we have the We attach kind of what we have a restrictions, we say, in terms of the banks that we can trade with the banks that are acceptable for us, from a trading perspective, and certain criteria. So yeah, credit ratings whether they're Tom T. 19:03 deposit, protected, certain banks in Germany. Come under this deposit, protection scheme. This kind of thing that we try and include as again base data to say if we're compliant or non-compliant with our investment policy at any one time. Emma S. 19:16 Culture. I read France as a country are being downgraded in terms of Tom T. 19:21 Over here. Oh yeah, not great. No, it sorry. Could go ahead. Emma S. 19:25 Yeah. Emma S. 19:27 But yeah, but so question here. If we assume into the credit ratings, how do you Tom T. 19:33 Yeah. Emma S. 19:33 currently maintain them? I heard something about an AI agent or Tom T. 19:36 Yeah, yeah I'm trying I'm sure it's the it's a classic AI agent that it works once and then it's not working again or it's it works once but then you refresh the data and it brings brings through different information. Yeah. So the essentially what I've tried to create is A Tom T. 19:56 All of the banks that you see here. So Citibank Deutsche Bank HSBC, uni credit, they all publish all have a area within their Tom T. 20:06 online work that their websites where they detail out credit ratings, the difficulty is, is that those ratings are Tom T. 20:14 Each so Citibank can have North America. Citibank again have Citibank Europe City Bank application and it also all of these banks use different language to reference different credit ratings. They have so you can have short-term credit ratings, long-term credit ratings secured and unsecured credit ratings like a long list. So it gets pretty difficult to get the comparable rate and actually Tom T. 20:38 the the right terminology credit rating across all of these different banks, that's where it gets really tricky. Tom T. 20:46 You basically have to instruct this AI agent to go into each. Tom T. 20:50 Website and pull out the exact right rate ting for your for your banking partner and using the correct terminology it is just it's not full proof. That's the difficulty. Emma S. 21:03 I gotcha. Gotcha. And how often would you like these two updates as a daily? Tom T. 21:08 No, we are. We update these where I might. Tom T. 21:12 Daily would be great. The daily would be the most preferred because essentially Emma S. 21:13 Yeah, but Tom T. 21:17 we only really use this as a reporting tool at the moment. What we're really not in a position to do, I'm not able to update these on, I don't yeah, mind on a manual basis, like on a daily Tom T. 21:27 Daily basis. Yeah, but that should be. The goal is actually that any time there Emma S. 21:28 Yeah. Tom T. 21:33 is like a movement in one of these ratings or a downgrade or even an upgrade, Tom T. 21:37 we'd want to get some sort of notification. So from a reporting perspective, I I update these on a monthly basis That's how like monitoring window, but I've, I've seen it in other Bloomberg, for example. The I don't know if you're familiar with that like the yeah, the screen and the terminal that a lot of, a lot of the Treasury departments. I've worked in have that Bloomberg terminal and Tom T. 22:00 they have all types of setups where if there's anything that conflicts with their, their rating scale, if they get a king and it, you know, it's a drive to to do something about it. Emma S. 22:14 But you don't currently have Bloomberg episode, right? Tom T. 22:18 No, no. It will be, we don't necessarily need it, but it's something that as the business. Tom T. 22:24 As a business grows and and you tend to get more complicated in terms of your investments, but also in terms of your FX and maybe different types of tools that you might be using like hedging tools, that kind of thing. You, you tend to need a Bloomberg tool or having the past. I don't know. Let's see what what AI Tom T. 22:42 comes up with to see if you actually need that. But, yeah, that's how it's been Tom T. 22:47 managed in the past. And they've had kind of a monopoly in on that information, as well. Like the Emma S. 22:53 Yeah. Tom T. 22:54 So it's quite expensive and we don't have the budget or really the need for it at this point. Tom T. 23:03 Yeah. Emma S. 23:04 It's an interesting one for sure. Also, thinking from our side, how we might Tom T. 23:06 Yeah. Emma S. 23:11 Get it as just to begin with. Tom T. 23:16 It's, it's a difficult one and I actually don't know like the Tom T. 23:21 The rating agencies themselves, like Santa Paws, Fitch Moody's. The, the most recognized ones, the three. They Tom T. 23:32 The you can sign up to those, but they're subscription-based as well. They cost money to sign up to and I don't like that's just as an individual. So if I was to go in there and Tom T. 23:53 B2B perspective. I don't know what the Tom T. 23:57 I think I'll be very expensive, honestly, to have that set up. Emma S. 24:04 Gotcha. Emma S. 24:07 okay, but this is helpful, we can Emma S. 24:11 let's see what we could come up with from our side, but I think, Tom T. 24:14 Yeah. Tom T. 24:15 I think that I guess. Emma S. 24:16 the very minimum you would Emma S. 