Adam Larson: 00:26

Welcome to Count Me In. I'm Adam Larson, and today I'm joined by Sharoon Thomas, founder of Fulfill, for a conversation all about the unique accounting challenges facing direct to consumer brands. Sharoon's experience with ERP systems gives him a front row seat to the complexities D2C teams deal with: massive transaction volumes, multi channel sales, tough revenue recognition, and the struggle to close the books efficiently. We discussed why spreadsheets and legacy ERPs aren't cutting it, how modern platforms and AI are helping, and what a good close process actually looks like. So if you're curious about how technology is transforming finance for consumer brands, you won't wanna miss Sharoon's practical insights and real world examples.

Adam Larson: 01:07

Let's dive right in. Well, Sharoon, thank you so much for coming on the podcast. Excited to chat with you and get to know you better and to talk about what's going on in the accounting space. But to start off, I kinda wanna walk back a little bit. You know, let's talk about the beginning of your accounting career.

Adam Larson: 01:28

How did you first land in this space and kinda what drew you into it?

Sharoon Thomas: 01:32

Yeah. So I didn't start as an accountant, but when you work with ERP systems and do ERP implementations a lot, you end up working with accountants quite a bit. And every time I worked with, brands, one of the things that we've noticed is the finance team's always asking for, hey. Can we get more reports? We are behind, and, we're not able to close.

Sharoon Thomas: 01:52

And there's this additional data there. That's kind of what got me more into trying to figure out what exactly is so different about D2C and why is it so much more challenging for them to do these things. That's how I got more into accounting, especially for the D2C space, which is where we are now.

Adam Larson: 02:09

So let's kind of dig into that. Kinda what made you really decide, like, hey. I wanna get into this direct to consumer space specifically.

Sharoon Thomas: 02:15

So having prior to Fulfill as well, I worked as an ERP consultant. So we'd worked with governments. We'd worked with service companies. We'd worked with manufacturers, wholesale businesses, and so on. And something that we found pretty unique about b to c was how some of these businesses were starting to scale.

Sharoon Thomas: 02:34

And as they scale, despite being a business to consumer sort of transactions, they were all starting to work at b to b scale. But most of the systems that they had in place were all b to b type system. So these are all really good systems, but they were all expecting something like, say, hey, 500 transactions a month. And a business to consumer business, a d two c brand would end up having 500,000 of transactions a month, and everyone's figuring out that, hey. This is not working, so let's go and summarize all these transactions.

Sharoon Thomas: 03:07

So your actual ERP ended up being a spreadsheet. And the way we would hear those problems from an operational standpoint is we need more reports. We need more reports that summarize certain way. Oh, now these reports don't have the returns included in it. So let's go ahead add the returns back.

Sharoon Thomas: 03:24

For a direct to consumer brand, especially this time of the year, they would sell in November and December, and the returns is gonna come back in January. So you can't really book that based on actuals, so you need a percentage. Now how do you calculate that percentage? It's not fixed for every business. So what we saw is more and more ask for strange looking reports, strange looking data requirements, and it it would almost make you wonder whether is this finance team actually closing, or are they trying to build data pipelines for every piece of data that they and, actually, that ended up being true.

Sharoon Thomas: 03:55

Like, today, finance teams and d two c brands, they all end up operating big data pipelines. It's a data collection process first, then a data analysis part next, then comes the closing. And if there's any time left, they focus on actually looking into the business. So that's pretty much what we saw, and we thought, hey. There's a better way to do this.

Sharoon Thomas: 04:14

If we can handle millions of transactions for operations, that operational data should drive financials, not the other way around.

Adam Larson: 04:21

Yeah. That's so interesting how you have there's so many I feel like there's so many accounting teams that are doing what you've described, and they haven't quite figured out how to kind of consolidate it altogether to say, okay. This is when I do that. This is when I do that. And how do you think that these these how do you think these organizations are able to actually close their books every month?

Adam Larson: 04:41

You know, there's so many there's so many different methods out there, but I feel like a lot of times we get stuck, especially as accounting and finance teams, we get stuck in the same way we've always done it.

