Skip to content

Personio - Reporting Discovery - 2024-10-03

Metadata

  • Date: 2024-10-03
  • Company: Personio
  • External Participants: Tom Thorn (Treasury)
  • Palm Participants: Emma Sjöström
  • Type: Customer Call
  • Domain Areas: Reporting, Categorization, Variance Analysis, Cash Forecasting
  • Recording: https://tldv.io/app/meetings/66fe87a8633ec80014b70c55/

Summary

Context

Deep discovery call with Tom Thorn from Personio Treasury about their reporting processes. Tom walked through their cash flow reporting workflow, categorization challenges, variance analysis with FP&A, and vision for automation. This was an early-stage discovery call to understand reporting pain points.

Key Discussion Points

  • Two main report types: cash flow reporting (transactional categorization) and cash/investments reporting
  • Operating vs non-operating split is a key categorization challenge - same vendor can have both types
  • T+1 reporting: manual best-effort on day 1 of month using bank statements
  • Full reconciliation from Workday ERP available around T+11/T+12
  • Data validation takes "way too much time" - reliance on memo fields with typos and inconsistent labels
  • Working with Fintech team on Workday automation with pivot tables
  • Bottom-up forecasting needed for daily funding decisions vs top-down FP&A monthly forecasts
  • Currently maintaining buffers on accounts due to forecast uncertainty

Pain Points

  • Manual categorization using VLOOKUP on free-form memo text - typos and inconsistent labeling
  • Data validation takes half a day+ before variance analysis can even start
  • Can't get accurate data until T+11/T+12 when Workday reconciliation completes
  • Ad-hoc bank transactions are "a myriad of everything" - hardest to categorize
  • Operating vs non-operating split is nuanced (same vendor, different treatment)
  • No forward-looking information in cash/investments reports - all backward looking
  • Buffers held on accounts because forecast isn't reliable enough for just-in-time funding
  • Customer inflows variance is a "black box" - many factors impact it

Feature Requests & Needs

  • Automated categorization on T+1 using ML instead of waiting for Workday
  • Enriched data points (customer name, segment, type) to help explain variances
  • Forward-looking investment reporting, not just backward-looking
  • Daily transactional forecast for just-in-time funding (reduce buffers)
  • Automatic refresh and summary generation from pivot tables

Jobs & Desired Outcomes

Job: Produce accurate cash flow reports categorized by type (operating/non-operating) for management

Desired Outcomes: - Minimize the time spent on data validation before variance analysis - Reduce reliance on inconsistent free-form memo text for categorization - Increase accuracy of operating vs non-operating classification

Job: Analyze variances between actuals and FP&A budget to improve forecast accuracy

Desired Outcomes: - Minimize the time from month-end to variance analysis (currently T+11) - Increase the ability to explain variance drivers with enriched context - Reduce recurring unexplained variances through feedback loop to FP&A

Job: Fund operational accounts with just-in-time precision to reduce idle buffers

Desired Outcomes: - Minimize buffer balances held on operational accounts - Increase investment income by deploying excess cash - Reduce uncertainty about daily funding needs

Domain Insights

  • T+1 reporting: Best-effort report on day 1 using bank statements; full accuracy at T+11/T+12
  • Operating vs non-operating: Key split for management - same vendor can be both (e.g., lease deposit = non-operating, rent = operating)
  • Cash burn focus: As a private, investment-reliant company, understanding runway and cash burn is critical
  • Ad-hoc bank transactions: Catch-all category in Workday that's hardest to categorize
  • Workday limitations: Journal entry lines lack enough info; rely on memo text which is inconsistent
  • Buffer mentality: Keep safety buffers because forecast isn't reliable enough for daily precision

Action Items

  • [ ] Continue building Workday automation with Fintech team
  • [ ] Palm to show reporting capabilities when ready

Notable Quotes

"Way too much of my time is spent on data validation." - Tom Thorn

"The starting point should be variance analysis, right? It should be the review of actual versus forecast. At the moment, that starting point is way way back where we're copying through, we're doing the data validation." - Tom Thorn

"If Palm was able to take our cash flow actuals and categorize those on day one based on its machine learning... that's a huge win. They get that information two weeks in advance." - Tom Thorn

"A system that can be a time saver, provide efficiency and accuracy... it pays for itself." - Tom Thorn

Full Transcript

00:49 Emma Sjöström: That's awesome.

00:50 Tom Thorn: Yeah.

00:50 Emma Sjöström: Excited to to hear more. All right, I'm just gonna start asking you questions. The idea is that I would like to

00:58 Tom Thorn: M.

00:58 Emma Sjöström: start like it's a little bit of a funnel, right? So start a bit broadly just like Describe the old day today and I'll hopefully get to probe a little bit. Unlike what parts you find more. Most annoying way, your biggest pain points are right now. And like, you. You wish and hopes for the future, right? Like how it ideally work out for you.

01:19 Emma Sjöström: So Would you mind starting just briefly walk me through your typical reporting process? Lot to fit. And I recommend that can be different types of reports. Feel free to kind of

01:33 Tom Thorn: Yep.

01:34 Emma Sjöström: Describe all or the most important ones to you.

01:36 Tom Thorn: yeah, so I'd say that the two main ones are firstly we have what we call cash flow reporting and that's really looking at a on a transactional base

01:49 Emma Sjöström: Yeah.

01:49 Tom Thorn: Item by item. We are taking information from bank statements, from our work day, integration all this, this stuff, essentially, going line by line across all of those statement lines to categorize that information, categorize, all of those cash flows and build up a picture of how our actuals felt within the month.

02:09 Tom Thorn: And that categorization is based on the preference from the company side, the management side and I have been a team in terms of how we want to separate things out. the Germany that falls into categories like customer inflow personal related costs to essentially, payroll supplier payments, Missing one VAT.

02:34 Tom Thorn: So tax, basically. And then there is across all of those categories. There is this like somewhat of a separate split, which would be operating and non-operating. So we have some, some supplier payments, certain things are operating related, so that's just the, you know, the normal set normal

02:48 Emma Sjöström: Of course, yeah.

02:50 Tom Thorn: side the business but then you'll have other things like chance to be like at least contract leases but offices that that kind of thing. That that

02:56 Emma Sjöström: Right.

02:57 Tom Thorn: fall under a non-operating. Budget. So that they have this separate

03:00 Emma Sjöström: Yeah.

03:01 Tom Thorn: budgets for both. And the point there is to get a good handle on what is relating to the core business processes and what is actually, you know, whether that be, you know, a system, it could be Google or Amazon web services something like that, that will fall under operating, but you have some invoices that were processing, that fall outside of that.

03:19 Tom Thorn: And the focus is to separate the two, which can be. I mean, talking about, That's one challenge. Can actually be understanding what's an operating customers are not operating costs. We can have relationships with certain supplies that can struggle both of those sometimes. And so, that that can be a bit of a challenge to separate out

03:34 Emma Sjöström: Well, do you have an example of one of those and

03:38 Tom Thorn: Yeah, so an interesting one. I'll go back to leases. What they, the FNA team have generally said is that a deposit. So that generally when you move to a new office and we had a few of those this year, the various offices, the deposit that we place with A, with a

03:53 Emma Sjöström: Right.

03:54 Tom Thorn: partner would be non-operating. Right? So, we're and generally because we're expected to hopefully get that back at some point or it would pay against a future monthly rental. So the that would be non operating the deposit that we place when we move into a new bit office space, the actual ongoing rent within the the office space is operating

04:16 Emma Sjöström: It gotcha, right.