24:19 Like that, at least. Tom T. 24:21 Yeah. Yeah. That's what literally what I was gonna say is that, if, if nothing Emma S. 24:25 In. Tom T. 24:26 else, what would be really helpful to do is that if we have these types of investments, or we have these types of balances is to in some way, attach the Tom T. 24:35 rate and the maturity, Tom T. 24:38 Within parts of the data like it's not just the balance itself. It's you know if it's an investment it has its it did those additional data points that we could then use to Tom T. 24:49 Yeah, to report on. Yeah, even if it comes to the manual update of the credit ratings, if we had that within palm, like if we and we were responsible for those updates again, we could still, we could get to a point where we can replicate the reports in Palm, as opposed to what we would likely have to do is Tom T. 25:07 Do a quasi, pull the data from Palm and then upload into a G sheet and then Tom T. 25:12 attach the other manual information. And that that's what I really like to avoid is if we're shifting to a system is even if that system needs to be manually maintained, I'd prefer to do it that there as opposed to doing a doing it in both both ways. Emma S. 25:26 I'm wondering because what we are. Exploring internally. I'll just Emma S. 25:31 It'd be fun to pick your brain about it as well, right? It's that we we're exploring the ideal actually modeling out the counterparties Emma S. 25:39 The banks as their own, sort of conceptual accounts investment account in Palm, right? Tom T. 25:47 Yep. Emma S. 25:47 So, you would have your Emma S. 25:49 Ages BZ. Emma S. 25:51 UK. Emma S. 25:53 Investment account. Emma S. 25:56 And I wonder if yeah, it would make sense to, then have some additional, you know, ability to add this credit rating across those accounts. And I'm wondering Tom T. 26:04 Yeah. Emma S. 26:08 just thinking bigger, and not saying we might do that, or, But is this relevant for any bank account? Not just the investment bank account. Yeah. Tom T. 26:15 Yep. It's relevant for all bank accounts. Yeah. So even if it's not an Tom T. 26:20 investment, say which we look at JP Morgan, for example, if you see on their we Tom T. 26:27 or maybe a better example is Tom T. 26:30 Let's say several is a an account or it? Sorry a bank account in Spain that we use for very specific purposes. It's not used for investment in any way, but we still care about the credit rating of the bank because we hold cash there. If Tom T. 26:46 they're downgraded to a certain level, we still need to move fonts out of that account and engage with the different banking bank account banking party. So investments are one thing. Like it's it's almost Tom T. 26:57 There's always like the Tom T. 27:00 So there's like the the scale of it like security liquidity. Yield is always the three points of an investment. So security security is always like the, This is Emma S. 27:06 Know. Tom T. 27:08 what you're seeing At this point is the counterparty risk. Liquidity is like the duration and yield is, is the last bit, like for a lot of people it's flipped people think the youth is the most important, but actually, it should be the security itself. But essentially, yeah, for any balance you have that yield could be zero. So it doesn't necessarily Tom T. 27:29 It's essentially always you need to see all of those combined together to get a feel for it and Credit ratings and the counterparty risk will apply to any balance. You have not necessarily just the investments. Emma S. 27:41 Yeah, that makes sense. Emma S. 27:44 Makes a lot of and so rates, When you were talking about rates, I will assume Tom T. 27:44 Yeah. Emma S. 27:49 you just want to get the rates from, for example, that you already have in your dock. Like, that's the rate that you or is there any other? No. Tom T. 27:56 No, no. That would be the rate and that, I guess that's the, the more challenging. Well, that's where they're, I think there would always need to be a point of inputting that, like, the deposit side is always going to be based on a trade that we make with the bank directly. So it's not going to be something Tom T. 28:12 that's maybe publicly available as a rate. So there would always need to be a certain element of us in pushing that information. The rate would be unique and probably that the duration of the deposit would be unique as well. When it comes Tom T. 28:23 to talk about deposits, something like money market fund is more interesting, I guess, because that information, the essentially the Tom T. 28:35 The way that you derive, the invest, like the the return on, on a money market fund is, Where am I looking? There? Were way that you derived. Yeah. The money Tom T. 28:46 market fund return. Would we have on Monday night is, is driven by. It's almost like a share value. It changes every day for them. And so we have a duration of Tom T. 28:57 our investment. We invest on A1, and then over the course of the lifetime, the shares quite should we say, the appreciate and value. And that's, it's only when we redeem the value that we actually recognize the, the investment return. But it could be that we invest Tom T. 29:13 for 29 days and appreciates an on day 30. It's lumps and we lose. We