Sharoon Thomas: 04:49

Yeah. So a surprising thing that I found is a lot of businesses, even at twenty, thirty mil, don't close their books. It's one close at the end of the year and sort of an audit that goes through. Then comes along an investor or a bank or a lender that pretty much requires the business to provide them with solid financials and very close to the immediately after the end of the month. That's usually when businesses start thinking about we need to be doing this thing called a close, and we need to be sending the financials to these lenders.

Sharoon Thomas: 05:23

And that's the first time some of these founders would hear about Gap, And that's usually when d two c brands start putting together a finance team that actually thinks about a close. So it one, it doesn't happen in a lot of businesses where it should be there. Two, even when it happens, it's usually late. We're talking about twenty days, twenty five days after the month end when these closes actually hap and it's a lot of work. It's a lot of data analysis.

Sharoon Thomas: 05:50

It's a lot of spreadsheets. It's a some of them have described processes, so at least there's an auditable way to prove how things go. But it's just a lot of sheer willpower and hours that these teams put together to close the books even if it's, like, twenty days after.

Adam Larson: 06:06

So how does technology play a role in kinda solving this problem that you've been describing?

Sharoon Thomas: 06:10

Yeah. One of the things that technology done well is really good at is dealing with this transaction volume. So what worked at 500 transactions, if you try doing that at 500,000 transactions, it doesn't work. But for technology, that is just a just a small step. So that's where you can use technology to do these to make your close itself faster.

Sharoon Thomas: 06:32

It usually looks at it from a few different lenses. The high transaction volume is the obvious part, but there are other aspects too. For direct to consumer brands, for example, one of the big aspects of complexity is the different channels they sell in. So if you sell on Shopify, your gross revenue, your discounting, your net revenue, your sales taxes, because most brands are responsible to collect them and remit them, is different. If you sell on Amazon, that's very different because the sales tax collected, Amazon's gonna remit that for you.

Sharoon Thomas: 07:01

So that's really shouldn't even be hitting your books. Your gross revenue, there's really not that concept at all. It's net revenue because Amazon's gonna pay you net of all their fees and transaction and advertising and all of the other business. These are very different channels, and your bookkeeping has to be completely different for both of these. And as brands scale, they actually scale by adding more and more of these.

Sharoon Thomas: 07:26

They start selling to Nordstrom. They start bringing in an Etsy channel, eBay. So each of these additional channels that they bring in just adds more complexity. One more thing, technology is good at because it can interface with these APIs and consolidate all of these transactions for you. The bigger aspect, though, for direct to consumer brands is the complexity of revenue recognition itself.

Sharoon Thomas: 07:50

So if you look at I thought b to b businesses with all their rebates and so on was complex. B to c is a whole different level of revenue recognition. One, you have discounts, and you almost always have some discount or the other that's going on. Then you have complex bundling that happens. Then you have subscriptions.

Sharoon Thomas: 08:05

Hey. You can pay for six months ahead of time and get the product every month, and you might pay for it every month. You might pay in advance for it. That completely changes, and you have a whole new other level of deferred revenue that needs to be recognized. But d two c brands also do a ton of preorders because they have a loyal customer base that they can sell to and raise cash without having to go to a lender and so That's another problem in revenue recognition.

Sharoon Thomas: 08:26

And, of course, there's always returns, buy one, get one free, then you have influencer orders where you wanna send the product for free to them so they would talk about you, and you don't wanna book that as COGS. Instead, you wanna book that as marketing. So the complexity of revenue recognition for a direct to consumer brand is just at a whole different scale. And with the high transaction volume and the different channels, it's just a complex night. Now the one thing that we didn't even touch about is where all your inventory is.

Sharoon Thomas: 08:56

You probably have it in, you know, three different three p l's. You're operating a warehouse, maybe you have a store. All of this is just so much complexity that technology is better equipped to handle, and it's not a job for a spreadsheet.

Adam Larson: 09:10

It's not a job for a spreadsheet at all. And I think the the larger it gets and the more complex all the different transactions and the different data coming from different places and being in different formats and all that stuff, Obviously, you need some sort of a platform to kinda put all this together and see that. Now do traditional ERPs are are traditional a ERPs able to handle all those transactions occurring from those different entities?