04:17 Tom Thorn: Yeah.

04:17 Emma Sjöström: So you cannot just rely on the vendor as a way to categorize.

04:20 Tom Thorn: Yeah.

04:22 Emma Sjöström: And also, yeah,

04:23 Tom Thorn: Exactly.

04:24 Emma Sjöström: Interesting.

04:24 Tom Thorn: And Yeah, a bit of a bit of a nuance there and that that certainly can raise a bit of a question, but essentially that that's what I'm doing. My point is taking all of that cash information, plugging it into essentially a spreadsheet and then based on certain data points and criteria that we have within that sheet is doing our best at Categoriz.

00:00 :

04:41 Emma Sjöström: Yes.

04:43 Tom Thorn: categorizing things accurately. And then I will report that the FNA team report that to our management team to then the big piece of analysis is then our actuals versus what they have budgeted within the same categories. And then we look at okay why did we why do we miss in terms of forecasting on personal costs this month or Why were our cash inflows down this month? That kind of thing.

05:06 Tom Thorn: So there's the the information load the comparison between budget and the comparison between budget and

05:13 Emma Sjöström: Right.

05:13 Tom Thorn: I'm so I that's that is a really talk about in focus. That's like the big focus of the company at the moment is to get is accurate as we can with that because Is we can with that, because So they come the position, the companies in the moment, still private still heavily reliant on an investment.

05:31 Tom Thorn: We need to have a really good understanding of our lead lead. Wayne in terms of cash burn in terms of where we're going to be in six months time of years, time to understand what. Yeah, you know, we are Rightly burning cash, you know, trying to expand and grow the business.

05:47 Tom Thorn: But it you know, understanding how much we're doing that and how accurate we can be with our forecasting is really important in terms of yeah projects and scalability all this good stuff.

05:58 Emma Sjöström: Of course, would you say? So it sounds to me that you're doing the data digging and, like, summarizing all the information and then is it like a joint effort with you and in a team or management team even also to do all this context or like digging deep.

06:13 Tom Thorn: Yeah, exactly. Is that. It's really a

06:17 Emma Sjöström: just kind of just like, who

06:18 Tom Thorn: Yes.

06:19 Emma Sjöström: Yeah.

06:19 Tom Thorn: So I will take in, I'll do the at the moment, somewhat the legwork in, Maybe I'll say that, that's that's the overall change. The what we're trying to present in terms of the reporting side, What we tend to do in reality is like a t plus one. So, on day, one of a month, I will actually do a best effort based on

06:42 Emma Sjöström: Okay.

06:43 Tom Thorn: our Bank statement information. So there's a few. We don't have Hundreds and hundreds of bank accounts. We actually have a few that are operational and based on the information that I have at hand. So we I mean even mid-month in September. We should have a relatively good at sorry.

07:00 Tom Thorn: Relatively good understanding of what are payroll. Flows were and same with tax. So there's that there's certain things that we have that we can strip out of our operational accounts and operational flows.

07:10 Emma Sjöström: Yeah.

07:10 Tom Thorn: So don't get somewhat of an understanding around. Okay, we we expected payroll to be this in September and ended up being this. It basically through some manual workarounds on T plus one I can get you know, relatively accurate in terms of the split across those main categories. What I can't do is say what's operating, what's going on, operating? So that, that piece comes later, I'm not going to dig into, you know, 2,000 transactions on an individual basis on one on day one.

07:40 Tom Thorn: That's, that's just not good use of time. So that's when we have that t plus one, which is really like a manual process from my side. Takes way too long, not my favorite, but it gives the, the management teams in comfort around where we've how we've performed in the previous month and then the Accounting team will work on the reconciliation within workday and then at a certain point in the month, maybe around T plus 11, T plus 12.

08:07 Tom Thorn: The data in the workday system the reconciliation will be completed then I can start extracting that information from workday and utilizing it for the more. Accurate reporting where I plug all of that information into a file and hopefully it spits out the right numbers, similar to what I reported on t, plus one that makes sense.

08:24 Tom Thorn: Plus one that makes

08:25 Emma Sjöström: Right. How hard is it for you to spot like discrepancies, though? Things that are all?

08:31 Tom Thorn: Difficult.

08:31 Emma Sjöström: How much time do you spend?

08:32 Tom Thorn: Yeah.

08:32 Emma Sjöström: Like, just finding out where you

08:34 Tom Thorn: Yeah.

08:35 Emma Sjöström: people even.

08:36 Tom Thorn: It's, it's a good question and it's one of the main. Again, It's a big focus topic for me at the moment. So a way too much of my time is spent on data validation. So I mentioned there's some data points that we have within workday. Essentially, what we're taking is like the base information that we can get like journal entry lines from our ERP system, which doesn't have enough information to truly.

08:59 Tom Thorn: You know, be as accurate as we would like to. We often rely on say statement lines or memo lines so that's like reference information on a transaction level to then like

09:08 Emma Sjöström: Yeah.

09:09 Tom Thorn: transaction, says the 80. Okay, that's a tax payment. If a transaction says healthcare, that's going to be a personal, It will come under personal cost, but that's assuming all of those payments are made uniform formally, which is not the case is assuming there's no typos, which is definitely not the case or all of this, this kind of stuff.

09:28 Tom Thorn: So it's one of the key challenges for me is and it takes up too much time. So you, I plug this data in something looks weird and I have to go and investigate for a couple of hours, you know, across these transaction around what's not, what's up, putting through correctly in across the vlookups and and if functions all that good stuff, so stuff, Actually the processes is down since I've mentioned is is that very it's labor-intensive and it's not not the most accurate.

09:57 Tom Thorn: What I'm actually working on with our Fintech team. Is to pull through.

10:01 Emma Sjöström: Okay.

10:03 Tom Thorn: It's like a, like, the next step in terms of automating how we do things is actually to have a live connectivity with our workday function. So I work day ERP, and then I can ref

10:14 Emma Sjöström: Yeah.

10:16 Tom Thorn: refresh that and it's all, it's all LinkedIn. There's no copying and pasting across, you know, thousands and thousands of lines. Good one but also for each payment category, each type of payment that we make when I say that, I mean Within workday with within the way that that system is set up.

10:34 Tom Thorn: There's different types of categorization of payments that they're making, so it could be something gets categorized. As a supply payment, something gets categorized as a customer payment. Those are relatively easy to spot right from my side. When I'm trying to pull that information across that data point fine, I can see that in a column, no issue.

10:50 Tom Thorn: We know what those types of payments are. There's other more, there's other payment types which is just fully ambiguous and they tend to fall in the ad hoc category and and you will have That that could be any type of payment. It could be a supplier payment, there's just been supplementary.

11:05 Tom Thorn: It could be a return transaction or where we have to

11:08 Emma Sjöström: Right.

11:09 Tom Thorn: Rerun a payment to a supplier, something like that. It could be personal cost. It could be at least payment, you know, all of these things for tend to fall under what they call adopt bank transaction. And that's just a myriad of everything that's where the big challenge comes in, but what I've been asking working with our accounting team, Is to start utilizing additional data points within workday.

11:32 Tom Thorn: So instead of just having ad hoc bank transaction, this should then be a supplementary data point to say bank fees, or investment income or

11:41 Emma Sjöström: Labeling, the transactions of Some

11:43 Tom Thorn: Exactly, exactly. So it take, it takes away the reliance on free free type text, you know, you don't you're not just relying on the payment reference or memo, memo text, you're actually relying on a Like drop down data point that they have to select when they when they book a transaction.