lose money. It could be the case, but that's where Tom T. 29:21 Money Market That information, let's say at least the net asset value of a money market fund is publicly available. So as long as you can attach, Tom T. 29:30 Or reference. Tom T. 29:32 Like correctly, the money. There's a lot of money market funds to invested but as long as you can reference the correct ones that would potentially be a Tom T. 29:41 A chance to pull that information from. Tom T. 29:45 Externally. Emma S. 29:48 That is true. Tom T. 29:50 Yeah, but I guess I have. I'm coming up with lots of suggestions, but I don't Emma S. 29:51 Just the question on them. I I do appreciate the ideas though, so it's keep them Tom T. 29:55 get me wrong. I appreciate how difficult it might be to bring all of this into the system, particularly like it. I could not an overnight thing, but Tom T. 30:05 Yeah. Tom T. 30:06 Yeah, just at least the ability to input this and so we have it in one. One area would be, would be amazing. Emma S. 30:16 coming. It's great. And I see the value of having that automated. So you would Emma S. 30:20 have a pub, you would have perfect visibility essentially. Tom T. 30:23 Yeah, basically. Emma S. 30:24 Yeah. Emma S. 30:25 How do you currently manage it in terms of like you go in and just check the value every day manually or like what working? So Tom T. 30:34 Yeah, pit periodic. Some of these I'm in I'm looking at this on a monthly basis so and some of them I'm looking at weekly some some daily it's Tom T. 30:44 When it comes to money market fund returns, it isn't the funds that we invest in are relatively. Tom T. 30:51 Then also volatile. Let's say so you're not getting these huge movements day today. You're generally getting this very, very slow but saw increase. I mean, Emma S. 30:58 What hopefully? Tom T. 30:59 think it's got. Yeah, I think it's cost. Um, but that that's essentially we're invested. So I'm quite comfortable to update this on a essentially, on a monthly Tom T. 31:06 basis. There are other market money market funds, which, you know, engage in more risky. Activity, the do that are much more volatile, so it. Yeah, I know Tom T. 31:19 you would never have anything like, in particular, just one customer but it's a thought that, you know, if there was Tom T. 31:34 Um, Bloomberg itself. And I know other other sites that we spoke to other providers to be spoke to, they can integrate with Tom T. 31:43 ERPs and so forth. So similar to a banking system you can, it doesn't necessarily have to be that you, you put all this information. It may actually be far easier than that as a one-to-one integration where you can pull the information as well. Emma S. 31:57 yes, that would be Emma S. 31:59 A very valuable thought, I think for us the leverage, what's already there. If Tom T. 32:01 Yeah. Emma S. 32:04 there's something. Tom T. 32:04 Yeah. Yeah. Emma S. 32:06 But I'm also keen just to understand. Do you get do the banks and you any Emma S. 32:11 reports? You know like I get stuff from my bank, right? Oh here's the update. You of your holdings? Tom T. 32:17 No, that you can you can get that and because some some banks are better than others. The Bloomberg, for example, they the the system that they have the update, the Tom T. 32:30 Reporting system that they have in their, in their site is really good. You could just click a few things and add in and you essentially get the report that you want to other banks. It's Tom T. 32:41 They seem to try and make it difficult to do the reporting because they tend to want to push for. Tom T. 32:47 Full integration or some sort of additional reporting suite that they have some, some banks are pretty good. They'll have like a liquidity portal but or like a certain module that does some of the reporting for you. But it's also Tom T. 33:02 I always struggle with that as a concept, because no, any any Tom T. 33:09 Corporate or any business. I generally not banking with one bank. So you would have so you'd have to integrate all of your other banks into that one bank, and then you become super reliant. On that one bank, you can't necessarily shift Tom T. 33:25 your business elsewhere, in an easy way because you get kind of locked in. Tom T. 33:30 I've never used. Yeah, to bonus. I've never used the reporting like that from that. From a banking suite it, we always pull the information out externally and Tom T. 33:37 do it yourself. Tom T. 33:39 Or I have done I don't know. I'd be interested to see what some of the the guy. Yeah. What the people at Palm think about that but I yeah I generally don't use Tom T. 33:47 the banking reporting. Emma S. 33:52 That's fair. Emma S. 33:54 Very cool. Emma S. 33:56 Yeah, I think we do definitely need to take into account those different types of investments, those different types of I mean, even from a forecasting perspective, you might be able to forecast yield from time like I'm deposit. Tom T. 34:10 Yep. Emma S. 34:11 The money markets funds. That's the forecasting bit of that is still a bit. Like, is that something you would care about forecasting, typically, or is it just Tom T. 34:21 I, I think it's Tom T. 34:24 Probably I mean to an extent yes but it wouldn't be always be for me. Well