Sharoon Thomas: 09:33

So if you look at traditional ERPs, they were also mostly built with the wholesale idea in mindset. So they're pretty good at revenue recognition, but when it comes to the scale of transactions, it becomes a challenge because most of them's their go to market model itself scaled by charging businesses for transactions. And now a direct to consumer brand could be doing the transactions what they expected a billion dollar company to do, but this is just a $20,000,000 brand. And they're selling a $20 widget online. So the go to market models of the legacy ERP systems itself is at a mismatch with what most direct to consumer brands need, and then they end up getting charged for a higher transaction tier and a higher transaction tier.

Sharoon Thomas: 10:14

So even when they can do the rev rec piece, it doesn't. The other gap is usually direct to consumer. It's all about scaling your channels. And if it's about scaling your channels, each channel is a new integration, and legacy ERP systems were not designed to for the level of integrations that are needed today. They were all designed at a time when humans actually got a fax, entered the order by hand, and send out an email as best as a confirmation.

Sharoon Thomas: 10:41

That's a a completely different era, and most legacy RPA systems are still licensed that way, used that way. So that makes it very challenging despite being a solid accounting platforms.

Adam Larson: 10:52

So how are the how are these new ERPs that are coming up, you know, kinda like the one that you run? How are the how are these new ones kinda tackling these new issues? Because, obviously, legacy ones have their way, they're great for what they do. But there's new channels. There's new things that are coming at them that you know, how are these new these new RPs kinda handle tackling these new problems?

Sharoon Thomas: 11:13

Yeah. So systems like Fulfill are designed from scratch to handle the scale that a direct to consumer brand needs. So we expect that even a simple sales order that happen on Shopify, for example, is going to result in four to five journal entries if you have to do the book at different times, if you have to do bookkeeping right. So as soon as that checkout happens, are we booking deferred revenue, and are we debiting the gateways that the money got into? It doesn't stop there.

Sharoon Thomas: 11:46

It doesn't it's not just as simple. You're collecting sales tax. You probably wanna carve that out. That's not part of your deferred revenue. So as it is, you have two journal entries now.

Sharoon Thomas: 11:55

Just you just got hold on an order online and you have two journal entries. At some point, you're going to split ship from two different warehouses or got to get these items out. So you gotta book your inventory journal entries. Then that's when revenue recognition comes into play. So a modern ERP system for a DTC brand should be capable of handling each of these different timing events and book a finance entries at each of these timing events.

Sharoon Thomas: 12:22

So when brands want to close, they're not exporting data from five different platforms and trying to group them together in a spreadsheet. They should be focused on, okay. We have up to date revenue data. Let's figure out if all the bank transactions are in. That's so it's it's all how connected are these is your data.

Sharoon Thomas: 12:40

How connected is your ERP to all of the other digital tools that you're using, including your credit card company and your bank and your channels and your warehouse. If all of these are connected, you have the data for a quick close. If these are not connected, you are going to be spending a lot of time on spreadsheets to find out what's the number to, book for a close.

Adam Larson: 13:00

I can only imagine how if you have them all connected and it's all in one place, it almost seems like this is too easy that it's all in one place that we can actually close our books now. It it seems too simple. Right?

Sharoon Thomas: 13:11

Our hope is that with the ability to close quicker, you spend more time on the business and looking at the nuances. For example, one of the things that I saw a gradual evolution with one of our customers is initially they wouldn't be booking anything for returns. There's no clear data for proactively booking provision for returns. They started by looking at, okay. Now we have the data.

Sharoon Thomas: 13:33

Let's look for a year of data and see what we should book. We should book 15% of the last month, 8% of the month before, and let's book that as provision for returns. Then they got even more nuanced because now they have all this data. So they start taking different percentages for different channels because your Shopify returns is not going to be your Amazon returns person. So once you start nailing down faster close, you can get more and more financially accurate, and the quality of your accounting, quality of your bookkeeping can substantially go up because now you're not spending time chasing down spreadsheets or chasing down a report which only one person in the company has access to from a sales sale.

Sharoon Thomas: 14:12

Instead, you are improving the quality of your books, giving the management more visibility, able to forecast. So we believe those are things that are better time times better spent on those than on trying to collect data for close.

Adam Larson: 14:28

Yeah. For sure. So what about things like when you're trying to do keep up with, like, gap compliance and and dealing with that many volume transactions and there's so much complexity across the different channels. How do you handle that in this type of a platform?