12:00 Tom Thorn: So when they're going through their reconciliation, it basically. Does a lot of the instead of having, like, rows and rows and reels, and reels of VLOOKUP trying to search for all of these various different buzzwords, you actually just have a column that is very clearly defined. What? That type of transaction is yeah.

12:16 Tom Thorn: That type of transaction is yeah. That type?

12:17 Emma Sjöström: It's like a human intervenes in the process and explicitly labels.

12:21 Tom Thorn: Yeah.

12:22 Emma Sjöström: Okay.

12:23 Tom Thorn: Yeah, it would always be so for certain transactions. So, for example, supply payments and

12:28 Emma Sjöström: Yeah.

12:28 Tom Thorn: that that goes through a payment run, it's but pretty much automated like straight through processing. It has stage gates for approval, all that stuff, but that those ones get pretty much automatically reconciled. There's no issue because it's a push from our side to work day will be information in there, but when it comes to ad hoc functions or these more ad-hoc type payments, it's then something that the Team, the accounting team will step in and book and post themselves.

00:00 :

12:54 Emma Sjöström: I, I

12:55 Tom Thorn: And at that point, they would have to select a purpose type.

12:57 Emma Sjöström: Should.

12:59 Tom Thorn: Yeah.

12:59 Emma Sjöström: You right? They've already investigated. So why why double why do double work like if they yeah?

13:03 Tom Thorn: Yeah. Yeah.

13:04 Emma Sjöström: sense. Very cool. All right. So like if If categories were not an issue, you were able to pull categorized transactions already. But what would be your next step then like what? just asking because it's because I,

13:22 Tom Thorn: The neck. Yeah, no.

13:23 Emma Sjöström: you but it's a big pain point, just

13:23 Tom Thorn: No definitely.

13:25 Emma Sjöström: the categorization but once that is

13:27 Tom Thorn: Just got Advisation.

13:29 Emma Sjöström: done like what's

13:31 Tom Thorn: Yeah, so that the next stage for me would be to go further in times in terms of refinement. So, essentially, what we would be able to do is we would have a basically, The way it's been out of the moment is across a series of taps. So each one of those tabs represents a payment type.

13:51 Tom Thorn: So supply payment, customer inflow up bank transaction, you know. So on what I then love to be able to do is you actually you have to refresh all of those separately and then off the back of those, I feel pivot tables to sort through the data and bring through the categorization type.

14:08 Tom Thorn: So, bank transactions, I have a pivot

14:08 Emma Sjöström: Right.

14:10 Tom Thorn: table for bank fees and for investment income. And so you get what you need and then you have that stood out month. The next stage is to essentially. Troubleshoot that because there are still some some tweaks that that are needed. But then actually have that File work in unison and in parallel with the very manual one I have at the moment to make sure the right information comes through.

14:33 Tom Thorn: But to your question, the next step with MP2 work with our Fintech team to have all of that information automatically refresh. It doesn't even need me to be in there and clicking through it. And ultimately, it would be good if they could help me. I don't through a macro or something like that.

14:50 Tom Thorn: Take the refresh information from the various different pivot, tables, and all that automatically into a summary, a summary sheet. She That would be the hope. So then you'd be able to say month or month categorization, all of these different and you can see the variances that when the actuals but essentially that would then get us to a point where we're automatically replicating, the manual file which is taking forever to build at the moment.

15:15 Tom Thorn: You would have that, hopefully very clear. The final stage would then be. You know, if there was a way to do it to plug in our forecast information as well.

15:25 Emma Sjöström: Right.

15:26 Tom Thorn: able to build a delta to your forecast information that that what that should be at the moment. That that what that

15:31 Emma Sjöström: So, when you say Del do you mean?

15:33 Tom Thorn: Yeah.

15:33 Emma Sjöström: the variety of Ryan's

15:34 Tom Thorn: There's a variance. Yeah. Parents yet.

15:36 Emma Sjöström: Couple.

15:37 Tom Thorn: linger and but yeah you'll be able to then it you know much earlier within the month and without too much reliance manual reliance on someone manually, clicking through or copying and pasting, whatever that you'd be able to get to the starting point, which is actually the starting point should be variance, right? It should be the review of actual versus forecast at the moment.

16:01 Tom Thorn: That's starting point is way way back where we're copying through. We're doing the data validation. So that that would be the hope is that we just get to a point where we can refresh. Refresh information have all of the pivot tables automatically, all of those pivot pivot, tables that automatically, populating, the summary sheep and then produce a variance to say This was your personal costs forecast for September.

16:28 Tom Thorn: This is what actually happened. That's going to have a look the actual benefit of that that file as well as. I don't know if you call but I mentioned on a manual. I'm copying pasting just really, really basic information across information. Now, we have things put out within that automated file in a way where we can bring through enriched data.

00:00 :

16:44 Emma Sjöström: Yep.

16:45 Tom Thorn: So it's helpful for me from a categorization point of view. But then it should also help with the analysis. So we can see, I can actually work closely with the NFP and a team to say it will. For example, it will include supply pay supply names, it will include customer names, It will include what type of customer category that they fall into, which should then help great deal with the actual analysis side.

17:08 Tom Thorn: At the moment where we have a variance that we have to go and dig into each month, but actually off the back of this automation of this new build, it should allow us to It should Should allow the effing 18 to get a much quicker. Much better handle on what was the driver of that variance? Basically, based up based on

17:25 Emma Sjöström: Right.

17:26 Tom Thorn: additional data points,

17:27 Emma Sjöström: Right, right? So you're saying to add more context where it's very helpful because other

17:31 Tom Thorn: Yeah.

17:32 Emma Sjöström: otherwise they would need to go and dig that up somewhere else.

17:34 Tom Thorn: No, no, yeah.

17:35 Emma Sjöström: And that, I guess, because human memory is not perfect. Even if only before you're gonna have to go and do it again at some point. Maybe like I'm just off, like this was a type of mix, but maybe there's more types of

17:47 Tom Thorn: Yeah, so no, I, I guess that it's not not entirely accurate but put it this way. So at the moment, if I was to take the information from the the data that's available and and works for our manual file, you may just have an invoice number in as a as a reference.

18:06 Tom Thorn: But if with this new automation, you will have a whole page. The whole tab essentially, dedicated to that. Customer invoice. I'm able to pull my information. Oh, sorry about that. Customer flow. I should say.

18:17 Emma Sjöström: Okay.

18:18 Tom Thorn: that. I get the information I need for my inflows but not it won't just have the invoice number. It will also have the customer name, the customer, the type of customer they are whether they're subscription based or renewal based or what might

18:35 Emma Sjöström: Right.

18:35 Tom Thorn: what what market segment they fall into all of this good stuff which then helps our FN 18 identify hopefully quicker and more accurately. What was the driver of the variance that the high level?

18:49 Emma Sjöström: That's very cool.

18:50 Tom Thorn: Well, that's the plan that yeah, if

18:52 Emma Sjöström: Right.

18:53 Tom Thorn: Yeah, if we can have that, it would be helpful. Yeah.

18:55 Emma Sjöström: So first like digging into all the at least so now you're at the point where you can actually identify very uses.

19:03 Tom Thorn: Yeah.

19:03 Emma Sjöström: Right. And then when you start digging into that you you've done this way, it's where you provide this additional context. So you can help provide explain why the most stuff like involved in this process of like finding the next investigating, why? This went off the rails like yeah. This went off the rails like yeah.

00:00 :

19:20 Tom Thorn: So that the I guess the the endpoint is then so that this is all driving towards a really all encompassing month-end report, they then gets fed to our finance director. And sometimes to our CEO and to explain the variances between all of the stuff and the key, the key question will always be.