at least at this point more of a nice to have because the rates that we get are so dependent on external factors like Tom T. 34:37 Central bank rates, so forth. So we what would be interesting to see is like, building in. It's assumptions on central bank rates, or if you had the ability to do some scenario, planning long term around. This is the band. So if you have Tom T. 34:52 cash balance of today and what we're investing today and then are expected, cash burn, or, you know, profitability or whatever in this system. And then also layer in assumptions around, ECB rates, like our average weighted average return on investments over time. That would be great. But that again, that's a talk Tom T. 35:10 about nice to have ice. That's quite far down the line. It's not not something. Emma S. 35:13 What? Tom T. 35:16 It's something that we budget as part of our planning process. We, I do some assumption-based kind of interest for casting. Tom T. 35:26 But it's very, very high level because it's it's Yeah, I don't have a crystal ball, you know, you can't see where rates are going to go necessarily what what I would say. Emma S. 35:35 but, Tom T. 35:36 Though on the forecasting piece is you you mentioned about deposits in an investments. It's like Tom T. 35:43 There's the the balances that we have over time there's then the transactional forecast, what is the like ultimate What would be that? The best thing to be able to view is the availability of cash. So we have Tom T. 35:56 So we have eight. Yeah, 80 millions worth of cash at the moment, but 40 million of that is locked up in investments for the next six months. It it's not like Emma S. 36:03 Yes. Tom T. 36:04 and we have a forecast where we're burning through cash, What helps us is then is to layer everything together to show when we have a deposit maturity. Emma S. 36:12 Yes. Tom T. 36:13 In the future and say, Actually, you know, that that that's really when you get to the point where you can make decisions on investments with certainty to say, Tom T. 36:21 Oh, we're comfortable not to invest overnight anymore. We can actually push that out and generate a better rate of return, because we know how much cash we're going to have in three months time and we have X amounts, deposit extra amount of deposit maturing in two months time, you know? Emma S. 36:37 No, gotcha. I'll video. Hey about that. Just like curious. Sorry, very, very Emma S. 36:42 specific question. But do you have any favorite? Emma S. 36:46 Way to visualize to maturing balances like, Is it a ladder? Is it like How is there like a good way? Emma S. 36:54 See this. And that makes, yeah. Tom T. 36:56 Oh yeah. Tom T. 36:57 I I think it for me, it's always on like a line almost. So if you have like, your transactional Tom T. 37:06 Value like but then you have, we add the moment, I superimpose it on the background. Everything looks overnight, because all of our investments are pretty short term. Money market funds are overnight. Tom T. 37:18 Yeah. Etc. But what I would always love to see is like a something that pinpoints, this is a maturing investment, if it's on the Tom T. 37:27 Y axis. It just says here we have maturity but then you also see a corresponding spike in your cache, essentially. That's all I really care about is is the movement of that cash flow over time. And if there is a yeah, maturing cash just Tom T. 37:42 to identify that in some way within there like a obviously, a spike in the amount of cash that you have, but actually to say that that is built on a Tom T. 37:54 I'm not thinking about it, hang on. Tom T. 38:03 well, but what I could potentially see is like, if there was a column graph, so kind of Cash balance, like by on a weekly basis or something like that. But you could say, This is a, You're expected cash balance at the end of each week. And if there was a significant increase, then you have like any investment maturity during that week in a different blog, maybe? So you layer it on top like a Tom T. 38:23 layered column graph, something like that. Emma S. 38:26 All right. Tom T. 38:27 But I, I would have to have to think about it, to be honest. Emma S. 38:27 Very cool. This is Emma S. 38:30 Yeah, at least think this is very neat to gritty but I think like these things Tom T. 38:33 Yeah. Emma S. 38:34 shouldn't be underestimated in like times of how can we visualize it in a way that really drives. Tom T. 38:40 Yeah. Emma S. 38:40 The app you're looking for is one thing looking for like when will I have the cash? But maybe it is another thing like Emma S. 38:47 Well what's my actual excess cash that I should safely be able to pull out into a time deposit so that I can locally in belong there because I know that it will be it was I will still be capable of operationally. Tom T. 39:00 I mean, if you go, Tom T. 39:14 historical flows but any movement where we pay out to a money market fund a week, we receive funds back in, they generate a Tom T. 39:23 That they should get categorized that that flow. But what would be also great is that if you have that the cash balance, then the forecasting piece link to each bank account, is that if we do have a maturity Emma S. 39:35 Yeah. Tom T. 39:36 That you actually see that cash