Sharoon Thomas: 14:41

Yeah. So GAAP compliance for DTC brands is a very, very tricky affair, and it starts first with revenue recognition itself. Right? So when you talk to brands, sometimes you hear somebody say, hey. We did 10 last year.

Sharoon Thomas: 14:54

But what was your revenue? They would say same 10 mil. But this is a number that's coming from the admin screen of something like Shopify, which is actually cash collected. So it's not even really revenue. It's just the cash collected that in probably includes shipping revenue, which needs a whole different revenue treatment.

Sharoon Thomas: 15:08

It probably includes all the gift cards that you sold. It has items that you still haven't shipped because some of those were preorders. So the GAAP compliance part of it is tricky for direct to consumer brands, and that comes from each of these transactions having multiple aspects to it. Sales taxes, which shouldn't be part of your revenue, but most systems will just report that in the top line number. You are selling these are platforms that bill based on GMV.

Sharoon Thomas: 15:34

So your GMV at your list prices are always higher, but you're always selling on a discount. That compliant revenues now takes a step lower. Then you have additional costs that go into it. Some of the revenue is from a you know, they added a gift card into the order of a $100. So it's not really the $900 a customer paid of after the taxes, it you should also reduce that gift card and recognize it when you sell it.

Sharoon Thomas: 15:56

Right? So all these different aspects, what we often find is something like when there there is a difference of, say, about 20 to 30% difference between what an admin screen of one of these platforms will show you as your, quote, unquote, revenue, which is really just GMV and your actual GAAP compliant revenue number at the end of the year. That is also very, very amplified for periods like Black Friday and Cyber Monday, which almost always is at the November. But the vast majority of those orders for which you collected cash will only ship in the month of December. So it looks for most brands that don't do the accounting right, it look very skewed because you spend all your advertising dollars in November.

Sharoon Thomas: 16:38

You acquired these customers in November. You got the cash in November, but you have to be recognizing your revenue and your COGS in December because that's usually when most brands will just ship out all of those items and all of those returns in January. So you what do you even book for? Something like the returns provision should be different. So, yeah, it's a it's a gap compliance is a tricky beast for even a 10 or $20,000,000

Adam Larson: 17:08

Yeah. That sounds and so, you know, we've talked a little bit about the tech well, how is technology playing a role, but maybe you can give some examples on how, like, things like AI are being used to kinda take these huge huge things that you just described that can be super complicated and confusing and maybe, like, maybe making those now easier to handle.

Sharoon Thomas: 17:28

So, our big thinking on AI is that in the near future, a finance team is a small team with an army of agents that does a bunch of things for them. They can be an agent that's responsible for making sure that your revenue recognition is right, an agent that's coding your bank transactions and just waiting for approvals on the ones that's needed, another one that's doing AP reconciliation and so on. But all of that is still a few a few steps away. Today, I think the most practical use case for finance teams with AI is just report generation. One of the nail problems with large language models today is converting text into SQL, and most data is stored in databases that understand SQL.

Sharoon Thomas: 18:14

But the problem has always been a finance user can either use all the canned reports that come with the software, which never really meet the requirements. So they have to export a bunch of these reports and system together with VLOOKUPs on a spreadsheet, or they have to describe what they want to a user to a person who understands SQL and databases, but not exactly finance. So now they have to go back and forth with them and iterate with them to eventually come up with this report. After you have done that a few times, now you don't wanna do that for the amount of the friction that's there prevents you from doing it for all of the different use cases. But with AI, finance teams today can describe exactly what they want and get a report back with the data, with the precise controls that they need.

Sharoon Thomas: 18:59

After they get it, wanna improve it, say, oh, I want warehouse also as a column, and I want this revenue report segmented by channel and warehouse and product and all of these different dimensions. You can just describe it and get a report. That's what we are seeing our finance teams cost our customers have the highest value in. Yeah. All of the other fancy stuff like, you know, coding bank transactions with AI and so on are there, but they all still require quite a bit of human in the loop involvement today to be accurate enough for modern financial reporting.

Sharoon Thomas: 19:27

But something that works really, really well out of the box today for finance teams with AI is report generation. It's the biggest quality of life improvement for any finance team.