19:39 Tom Thorn: When we see a variance in forecast, is that offsetting You know, so if we have a treat like a supplier prepayment that we're making in at the beginning of a quarter and that's gonna last for, you know, that's paying from. So we make the payment in July but that's covering invoice in invoice value for July, August and September.

00:00 :

19:57 Emma Sjöström: Yeah.

19:58 Tom Thorn: and our forecasting team had had split it out over those three months. So you have a really big upfront payment and then no payments for the second, two months that kind of insight is helpful to have, but it's also key for that team to say, You know, it's a bit of a relief, I guess to say Oh we've identified that this, you know, this big prepayment was the cause of our variance but it will offset in future months, you know it.

20:20 Tom Thorn: Well we will still be accurate net at the end of the year. It's good. It's helpful to be able to identify those types of variances versus actual. We hand forecast, this payment it was a surprise. That's cash out the door with it, we're not going to get back.

20:33 Emma Sjöström: Of course, of course, you can count

20:34 Tom Thorn: Yeah. So

20:36 Emma Sjöström: it back home some way. Oh like yeah.

20:38 Tom Thorn: Yeah. Yeah.

20:39 Emma Sjöström: That's always expression. I don't know.

20:39 Tom Thorn: I

20:40 Emma Sjöström: I try to directly translate to English, but yeah.

20:44 Tom Thorn: Yeah, and

20:45 Emma Sjöström: so,

20:45 Tom Thorn: And and the other I guess, The other point of that. So that's just feedback and reporting to our, you know, the standard stuff. But what it should help to drive is a more accurate forecast as well. So as we get more learning, and we get more understanding of how our cash flows are working.

21:00 Tom Thorn: We're able to connect more with our AP team or AR team with our forecasting teams and stuff. Building that that common understanding of how cash flows are working it.

21:10 Emma Sjöström: Right.

21:10 Tom Thorn: It should in time, it should be like a Actually, right, that it's going going to our fb-18. So next year, for that same invoice, they don't, they don't Plan it over three months cycle. It's actually planned as a prepayment, so we can be hopefully more and more accurate and drive down the variance values that you see.

00:00 :

21:27 Emma Sjöström: Cool. So like how much forecasting are you is like Treasury producing at this moment?

21:32 Tom Thorn: Yeah.

21:33 Emma Sjöström: Much.

21:33 Tom Thorn: So so That I would say that's like the

21:37 Emma Sjöström: Yes.

21:38 Tom Thorn: group. Well, that's like the top down forecast, and I'm part of a bit of a cog in that wheel in terms of then the actual what I would be interested in from a Treasury, point of view, is way more on a transactional basis. So they're looking at customer flows, like, they're being big numbers on a monthly basis.

21:57 Tom Thorn: My myself in order to move cash and funding across the business, accurately and efficiently. I'm I want to know at a very granular level level When these payments are being made within the month. So, for example, if you have a payroll payment or you have a payroll outflow of 10 million for the month, Right.

22:19 Tom Thorn: That would be what I would see in the budget side. Something like that. But when is that payment being made in what currency from what accounts and actually that payment isn't one 10 million payment. It's a series of much smaller payments that can happen from the first of the month, all the way up to the 30th and it can happen a number and a number of different currencies or from a number of different bank accounts.

22:41 Tom Thorn: And so then I what treasury to? Well I would tend to be more involved invested in is a bottom up forecast. So actually taking invoice data or it confirmed payroll information and building up a transactional base forecasts based on you know, on the input where I can be comfortable and understanding actually not just the 10 million but each Bit of that 10 million that's made up.

23:06 Tom Thorn: Where is it coming from? Where is it? And when, What is making up that, that value and where? And when is it, basically, so that that's actually something that I tried to build up over the last few months again, very, very manually at the moment and that's something and And By most companies, I feel but we tend to run a buffer on our bank accounts.

00:00 :

23:29 Emma Sjöström: Right.

23:29 Tom Thorn: So we, we maintain a buffer because

23:30 Emma Sjöström: Yeah.

23:32 Tom Thorn: worst case scenario, I get something wrong within my forecast, we don't want to be in a position where we're not me. We're not able to make a payroll payment or we're not able to make a supply payment but that inherently is inefficient, right? So if we have a more accurate and reliable forecast, we can drive down the buffer, the buffers that we have and then we probably get a, should get a benefit in terms of interest income as well.

23:52 Tom Thorn: As We can invest more over that cash, that's currently on the buffers, we can put that within a interest-bearing account, or a deposit, or whatever, it might be, it's a benefit. So, that's where some Thing like, Palm comes in. Is that yeah.

24:04 Emma Sjöström: Yeah, I was going to say this, you internal pitch, right?

24:08 Tom Thorn: Yeah, no it is it definitely is. That's that's basically where that so much and be so accurate. that comes from is that I can only do

24:12 Emma Sjöström: Yeah.

24:16 Tom Thorn: harm has all, you know, Treasury systems have is some machine learning or they can actually pull in invoice information. So, for example, a good. Yeah, a good example of this is I hold a buffer at the moment. We have a budget to supply payments. It's that forecasting piece we have We 10 million for that.

24:34 Tom Thorn: I'll use the example again, 10 million for supply payments across the month. I have to assume I essentially take that number. I do some checks from previous months in terms of what we paid out for GBP, what we paid out for USD and then,

24:47 Emma Sjöström: Right.

24:48 Tom Thorn: And then another, I split that out. So then I have like, like roughly my Euro GBP and USD amount.

24:55 Emma Sjöström: Yeah.

24:56 Tom Thorn: And then I aggregate that, I customer. That's the best I can do, right. I don't know exactly. We'll have some big invoices within there. We'll have some like 500k invoices, maybe a 1 million invoice. I don't know exactly when they're going to be paid during the month. So I stood it out equally across the month, across the working days and

25:13 Emma Sjöström: Yeah.

25:13 Tom Thorn: that that's where you have to have a buffer because if I was to try and run a low of value on my bank account, I could be called out to say If we have 10 million and seven million of that is euros that I've assumed for the month and then I split that out over 22 days.

25:28 Tom Thorn: Your daily supplier. Payment value is no a few hundred thousand. If I just maintain that on my operations bank account and we have however much goes through and one particular day. It could be multiple, big, big supplier invoices. I don't have enough. Whereas, if I have the Transactional, kind of really bottom up, so you have invoices feeding into your system to say, we know we're going to pay Amazon.

25:51 Tom Thorn: We know we're going to pay Google on the 10th of the month, but around those days we have very little in the way of supply payments. I can organize my my funding in a way that We get the funny just in time.

26:01 Emma Sjöström: Right.

26:01 Tom Thorn: essentially for those big payments but then we don't overfund and we don't hold the buffer around it. Unnecessary.

26:06 Emma Sjöström: How would you define just in time? Would it be like on a month? Like with a monthly granularity? Be okay. Should be weekly. Like how? Yeah.

26:15 Tom Thorn: I would define it daily actually.

26:16 Emma Sjöström: Daily. Yeah.

26:17 Tom Thorn: Daily. And that's particularly for a company like us, like, I, we're not so complicated, to be honest, but we there's, there's not endless, endless bank accounts and endless endless currencies. Uber did have that which made it challenging, but we should be able to be accurate enough that we can say how much we need on a daily basis.

26:40 Tom Thorn: It's not the there's a couple of I guess caveats to that, you have tax payments that are often direct debit and so they can fall within a window, right? They maybe the tax authority say we will debit you either on the 15th of the month or at least within five.