flow showing as a like a forecast value there as well. That would be awesome. Emma S. 39:45 Would that also apply to if you have it? Emma S. 39:49 Any associated fees being an outflow and any associated? Like yeah, rates like Tom T. 39:55 Yeah, yeah. Okay, like the So if you have the debit on day day one let's say of Emma S. 39:55 read. Tom T. 40:01 the principal value. But then yeah, in two months time or whenever the maturity is you have the, the principle of return plus the the interest that turned. Yeah. Emma S. 40:10 Very cool. Emma S. 40:12 all right, could I show you another prototype that is Tom T. 40:15 Yep, these two, let me. Yeah I'll just get out of this. Emma S. 40:17 Yeah. Emma S. 40:18 Thank you for sharing this. Tom T. 40:19 No, no problem. How do I always get this? Tom T. 40:23 I'm still getting used to Google. Sorry. Okay. Emma S. 40:26 Don't worry. Emma S. 40:28 It's yes. So okay. Emma S. 40:32 Here is the other one. Emma S. 40:36 So, what we're Emma S. 40:38 Looking at here, whoop. Emma S. 40:41 Is imagine. This is the current account page in Emma S. 40:45 In palm. Emma S. 40:47 and I just thought wasn't very clear but I clicked collections Emma S. 40:51 And that opened up this sidebar. Tom T. 40:52 Hmm. Tom T. 40:54 Yeah. Emma S. 40:56 And what we are just exploring now is to play around with this concept of details around the forecast. Emma S. 41:06 And this division between like, short, and midterm. Tom T. 41:09 Hmm. Emma S. 41:09 kind of, Emma S. 41:12 Some additional information and then maybe some explanation here at which, well, Emma S. 41:18 you know, what's the system forecast or the palm forecast behind all this? Emma S. 41:24 It could be a machine learning model, that is called so and so or it could be a statistical model or any, right? Depending a little bit on the category and the available data and kind of what drives the Best outcome in terms of variance. Emma S. 41:42 we're also looking into some of these more, like, Emma S. 41:46 Easy access focus versus actual. Emma S. 41:49 Things. But Emma S. 41:51 But for now, it's actually this configuration part that we've prototyped out here. Emma S. 42:01 We're aware that it's likely gonna need to be maybe two to four weeks for whatever is considered a short time forecast. But Tom T. 42:09 Sure. Emma S. 42:10 Here or do you if you disagree, let me know but Emma S. 42:15 let me just try and explain it first and maybe Emma S. 42:18 so what you see here are a bunch of lines Emma S. 42:23 We have the system forecast, which is this blue dotted line, and the system forecast is essentially palm. Tom T. 42:29 Okay, yep. Emma S. 42:30 If you see that the official forecast, it actually follows the blue dotted line Emma S. 42:35 except for here. Emma S. 42:37 where we have like a Emma S. 42:40 We're still exploring naming of things but imagine that one off here. Tom T. 42:42 Yep. Yep. Emma S. 42:45 So the one of that you added, which pushes the kind of official focus line up, but otherwise it's currently just following the system forecast at Palm generated. Emma S. 42:57 But you also have your we currently call it confirmed forecast. Imagine it's Emma S. 43:04 Some sort of ar uploads that you did. Tom T. 43:06 Yep. Emma S. 43:09 so, Emma S. 43:11 You can see or kind of numbers beneath the graph here and then what we wanted to experiment is, you know, allow you to depending on the Emma S. 43:19 Short, or midterm. Emma S. 43:22 You might want to, you know, say Hey I have amazing AR data. Emma S. 43:27 And I want to, I just want to rely on that. So now the line here, the official forecast line updated to, for the first two weeks, Emma S. 43:35 Might need to be four. Emma S. 43:39 It, you're just going to say that whatever activity whatever forecast for this account. Emma S. 43:45 For the short term, I always want to rely on my kind of booked actuals or confirmed forecast data that I have. Emma S. 43:54 Whereas, you can see midterm. Emma S. 43:57 Okay, then we can use the machine learning model. Tom T. 44:00 Okay. Emma S. 44:02 The third option that we thought about was a little bit around like allowing to make an explicit choice regarding if you want to be conservative or take higher risk. Emma S. 44:14 so we were imagining this concept of a hybrid forecast, where you can blend your system and confirmed forecast, Emma S. 44:22 And what we mean there is you could, for example, since this is an inflow category, you could say, I want to be very conservative, so I want anticipate Emma S. 44:30 that we're going to get the lowest absolute amount. Emma S. 44:35 And and it will just pick the lowest absolute amount between your confirmed or palm like machine learning model essentially so that you can be a bit. Tom T. 44:42 Yeah. Emma S. 44:46 Safer. Tom T. 44:46 Executive. Yeah. Emma S. 44:48 And inverse, of course them for for accounts. Payable, for example. Tom T. 44:54 Awesome. This is really interesting. Sorry, Karen Emma S. 44:56 Yeah, no, please feel free. Tom T. 45:00 No, I was just gonna sit. This is like how Tom T. 45:04 I'm already kind of thinking around how it could work with, with the information Tom T. 45:08 that we have because we the way Tom T. 45:12 the way that we forecast at the