Adam Larson: 19:39

Well yeah. Because especially the the out of box reports a lot of times, they're they're they were made with something specific in mind. So being able to go to it and say, hey. I wanna see this with these numbers and these things. You know, I the other day, a colleague of mine was trying to do a report, you know, in a platform and just wanted to add an extra column, and they couldn't add the extra column.

Adam Larson: 19:58

We reached out to the support of the service. They're like, this report is not customizable. I'm like, what's the point of having this report if I can't add other columns? If I have to take two spreadsheets and then combine them together for VLOOKUP, which we've all been doing for for all the years that we've had those platforms. But what's the point of having modern software if we can't customize a report in that way?

Sharoon Thomas: 20:17

So even from a software vendor's perspective, every one of those changes will upset half of the users and by making the other half happy. So they always say, okay. Let's not change this report. Let's add one more version of this report, this one extra column. And then you have thousands of reports inside, and nobody knows which report to use.

Sharoon Thomas: 20:34

And Exactly. Ends up just being complex and more complex. And from an end user's perspective, all of these reports do something, and there's gaps, like you said, you know, it's like endless v lookups. And every time you wanna pass that knowledge onto a new team member who joins the team, now you're sending, hey. You gotta export from all of these different places.

Sharoon Thomas: 20:51

It's more and more. That's where I think AI is just if you don't have are not doing this today, you probably don't have a data lake, and you probably don't have something like Claude that should be doing this for you today. This is not a futuristic thing. This is something that you should have today. Every finance team should be equipped with this.

Adam Larson: 21:08

So, you know, you've talked a lot about closing better. You know? So for a d two c brand, you know, maybe you can just, from your perspective, what does the healthy close process look like for those teams?

Sharoon Thomas: 21:18

Yeah. If you have a healthy close process, you should typically be closing in less than five days. And that five days are even there. You should probably have most of this process documented. And the five days is usually a buffer decent buffer because some platforms, some, especially things like Amazon, for example, will take a while before they generate their accurate inventory numbers for a close.

Sharoon Thomas: 21:41

They generate their settlement reports and so on and send it to you. So a five days is a good target for a close for the pay. For d to c brands, the largest transactions are always going to be the revenue recognition transactions from the different channels. So you shouldn't be waiting till the end of the month. You should be able to do that throughout.

Sharoon Thomas: 22:01

So at the end of a month or day after, you should probably put in a soft close where revenue recognition and inventory transactions sort of end up only locked for just the finance team. That'll allow some things like, you know, your warehouse reporting of inventory numbers and so on to come in. And by day two or three, you have your inventory numbers also closed. Inventory is probably the outside of cash, inventory is probably the biggest asset in your books. That looks solid.

Sharoon Thomas: 22:29

Cash balance from bank feeds. If you should if you don't have a bank that gives you a feed, you should probably switch to a bank that gives you a real time feed of your account transactions. That helps you close a bit faster. That leaves you with revenues closed. Now it's all below the operation below the gross margin line.

Sharoon Thomas: 22:47

So those are ones that take a little bit more tricky. So there too, if you wanna have faster close, what we recommend is using tools that give you all of this data and integrate with your existing stack. For example, if your payroll software is not giving you an easy to export integrated transactions, you probably should switch to a more modern payroll system that will do that for you. So each of these things being checked off as, hey. These are all our major line items, and this is what takes the highest amount of time.

Sharoon Thomas: 23:16

You start knocking them down one by one and have an integrated process. You would see that by your soft close, the vast majority of your transactions are done. That leaves you with how do we make sure everything is in here, and that's a bunch of custom reports that ideally are all AI generated. That's giving you summarized information that you can easily cross check to get to a quick close by five days. That is a good process for most brands.

Sharoon Thomas: 23:41

Of course, each of these things need to be a pre prepared checklist that you're going through and knocking off as a team, and you can get there.

Adam Larson: 23:49

Yeah. You can't it's not something you can just turn on saying, okay. We're gonna start the five day close tomorrow. There has to be a process to build that foundation correctly the way you're describing it.

Sharoon Thomas: 23:57

Yeah. If you are not closing right now, that's probably the first thing to start. Hey. We wanna get to a close end for businesses that we have customers that come in from something like using QuickBooks or Xero, and they are we're getting all of this thing together and have never had a close process to being their first system that helps them do that. We also have people who come in from existing ecosystems where they have a close process, but that was purely based on summarized transactions.