00:00 :

26:57 Emma Sjöström: Then.

26:58 Tom Thorn: days. So you get a window. That's generally a bit frustrating. It's not the best because you don't know exactly when it's going to do it. But for the most part you can have an idea of when you need to find your account and when when the balance is need to be in there, and even if you talk, Like a couple of extra days worth of fun, like not holding the buffer on your bank account week, on week and month and month that was up to quite a big.

27:21 Tom Thorn: Saving so I it should really be daily.

27:25 Emma Sjöström: Super interesting. I would love to talk more about that as well, but ultimately, I have to like go back to reporting that. It's really interesting how

27:31 Tom Thorn: because,

27:32 Emma Sjöström: everything is so interconnected.

27:34 Tom Thorn: It is. Yeah but again it's it's I guess it's just modern way it's but when I started it was definitely weekly. It was definitely longer term than that. But now it's just the the availability of data in the the Upsc

27:47 Emma Sjöström: Yeah.

27:48 Tom Thorn: upskating of systems really helps you to build out more accurate forecasts, where, you know, they there's really no excuse, if you have a good system in place to not do that. The only excuse I guess would be capacity of things, that so complicated that you can't look at every bank account, can't do that every currency or whatever every single day, that might be an excuse, but I actually, for our company, it's not we're not in that position yet.

28:11 Tom Thorn: So we should be able to do daily and pretty accurately.

28:14 Emma Sjöström: Very cool. Looking forward to hear more about that.

28:17 Tom Thorn: Yeah.

28:19 Emma Sjöström: I'm gonna so you've been very like informative actually. I'm thinking

28:25 Tom Thorn: I'd sorry, I I know we

28:26 Emma Sjöström: No.

28:27 Tom Thorn: we've focused on that was the cash flow reporting. I was mentioning at the beginning. So that's the budget one. There is another one that I'm doing which is really just more based on cash and What I'd say is more core. Treasury reporting is actually cash and investments like where we hold our cash.

28:43 Tom Thorn: So our bank balances are deposit. Balances the rates that we're getting the currencies that we're holding them in. I'm taking those from me we. Yeah, I generally taking those from bank balances, you know, reports from our banking systems and then plugging those into the spreadsheet and pulling those out.

28:59 Tom Thorn: I'm doing that on a monthly basis and then adding some additional context. So, if there's, I know a great cut in, in the Euro ECB, right? Or the Fed, they're cutting interest rates. That's gonna impact our investment return month. So just calling out pieces like that. That that's I wanted to point out.

29:15 Tom Thorn: So that's you have your cash flow reporting which is really more of a forecast versus actual. I'd say, budgeting fb&a thing which I which I I'm I have input into and and contribute to Where's the cash investment reporting is more. I'd say the Treasury related piece, the stuff that we We produce ourselves? Yeah.

00:00 :

29:38 Emma Sjöström: Interesting. But both like information from both of these could be presented or like

29:44 Tom Thorn: Yeah, yeah. Yeah.

29:45 Emma Sjöström: and Yeah.

29:46 Tom Thorn: drivers so the the

29:50 Emma Sjöström: Yeah.

29:51 Tom Thorn: at Cashburn and how we performing against our forecast the second is really looking at what we're doing with our cash and are we being as efficient as we can be. efficient as we Yeah.

30:03 Emma Sjöström: Super nice. And And just so, I understand like, because this whole all this context and like all, why and all of those parts and how is it at your company? Like, are you stakeholders very probing or like if how how do you feel, you know, before you spending

30:23 Tom Thorn: Yeah.

30:24 Emma Sjöström: out these numbers? And you've added as much, I guess context as you can so that they will. And like how does that hold dance?

30:29 Tom Thorn: Yeah.

30:31 Emma Sjöström: work for you guys and like, And

30:33 Tom Thorn: I'd say the. They detail is is pretty high here that or the focus on detail is pretty high here. Like the materiality I would argue is higher sorry, lower here or the materiality thresholds we say is lower here. Then I've experienced in previous companies and but that's also in the context of where the company is in terms of its growth and and where it's going.

30:58 Tom Thorn: So yeah I'd say there is a back and forth. Certain things get called out where I would consider that like a immaterial value, but they want to understand. So you have to go digging in a little bit more. I think there is generally an appreciation that we don't have the reporting, the information, the data that supports all of this reporting in a really good situation yet.

31:23 Tom Thorn: And a lot of it is kind of Bandages and and ad-hoc pieces of analysis that drive and give context. And so that I think there is an appreciation often that we can't cover each and every euro of variance you know we had there has to be often times. That's something that's going to go on a list for review at a later point in time.

31:48 Tom Thorn: But there is a pretty in terms of team-wise, there's pretty big effing 18 who are engaging with the the broader broader business to understand, forecast, values, and trends, and all this kind of stuff. So, That's what they're tasked with is being as accurate as they can. So it kind of makes sense of effort.

00:00 :

32:04 Emma Sjöström: Yeah.

32:06 Tom Thorn: They they want to be reliant on the numbers and be it. Well sorry. They want to be able to rely on the numbers.

32:11 Emma Sjöström: Yeah.

32:12 Tom Thorn: understand it. It can be a challenge particularly when I joined uber was Had much much higher thresholds. And so you're not getting quiz, questioned or quiz on some of the things that we we are here, but it's also just that. Yeah, the nature of things, I wasn't cash a reporting for sure.

32:27 Tom Thorn: Who's not something I was involved in in, I do.

32:30 Emma Sjöström: All right.

32:31 Tom Thorn: But I said just getting used to how, how deep things go? There can be a challenge but generally that they're pretty happy. If yeah, if you have variances and supply payments, and you can explain the majority of that. Okay. But, you know, we did, we did a good job.

32:45 Tom Thorn: One of the challenging ones at the moment is understanding customer inflows.

32:49 Emma Sjöström: Okay.

32:49 Tom Thorn: and the variances that we we see between our forecasts and our actuals there, because Yeah, you go, you go. It's bit of a black box as soon as you you go into this so many things that can impact it from my, you know, from a transactional timing point of view, it can impact it from Yeah.

33:10 Tom Thorn: Name the name, the nice is huge number of things that can impact to it in terms of customer segment. Market segments, renewals information, their assumptions around those renewals all different types of stuff that can impact their forecast. stuff that can impact their forecast. Then tying that up to what they actually see is is a challenge that's one of the bigger focuses.

33:30 Tom Thorn: So we we have the cash flow reporting focus over the last few months to get that into a better shape, but the priority within that project is being customer imposed. Inf

33:39 Emma Sjöström: Yeah.

33:39 Tom Thorn: Yeah.

33:40 Emma Sjöström: Wow. Thank you. I'm gonna ask you some like A very open-ended questions. Again, I'm just going through, you've

33:49 Tom Thorn: so,

33:50 Emma Sjöström: You've covered a lot already. So I'm just And the right? So like, If you could improve one part of the reporting process, what would it be?

34:06 Tom Thorn: The. can I have to I I would say they kind of go that like this kind of go together but accuracy and automation of the data

34:18 Emma Sjöström: Why?

34:19 Tom Thorn: yep, 100% 100

34:26 Emma Sjöström: What would you? Okay, So it's a little bit repeating but it's really nice to just get Like What them what what does success Like What them what what does success reporting? Like it could be

34:38 Tom Thorn: Yep.

34:38 Emma Sjöström: Accuracy, stakeholder satisfaction other things. Like What how would you know like Ah This was a successful reporting cycle or like or didn't hate it, you know.