moment or the you know, very, very rudimentary Tom T. 45:16 forecast is that I take a Tom T. 45:19 The Fiona team that we have, they work on like a an update on a monthly basis which is transferring all of the revenue information and all the supplier information, all of that stuff into what they expect is going to have happen within the next month. So, actually short term over the coming month, we've relatively accurate. But all I get is like a single line entry. So I have my AR Tom T. 45:40 data for example is to say we think we're going to collect 20 million in November. And we can be quite confident that it's going to be that value. But I, what we don't have is then the system base forecast, which I guess would be. Here is Tom T. 45:53 like when that's going to happen. The certain days within the month that we collect more than others. So what would be interesting to see is like, we input essentially that 20 million value but then the system takes over. Tom T. 46:08 Or we have the system-based forecast, let's say maybe on a conservative level, Tom T. 46:13 which then shows the value that we of each day. So I'm working day, two, three, four, five, six, seven. When we collect the majority of our monthly inflows, we have kind of Tom T. 46:25 Like higher balances and then over there over the next two that the final two weeks of the month or whatever it kind of drops off more. So I don't know if that would that would also be a possibility to link up. Tom T. 46:36 Link those up in a way or you have in the short term. Tom T. 46:40 Like a yeah, you rely on that system-based forecast because it has that historical trend data of when you collect wait within a month, actually, when you receive your cash, but we have that reference data. Tom T. 46:52 Essentially, we assume each day that we collect the exact same amount which is not correct, but we can actually over the course of the month, we can see how far away. Tom T. 47:01 On a, on a total basis. Let's say that we're trending away from from that value. That, that would be an interesting interesting one for sure. Tom T. 47:09 Does that make sense in any way? Emma S. 47:12 Yeah, it does, it does it does make sense but the currently not thinking too much about you, I'm just being substant with you about how this big one number. Tom T. 47:20 Yeah, sure. Emma S. 47:25 And trying to kind of spread that. So what I'm what I would imagine you would Tom T. 47:28 Yeah. Emma S. 47:30 like to do is sort of Emma S. 47:32 Again. Emma S. 47:34 I mean on the other hand and what I would imagine you to do is I actually Tom T. 47:42 Yeah. Emma S. 47:42 Or maybe the system's forecast should capture. Emma S. 47:45 That. Tom T. 47:45 That's my yeah, exactly, that. That's my thinking. Yeah, so that that's Emma S. 47:50 So it comes down more to maybe sanity checking against. But yet like, okay, oh, Tom T. 47:50 Like one. Tom T. 47:54 Yeah. Emma S. 47:55 totaling over those four weeks according to the system forecasts, Okay? The Tom T. 47:59 Yeah. Emma S. 48:00 system forecast says 38 million but it says 40. No. Emma S. 48:06 It's not for, or Tom T. 48:07 Yeah, that that it does it does make sense. I think the the only Tom T. 48:12 Maybe note I would have there. It's not. Not each month is the same when it comes to our collections. So there's certain months within the year where we we Tom T. 48:20 collect more than others. So I'm particularly year on year if you're looking Tom T. 48:25 Say November 2025 versus November 2024. If that was to be like an indicator on the system-based forecast that you would hope there would be an appreciation in terms of what we're Tom T. 48:35 What we're collecting also at times within the year, particularly in January, Tom T. 48:40 there's a really significant uptick in our collections within that month. Because of, essentially, we just have a lot of yearly customers, who pay on a yearly basis pay with in January and so yeah, I guess that would be the one. Tom T. 48:54 Maybe the one point on those like if you have an aeon number that you're relatively confident, is going to come in, if there was ever an opportunity to then like layer that down, and have the machine learning piece applied to it, but actually, just split out across the Tom T. 49:13 Forward-looking forward looking days. I'm not explaining this very well but you have that number for November, let's say that. But based on the recent trends, in terms of actually, when on the days that you collect within the month, if you're going to apply that to that number, that would be a, that would be great. Emma S. 49:28 Yeah. No it makes it makes perfect sense. I think I think we should actually Tom T. 49:29 That would be like that would essentially move away from our like, assumption at the moment that we collect the same every day into Tom T. 49:38 I guess what the hybrid model in a way where you have the A on number which we're relatively comfortable with. But we also have from my perspective that what I'm interested from a Treasury side is actually within our like, the more granular data will actually when those spikes are in terms of collections, within the month, if that makes sense. Emma S. 50:03 probably dive deeper into it. When we also go