Sharoon Thomas: 24:20

And now they have a flood of these detailed transactions that's coming in, and they never had a process for chargebacks. They never had a process for what's called a green return, which is effectively the issuing a refund, but you're never expecting the customer to return the product to you. So all these gaps suddenly start surfacing, which all previously hid somewhere inside revenue recognition and return. So if it both of these process changes, we help these brands start thinking about how to get a close process together first and then start accelerating that close month after month. And I think it's a good place to be if you are able to get close in five days.

Sharoon Thomas: 24:55

Mhmm. A hard close.

Adam Larson: 24:57

Yeah. A hard close. So you you talk to professionals all the time in accounting teams. And, you know, if you're and if you were able to sit down and talk with, you know, a direct especially ones that are direct to consumer brands right now, you know, what's one thing you wish they could understand about their business that they could change their and they would have changed their approach to work to be to take a step in the right direction.

Sharoon Thomas: 25:17

Yeah. What I would tell DTC brands is don't try to take finance and use that to make change operations. You should first sit with your operations folks and then have operations work well enough that it comes to a good finance close. I'll give you a couple of examples. When this is a season full of discounts and promotions and and holiday and all kinds of offers out there, we always tell merchants, hey.

Sharoon Thomas: 25:46

Plan what the cadence for your discounts are going to be and the type of discounts that are going to be there. A classic one that gets finance teams catches them off guard is when you suddenly start seeing a whole bunch of transactions where there are order lines for $0. These are buy one, get one offers. Now you're not gonna start recognizing revenue for them as zero. You still got to recognize revenue the gross revenue at, the full value and discount it.

Sharoon Thomas: 26:12

So if the operations team is gonna go ahead and launch a marketing campaign that does buy one, get one, and the finance team is not, part of it, you end up with a month and close nightmare where suddenly it looks like they sold a bunch of product with value, and then you're going back and calculating that and so on. What works really well is when good operations data is available to you, a plan that's come where the operations team and the finance team's working together to, to figure out how the business operates and how these changes and if whether it's a marketing plan or operations plan is rolled out. Are you bringing onboard a new warehouse? Like, how accurate are the inventory numbers, and how are they gonna be able to send adjustments? How often are they going to do accounts so you have accurate numbers?

Sharoon Thomas: 26:54

That's what finance cares about. So good operations should drive financials, especially for operations heavy businesses like d two c brands. So what I always encourage them to do is work with the operations folks and make sure that the operations is tight, and that gives you the solid data without you having to make up data for your close. And you start seeing surprising things happen once once these two teams start working together. Something very interesting that I saw in a company where this is the case is every month, it would take the finance team weeks to close because they're waiting for Amazon to settle funds to them.

Sharoon Thomas: 27:31

Once they describe their operation, operation is like, we can always trigger a close on Amazon closer to the end of the month and have that settled. So those kind of small operational changes immediately make a difference to the brand. So now more or less, it's like those credit card statements that span over months, and you always have to split them. Amazon settlements are the same way. Once the operations team can start aligning those closes towards the end of the month, now it becomes an easy process for finance to close.

Sharoon Thomas: 27:56

So always working closer with the operations folks and describing your challenges and where that's coming from. They will come up with creative solutions to make process easier.

Adam Larson: 28:06

Yeah. It sounds like as you diversify and have all these different channels because you can't just sell in one place anymore, you have to sell in a number of places to go where the consumers are. You have to find a way to kinda easily reconcile it altogether. It can't just be one place in one there's no there's no one source of truth anymore. You have many places that's coming.

Sharoon Thomas: 28:23

Yeah. Yeah.

Adam Larson: 28:24

Well, Sharoon, I thank you so much for coming on the podcast. It was great having you here. Great hearing your insights, especially in the direct to consumer space, and I really appreciate you coming on the podcast today.

Sharoon Thomas: 28:34

Adam, thanks for having me. I really enjoyed talking about, legacy brands and their finance challenges. Thank you.

Announcer: 28:43

This has been Count Me In, IMA's podcast providing you with the latest perspectives of thought leaders from the accounting and finance profession. If you like what you heard and you'd like to be counted in for more relevant accounting and finance education, visit IMA's website at www.imanet.org.