34:46 Tom Thorn: Yeah.

34:48 Emma Sjöström: like

34:50 Tom Thorn: I I would say what success should look like for me is that the the time is spent on the variance analysis. Not the data validation. So time spent on date on the variance analysis, that is then reported on and the variances understood accurately understood in communicated. And then the feedback works, you know, if there's a an issue or an update that needs to be made within our budgeting engine that thing gets made to.

35:15 Tom Thorn: It goes back. It's basically a circle.

35:17 Emma Sjöström: Right.

35:18 Tom Thorn: This is, we're taking, essentially, we're not spending time on data, validation and clean up. We're spending it on the analysis, we understand why our variance was caused between actuals and and forecast. And then we can act upon it and so next time it comes around, it shouldn't be an issue.

35:33 Tom Thorn: So over time, if you have a number of different periods a number of different cycles like that, then, at some point you get to really, really reliable stable forecast

35:42 Emma Sjöström: Yeah, makes sense makes a lot of sense. So, would you say there are any like Metrics or type of feedback that you wish you had to know, kind of reporting works or meets expectations for you or for your oldest.

35:59 Tom Thorn: Yeah. That's a good question. I guess it would probably be the number of comments that we get, if, you know, if we getting feedback or So successful, in terms of the actual reporting. The, in terms of the stakeholders are receiving it. If they're comfortable with what we're giving them and they they don't have comments, something doesn't make set.

36:20 Tom Thorn: That's right. If they if they have, if they have comments or something doesn't make sense to them, or they want more information, we haven't set the context, right level. We've missed something within the information that they want would like to see. I guess that's a metric. It's a bit of a Broad one because you know, it could change month to month, you know, people have more time to review a report or something like that but overall I'd say those metrics.

36:45 Tom Thorn: And then I guess over time it should be reducing the variances. So I guess it goes slightly into the previous answer. I gave but over time, you don't have so many varians because lessons have been learned from previous reporting cycles.

37:01 Emma Sjöström: If that's great. Just taking some notes here. But so how how much time do you currently spend? You say just to the point the way you can start the variety.

37:19 Tom Thorn: yeah, I I would say started to be

37:21 Emma Sjöström: much like

37:24 Tom Thorn: able to actually start the analysis.

37:26 Emma Sjöström: Huh.

37:27 Tom Thorn: I would say, like, Half day. Something like that that's probably conservative. It can be more than that. The review and the variance understanding can then take x and up, as I say, the moment, it's the focus is being on. Understanding the variances and like, cash inflows and that is taking up.

37:49 Tom Thorn: I don't know how much of my time recently. The hope is that it's like a I guess with a lot of these, the hope is that it's somewhat of a one-off effort. Even if that one-off effort is gonna take two months of your time or something like that of intense review and and analysis, the hope is that you did lessons that you learned from that process, then do get fed back into their forecast and you don't need to go through it again.

38:12 Tom Thorn: That's, that's what I've got to clicking on to but yeah, that I'd say the just to get to a position where we're comfortable with the the data half a day plus probably

38:23 Emma Sjöström: Yeah.

38:24 Tom Thorn: And then probably more than that actually. And then the Yeah. The the review and and understanding of the variance is being caused particularly for some of those line items, those categories. Yeah. X amount at this point. X.

38:38 Emma Sjöström: Yeah, and when you say also, I want to ask you about that like for the caching clothes like lessons learned. Is it for you? Like your intuition that's developed? Or is it like something that can be shared? Like how would you

38:52 Tom Thorn: Yeah.

38:53 Emma Sjöström: You share and keep all these lessons. Like so for the next time? Yeah, like you said it will be a smoother process.

38:59 Tom Thorn: So it's, it's not necessarily for me, but it's for the, the forecasting. So, the FB and 18, essentially, should be taking those lessons in into account. So, for example, a couple of, well, a couple of examples around that is that they look at each month. Like, it's the same.

39:14 Tom Thorn: So they approach January February, March April, June, regardless of the month, they're looking at it? Okay, that's a freshman. That's, that's how things are going to work. Where From my side of things, there are some various to that, right. You have bank holidays. You have weekends that fall the big one for us.

39:32 Tom Thorn: At the moment, is weekends that fall over a month and so you might have the 30th and 31st of March, for example, we're a Saturday and Sunday.

39:42 Emma Sjöström: Right.

39:42 Tom Thorn: We, we have invoices and invoice invoices that are dated for those days. It's very unlikely that we're going to be able to get cash inflows for those days. So when you come to look at it from a from a forecasting point of view, we're completely off, right? We've got we're missing two days worth of collections which is pretty significant.

40:00 Tom Thorn: But for my sides, okay, we should never have expected to receive those. That should be built into the forecast in terms of just like a date, assumption or something like that to say that you know, the invoices from those days, we should just be assuming that they're always going to be collected in April, right?

40:17 Emma Sjöström: Yeah, that makes sense.

40:18 Tom Thorn: Yeah yeah but it's it's kind of a It's like a two-way thing. I had to get trained a little bit on their cash engine to then be able to point that out. And then there's there's a there's a lot of other things similar to that from my side which are just like almost like timing to say.

40:31 Tom Thorn: It's not that we're not gonna receive this cash but we're assuming it's going to come in incorrectly. So there may be So there may be say another example, say We have a customer who renews their package and maybe up updates that package so they You know, they can actually use a benefit to us, they're gonna pay us more but the way that we run that kind of like a unique way of doing it.

40:55 Tom Thorn: Is that from a billing and invoice point of view, we would give them a credit And then we would send a new invoice to them, but that new invoice. Say the original invoice was due to be received in September. The new invoice actually may have now a date with October.

00:00 :

41:09 Emma Sjöström: Okay.

41:09 Tom Thorn: So we get we get more cash in but it's later.

41:12 Emma Sjöström: Yeah.

41:13 Tom Thorn: and so that, that kind of Shift isn't necessarily then appreciated within the cash engines, say Where is that invoice? Why didn't they come through? It's not going to come through, but it's been rebuilt and it's going to come in later. And it's So essentially providing that kind of insight to the FB and a team.

41:30 Tom Thorn: There's then the AR team who are also working on their side to provide some information to the FNA team, the FB in a team are then looking at their side in their data. To make the actuals, basically from their side, more on a granular basis, around each customers, renewals all this kind of stuff basically, to over time tweak and improve the cash cash, inflow engine.

41:50 Tom Thorn: They call it where it gets more accurate and we can be more reliant on it.

41:54 Emma Sjöström: Yeah.

41:55 Tom Thorn: Yeah.

41:55 Emma Sjöström: Very cool. Thank you. I mean, there are We've got recovered, most things. Most But then so you said you like we've been talking about validating data and this is gonna sound made to you, but I just want to dig everything out of you so like why? So Why would you want to validate data, right? What? What could happen if you don't like, What are you?

42:25 Tom Thorn: Hmm. Yeah.

42:25 Emma Sjöström: Yeah.

42:25 Tom Thorn: is in terms of the the manual flow. I have sorry, the manual file I have at the moment say we have a say we have Corporate income tax is a good example. So say, for example, the accounting team make a tax payment and they label it as tax, that could be v18, which we would consider operating, or it could be corporate income tax, which is considered non-operating.

42:51 Tom Thorn: So if if we If we don't validate, I don't go through each and every line for sure. But if I if we're able to spot big flows like that, which don't, you know it in a general. Like I've seen it all throughout the year, the tax decide, the tax indeed.