more towards the daily granularity of things and and there are also we have some ideas around Tom T. 50:08 Sure. Emma S. 50:17 You mentioned before kind of assumptions or driver based? Emma S. 50:23 Modeling or like putting drivers and assumptions into the forecast as well. And I worry if there's room for something like this there Tom T. 50:31 Yeah. Yeah. So yeah, on the AR side, for example, so you have lowest amount Emma S. 50:31 Like yeah. Tom T. 50:37 possible like if there was a Tom T. 50:40 Is that they're ever being like an option for like a percentage value, let's say, or something like that. To say If you have or something Maybe you're yeah but something that you have your assumed AR value. But actually, we're gonna decrease that by 5% or two and a half percent or something like that. Just so Tom T. 50:58 we're You know, more safe, always more safe on the Treasury side. Emma S. 51:04 No, no. I I hear you. I am. Emma S. 51:09 I, I don't know. This is a good or bad idea to show you this, but Emma S. 51:13 I don't care. Emma S. 51:17 That the I just need to see. Emma S. 51:22 Which one of all my prototypes it is? Emma S. 51:28 sorry, I I just wanted to sanity to check something with you, apologies of Tom T. 51:34 No problem. All good. Emma S. 51:40 Was. Emma S. 51:42 Let's try this one works. Emma S. 51:47 Okay, I'm gonna just share something that looks horrible. This is lovable with many months ago. Emma S. 51:56 What I'm hearing is that, maybe something with this. Let's say collections. Okay. Tom T. 52:01 Yep. Emma S. 52:01 the prepare yourself that this data will make absolutely Tom T. 52:04 I all good, all good, no problem. Emma S. 52:08 And for now, let's just say Oh for all my accounts, maybe doesn't make sense at all. But I can add an assumption. Emma S. 52:16 and I could add like, Emma S. 52:18 a name for my assumption but it increase blah and I can say that Tom T. 52:21 Hmm. Emma S. 52:25 yeah, by assumption the same assumptions that drivers Emma S. 52:30 That number you want to apply the same percentage change across? Let's say this the Palm System. Forecasts that I seeing we are capturing those patterns Tom T. 52:38 Yep. Emma S. 52:41 already. Emma S. 52:43 For example. Emma S. 52:45 so I could say, like, Emma S. 52:47 For the next, it could be one month as well. You can spread it evenly. You could explain something. Tom T. 52:54 Yep. Emma S. 52:55 Here. And I can save this assumption. Emma S. 52:58 And this didn't do much but then I could essentially preview how that would affect my forecast. Tom T. 53:05 Yep. Emma S. 53:06 And if I like it, I can sort of apply it to my forecast which now has Emma S. 53:11 essentially updated my forecast to to include this assumption. Tom T. 53:16 But didn't know that. I mean to your point. Yeah, you could guess this would be, Emma S. 53:18 Is this anywhere near what your imagining? I do realize that you're also just asking to add a percentage number, Tom T. 53:32 this would be great you can actually see. Yeah, I I appreciate we use like, but in terms of being able to add the assumption in in that way, and then essentially get a view, like, a go, no. Go to see if it makes sense. Then then, Tom T. 53:45 yeah, that would make sense. And if you're able to, I can see that you've kind of got different categories that you you can apply it to. You can say we're expecting to be slightly negative on collections that say by five percent. But Tom T. 53:57 actually pay let's say payroll we assume we're going to pay out more than we we've planned. If you can apply it in that way like a series of assumptions that essentially you get to like a worst case scenario across all of them that yeah exactly exactly what I would have in mind. Emma S. 54:12 Something like that. And yeah. And maybe you could say I'm going crazy here but you could save maybe your Emma S. 54:18 Breasts. Worst. And best case those picking mix and choose depending on the Tom T. 54:21 Yeah. Yeah. Emma S. 54:24 category. And how what? You, okay, apologies that was very deviate, but Tom T. 54:31 that, that Emma S. 54:33 Just to be clear. What we're currently focusing on this. Kind of like we're imagining that we want the base forecast to be in great shape and then like, Tom T. 54:40 Yeah. Emma S. 54:41 like, Emma S. 54:42 What can we do to allow to model these assumptions on top? I'm not saying that it's out of the question that we wouldn't be able to add like, Emma S. 54:52 Something here as well, right? Emma S. 54:54 Increase by 10 percentage or something like that. We had this Tom T. 54:57 Yeah. Emma S. 55:00 Notion of It's not working right now. I apologies. But recurring for us entries Tom T. 55:03 but, Emma S. 55:04 that you could just say, Hey every Monday, I collect this. It grows by this presented month of a month or week or a week. So that's all something where Tom T. 55:10 Yeah. Emma S. 55:15 exploring a little bit. Tom T. 55:16 Yeah, that makes sense and also see that the single forecast entry there is is that would definitely be an important one. So there's there's things that happen on a Emma S. 55:24 Yeah. Tom T. 55:26 On an ant, maybe on an even on an annual basis, but not necessarily within the same month each each year. So those kind of things like corporate tax, basically things like that or vat reimbursements, whatever it might be, but we have assumptions again from our budget process from our planning process where we have somewhat semi-accurate, understanding of when those are going to hit at the moment. Tom T. 55:51 I don't. Tom T. 55:57 to manage on a manual basis. I don't really have the, the ability to do it, but if, yeah, if we have that. So, like, on a yearly basis, we have like a one-time or maybe every six months or so, we have like a refresh of those. Those inputs. Tom T. 56:11 That would be a big, big one. So you can kind of see when we're expecting those. Tom T. 56:17 Less frequent periodic debits or credits to come. Emma S. 56:22 Makes sense. Tom T. 56:23 Yeah. Emma S. 56:24 but, so for the collections, Emma S. 56:29 Just to be super clear, your current like the current collection vocals are you would potentially upload to palm. It's basically a number like one number for Tom T. 56:39 One number. Yeah. Emma S. 56:39 them. Emma S. 56:41 And then you would like, or Emma S. 56:44 What would make logical sense is somehow distribute that number evenly based on the whatever there was last January if it's January now. Tom T. 56:50 Yeah. Tom T. 56:55 oh, Emma S. 56:56 Or. Tom T. 56:57 yeah, yeah, I I think it tends to be Tom T. 57:01 So yeah if thinking if like in an idea, let's say in an ideal world if there was the ability to take our November number, for example, upload that into the system is like our expected collections. Over the course of November that's like our number to am at but then we have the system learning from Probably not even last November, but maybe even like the last three months or something like that to say, what the trends are in terms of the working days that we'll be. We actually collect those figures that will probably be the best. And then to layer on top of that initial number a percentage. So whether it's in Tom T. 57:36 that other screen that you were showing me but say, Okay, we think we're going to collect 20 million. That's take this down by five percent just so we're comfortable, that would be in an idea. Well, but actually based on what you have Emma S. 57:45 Cool. Tom T. 57:47 now is already a hundred times better than what you're showing me here because what I would expect to do then is instead of having Tom T. 57:54 That one number, what? I could input as like the external forecast is just what we do at the moment which is a single number. Like, we have that 20 million number but I just spread that across the number of working days that we have within the month and that's what you would probably then see as the confirm. Tom T. 58:11 Well, yeah, confirmed forecast, I guess, I in this, in this example, but then we would have the system base forecasts in addition to that, which would drive my probably drive closer, my decision making when it comes to actually how much we think we're going to collect. So even if it's Tom T. 58:26 Further away from the number because it's more based on historic flows than the AR number that we have. We still have the timing of when we think we're going to collect more within the month. Emma S. 58:36 Yeah. Tom T. 58:37 if that yeah so there's a way that this this still works for us, I guess just saying it that other that first option and an ideal world Tom T. 58:46 Would probably be preferable but it's not again. That what you're showing me now is is a million times better than without. We do things At the moment. I said a Tom T. 58:54 hundred times. It's yeah, it's a lot more than that. Emma S. 58:57 Alright, super cool. Emma S. 59:01 Was there any? We I think we have at least covered. Tom T. 59:07 The agenda? Yep. Emma S. 59:07 The main. Yeah, the points of the agenda. So and I know, I know we're out of time. Emma S. 59:14 But this was super helpful. Tom T. 59:16 Good notice. It's very interesting for me to see as well. Like, I've gone Emma S. 59:17 For sure. Tom T. 59:21 through aspects of this with Christian if you've gone through the setup, but also having in mind like what you're working on in the background to, to build things out, is really interesting for me to see and it's, it's good for me, I guess. Tom T. 59:32 To know what's coming in the pipeline. But also then from my own internal discussions, I can say like in in a few months time, we might have access to like a reporting model that we can start using or that this updates to come on the forecasting side. Yeah, that's really good. Emma S. 59:47 Typical. Emma S. 59:48 All right. Tom T. 59:49 It wasn't. Emma S. 59:50 All right, you're giving me much to think about. So thank you for that. Tom T. 59:54 No pleasure, I'm happy to do more of these sessions. Like if, if you, if you ever need to pick my brain or anything like more than happy to do that, no problem. Emma S. 1:00:01 Sounds great. I appreciate it. Tom T. 1:00:03 Awesome, cool. Well Tom T. 1:00:06 Yeah, not, I hope you feel better. I would say before before before you go away and get get some rest this weekend. Hopefully you're feeling better for physically. Emma S. 1:00:15 Thank you so much. Emma S. 1:00:17 All right. Well, see you soon I guess. Tom T. 1:00:19 Yeah, see you soon, right?