43:08 Tom Thorn: A P team process. These payments with all different types of labels. Depending on, who's doing it? That's awesome.

43:14 Emma Sjöström: Okay, so outside of the human

43:14 Tom Thorn: improvement and but they

43:18 Emma Sjöström: intervention, right?

43:19 Tom Thorn: Yeah, for sure. Yeah. But they Well the set so maybe to go back so that they're my manual files rely on that memo field, right? Which is like a free form text

43:31 Emma Sjöström: Right.

43:31 Tom Thorn: Which is like a free form text. Someone might put corporate tax in there. Someone might put cit, it's very difficult when we're doing, like a VLOOKUP and based on, based on that, when they're using different.

43:38 Emma Sjöström: Yeah.

43:41 Tom Thorn: every time. The update. Yes. Pain. But the update from them and the move to the automation files, that they, I don't really care what they put within the memo field because supplements, supplementary to that. There should be this drop-down field to say this is a corporate income tax payment.

43:58 Tom Thorn: payment. That that sorts out that validation

43:59 Emma Sjöström: Yeah.

44:01 Tom Thorn: piece, but it's the reason that I'm

44:02 Emma Sjöström: Yes.

44:03 Tom Thorn: having to dig through the data at least a little bit at the moment to make sure that for some of those bigger flows for categorizing them correctly really between operating and non-operating. Similarly, it can quite easily happen. That We have Hostings that are no one knows what they are.

44:24 Tom Thorn: So going back into earlier in the year, you might have Big, big flows, like return transactions that are paid out.

44:32 Emma Sjöström: Right.

44:33 Tom Thorn: So a customer has paid us. We've returned the payment or there may be a supplier payment that seriously was happening. Sometimes the pair of supply would pay us. We have paid them back. They appears again, we've painted back and you had like Yeah, just

44:45 Emma Sjöström: Well.

44:47 Tom Thorn: these oddities of people getting invoicing and all this kind of just real really, really messy. And but from a month to month, standpoint that can have a really big impact on on castle. So say that's a 500k or 1 million pound supply payment, and it's going in and out in in and out.

00:00 :

45:04 Emma Sjöström: Of course.

45:06 Tom Thorn: variances on variances to to our forecast. It's, you have to go back. What the hell is going on? So, it's, it's really when I say data, validation doesn't necessarily just mean, is it falling within the right category? You can actually be a bit of follow-up to say, Is this line? Accurate is someone just is this a bug? And the system or is this, is this someone genuinely posting three and four, inflow and outfit payments of the same value or purpose.

00:00 :

45:32 Emma Sjöström: Yeah.

45:34 Tom Thorn: Yeah.

45:35 Emma Sjöström: That makes a lot of sense and all. So you mention like the management team or bought like they don't expect every euro every cent to be covered. So that would be considered an issue.

45:43 Tom Thorn: Yeah.

45:45 Emma Sjöström: However, being able to make sure that you catch these potentially issues and then write in the explanation, right?

45:50 Tom Thorn: Yeah.

45:52 Emma Sjöström: So that's

45:53 Tom Thorn: Exactly.

45:54 Emma Sjöström: but listen, why you wanna validate data, like You don't want to miss any?

46:00 Tom Thorn: Yeah, you, yeah.

46:01 Emma Sjöström: Yeah.

46:02 Tom Thorn: I I guess the concern is that there's something that happens and then based on the The categorization of that, that's manual, horrible file, I'll call it be honest with it, that horrible father, we have at the moment that does the work and there's quite easy, they can quite easily happen that a flow falls into the incorrect bucket.

46:21 Tom Thorn: It's not it's not so full proof that It's not it's not so

46:26 Emma Sjöström: Bread.

46:27 Tom Thorn: That I could be completely 100% comfortable with it. So, I yeah, we go through it. We don't go through every single line this, for example, There's for I mentioned at the beginning that certain the payments of categorized is like a supply payment or customer payment stuff like that. I won't even bother looking at those, but there are certain transaction types.

46:45 Tom Thorn: Excuse me. Like I'd help like ad hoc function

46:46 Emma Sjöström: So, it's

46:48 Tom Thorn: transaction which may have, I don't know, 200 payments in there. I won't again. It won't go through line by line. But anything that looks odd within there, I will validate. I will make sure if there's a big flow in flow outflow in flow outflow, or if I, I might do a search for corporate tax, you know, that those kind of things, I might just double check to make sure that we've got Much as we can cover covered.

00:00 :

47:16 Emma Sjöström: Makes sense. The you already mentioned. like your metric like minimizing the minimizing the number of like, I guess probing comments or questions

47:26 Tom Thorn: Yeah.

47:27 Emma Sjöström: reports but like how often do you get this kind of feedback today and like is there any like recurring themes or anything like

47:36 Tom Thorn: Yeah, we do get them quite a bit. I would say that there's so much shared between myself and the FNA team, so it really does depends exactly who they're directed at. But there are a lot of comments for sure, the we there are repeating ones as well but they, they tend to be Things that are already known.

47:56 Tom Thorn: It may just be a reminder or checking like how we doing on this review. So customer inflows, for example. We've been looking at that for a while already so it's gone through a few already through a few reporting cycles. Where that well, you know, various different team members have been involved in reviewing and it may just be the case of the financial directing one, say an update on how things have gone and what changes have been made since last time.

48:18 Tom Thorn: So it, there are, there are repeats. Definitely repeat points in there, but I guess that would be another metric, right? Yeah, probably would be so the

48:27 Emma Sjöström: Yeah.

48:29 Tom Thorn: but then also the, the volume of comments. But then also, any ones that are repeated that with failing to, Preempt.

48:38 Emma Sjöström: Right. Exactly. That up. And I mean, like, I've gone through most of my questions or you've answered most of them.

48:56 Tom Thorn: Cool.

48:57 Emma Sjöström: and the

49:01 Tom Thorn: Hope other. Yeah, I I hope it was clear. It's that it's not the most

49:02 Emma Sjöström: Well.

49:06 Tom Thorn: straightforward process that we have a bit made sense difficult to do without like a visual thing. So, but I hope it hope those those made sense at least a bit.

49:16 Emma Sjöström: Thanks a lot of sense, for sure. I mean I could quit I could keep asking you like very similar questions, save something new pops up but I'm not sure it will. not sure it.

49:28 Tom Thorn: Okay. Sure.

49:32 Emma Sjöström: When you review a completed report. How do you judge if it's successful and need improvements still?

49:39 Tom Thorn: Hmm when I review so before maybe comments have come in from the management team and so forth. Yeah.

49:47 Emma Sjöström: I mean, before you send it off to them or like somebody sends it up.

49:51 Tom Thorn: Yeah, I guess it would probably be something you know, if Similar to maybe similar to some of the questions that you had like in an ideal world if I had all all of a like if I had everything available to me what else would I want to include in here for that one? investment report is that I don't I think more around the cash and have That the data comes to in a really mind again, really manual way, I'd love to be able to include some more interesting stuff within their in terms of metrics and expectations around where, where our investments are going to go.

50:23 Tom Thorn: So the moment it's all back with looking and I'd like to be able to get a position where we can have some forward looking information. And so if I if I'm producing a report but for example I think it's just limited, it's not giving like it as a stakeholder but okay branding a bit but say for example are cfcfo or my my manager directed Treasury and they may not know that that would usually be included so they've not necessarily always worked with Treasury professionals or I've had this type of report coming across to them so they maybe they just value whatever.

00:00 :

50:57 Emma Sjöström: Right.

50:59 Tom Thorn: I'm giving them And which they, you know, that they're happy with the report then giving them. But for my side, I would say that, I know if from previous systems that we had or previous setups that we've had I've been able to give more insightful information instead of just backward looking for the previous month.

51:14 Tom Thorn: It's also then forward looking information, based on expectations, on our forecast around investment rates or XYZ, we're not in a position to do that at the moment. So where I feel like the report is limited and not necessarily covering everything that I wanted to Yeah, it's probably like a comparison like similar, but previous reports that I've seen in other companies.

51:34 Tom Thorn: I've seen in other

51:35 Emma Sjöström: all right, would you say that I kind of limits how your ability to even present like what the Treasury function can help with like, is it's

51:42 Tom Thorn: Yeah.

51:44 Emma Sjöström: also does it deeper than like what's like the why you will want to present this like

51:49 Tom Thorn: You know, absolutely. I mean the Treasury function for me is is like overall as a support function so supporting there's I guess there's a couple of aspects to support support function, really primarily to the the finance team, but also to the legal team detect, various different teams to make sure that the business operates efficiently like that.

52:09 Tom Thorn: All the accounts are open when it needs to be good. Example, of that is uber used to have new markets every two weeks, and if we need to bank account, if we need, you know, all this kind of stuff. So you have to be really Food in, with all of those various different teams and really in sync with them, to get that up and running.

00:00 :

52:27 Emma Sjöström: Yeah.

52:27 Tom Thorn: Then there's the facilitation of all the cash flows, making sure payments like funding is in the right place for payroll and and supply payments and stuff. We mentioned the beginning, the other part of Treasury for me is then more of a value, add and advisory. So that's like the cash management pieces is the format making sure that we can all liquidity structure and the account structure and all this stuff supports the business in what it needs to do day today.

00:00 :

52:49 Emma Sjöström: Yeah.

52:49 Tom Thorn: side is, then the advisory stuff when you get towards like FX investments and all of that good stuff where you have way more specialism within Treasury and you should be able to like, There should be an element of my job where I'm taking in analysts or forecasting information from the market to say.

53:08 Tom Thorn: This is what we should be doing with our cash or This is trend within the market that we should be following or, you know, x y z that we're not doing at the moment that so and I I don't have the capacity to do that at the moment.

53:19 Emma Sjöström: No.

53:20 Tom Thorn: So, A system then comes in to say a system, actually supports both sides of it, anything that saves time allows me to do more of the advisory piece and the value out piece, and that's going back to the cash investment reporting. That's what's missing there at the moment, it's all just backward looking

53:35 Emma Sjöström: Yep.

53:35 Tom Thorn: How we did in the form of month, but actually on the advisory side and value outside, we should also be commenting forward. Looking, This is what we should be doing or should be thinking about, or this is how what we need to update within our Treasury policy to allow for this new type of funky investment.

53:49 Tom Thorn: That's going to save us, lots and lots of money, you know.

53:52 Emma Sjöström: Yeah, it makes a lot of sense.

53:55 Tom Thorn: Yeah.

53:56 Emma Sjöström: Very nice.

53:58 Tom Thorn: yeah, well I mean we want one could dream I don't

54:01 Emma Sjöström: Yes.

54:02 Tom Thorn: I don't, I don't think that's, I don't think that's in the short term but yeah, that would be the hope. We get some position where we can start doing so that stuff, So that stuff,

54:11 Emma Sjöström: I think so. Do you have any? Wrapping up. I know on the Top 40.

54:18 Tom Thorn: Um, no, no, yeah. Nothing I guess that we haven't already covered but I guess you know for the context of the conversation I think that's why for all of this stuff all of the the effort that goes into supporting a process that everyone recognizes is less than perfect. What would really improve things is that and that's why a system becomes so important is if you have something

54:41 Emma Sjöström: Yeah.

54:42 Tom Thorn: that can be a time saver, it can provide efficiency and accuracy, all of that stuff. Then it's It's super valuable and it pays for itself, right? It should eventually pay for itself or maybe not all of it but it should, it should pay significantly like a significant proportion should be either build back from investment, improved, investment income, or from the capacity that I then have to spend on other value add pieces.

55:10 Tom Thorn: So, In the context of all of that reporting, all of that manual, intervention the validation. If a system can do this stuff automatically for you. I actually have this conversation with Gidget earlier this It's

55:21 Emma Sjöström: Right. All right.

55:22 Tom Thorn: Say for it. Say for example I have I build out this new automation and hopefully that's going to improve things. A lot for me on a day to day, but it still means Unfortunately is we're relying on workday data to be available when it's available and that's generally in the middle of the

55:34 Emma Sjöström: Yeah.

55:35 Tom Thorn: month. If UM, was able to take our cash flow actuals and categorize those on day one based on its machine learning and and it's understanding of our of our flows. That's a huge win.

55:48 Emma Sjöström: Right.

55:49 Tom Thorn: They get that information two weeks in advance of what they get into the moment. So that's yeah, I guess that's where assistant comes in. So you can do all of your as much as you want as much as you can do within. The information and the the data that you have available.

56:07 Tom Thorn: But there's always other improvements out there and if you have a specialist system that that really does this for, you know, it's it's better spread of butter, it improves things hugely so hugely. Yeah, I guess it's a good place to wrap up, right?

56:21 Emma Sjöström: Well I think I'm super excited I can yeah I also like quietly register all these all these things that we already have that I feel like would probably make a difference for you guys, right?

56:32 Tom Thorn: Yeah. Yeah.

56:33 Emma Sjöström: Yeah that's super cool to be

56:34 Tom Thorn: Indeed.

56:35 Emma Sjöström: To be honest.

56:36 Tom Thorn: Yeah.

56:37 Emma Sjöström: Yeah. Hey, thank you.

56:38 Tom Thorn: Yes.

56:39 Emma Sjöström: again.

56:41 Tom Thorn: Pleasure. Yeah, no problem. No.

56:43 Emma Sjöström: This is super helpful for me and yeah I'm you know, can't wait to at some point. Hopefully also be able to show you what we're building when it comes to reporting.

56:52 Tom Thorn: Yeah, no, I'm excited. I know graduates, super excited as well. He was even mentioning something like We're talking next year in terms of any implementation. But, you know, timing wise, if it adds up there may be something already to show on the reporting side which would be. Yeah, super interesting to see that.

57:08 Tom Thorn: I mean, that would be for me, quite a big selling point if it was something we could already, you know, essentially replace our reporting out of work, work day. If we could place that something that's there on day one and super active. That would be huge, huge benefits. So yeah, excited for that.

57:23 Tom Thorn: Hope it goes well with the, with the, all the interviews. And then the build, I'm sure it's not a Noticeable tasks. So yeah, hopeful that goes well Hope all

57:33 Emma Sjöström: A huge topic, right? It's the video.

57:34 Tom Thorn: Yeah.

57:35 Emma Sjöström: So just navigating this finding a starting point and then go from there.

57:40 Tom Thorn: Yeah, I can imagine.

57:40 Emma Sjöström: Yeah. Yeah.

57:41 Tom Thorn: Cool. All right, thanks.

57:42 Emma Sjöström: Cool.

57:44 Tom Thorn: So let me know. I'm yeah, let me know if you have any other questions or anything comes up.

57:47 Emma Sjöström: Thank you, appreciate that.

57:47 Tom Thorn: Always have. Yeah.

57:49 Emma Sjöström: Let's keep that.

57:51 Tom Thorn: Yeah, sounds good. Thanks that see?

57:53 Emma Sjöström: See ya.