2026.10.06 | TB | Software | AI-Native Accounting Platforms Outlook

Analyst-led • 29 Sep 2026

Companies: XRO AU, BILL US

Adam Daigian (AD) Head, Go-to-market Engineering at Altruist Corp

Poorvi Banka (PB) — Third Bridge Sector Analyst

Agenda

• Key industry trends
• Growth outlook
• Competitive dynamics

Key Insights

• AI-native ledger platforms such as Puzzle, Digits and Kick hold negligible market share – well under 1% – compared to QuickBooks' 7-8 million customers and Xero's roughly one million
• "Puzzle was started seven years ago. It's basically using what is already considered an outdated software stack […] it was never an AI-native company". "Puzzle has no idea what it's trying to become strategically […] it's just trying to survive at this point"
• Roughly 50% of AI accounting market growth is flowing towards workflow reconciliation tools, benefiting players such as Basis, with general ledgers and vertical extensions splitting the remaining 50% roughly equally
• PE-backed accounting firm roll-ups aren't adopting next-gen ledgers because these platforms can't handle diverse business types beyond venture-backed software startups
• Expert thinks Digits or Puzzle won't be discussed in three years – a new set of companies will enter the market with agentic coding

Introduction

PB [00:00:03]: Welcome to Third Bridge's Interview entitled PE Sector Update: AI-native Accounting Platforms – Trends & Outlook. I'm Poorvi Banka, and I'll be hosting today's Interview with Adam Daigian, former Head, Growth, Revenue Operations and Analytics at Puzzle Financial.

Adam, before we start today's Interview, please state I agree or I disagree to the following statement: You understand the definition of material non-public information and agree not to disclose any such information, or any other information which is confidential, during this Interview.

AD [00:00:32]: I agree.

PB [00:00:34]: Could you provide a brief introduction to your background in this industry?

AD [00:00:41]: Yes, sure. As you mentioned, I ran the go-to-market organisation for Puzzle Financial, which was an AI-powered software accounting platform. I ran the direct-to-business motion, the partnerships motion and the accounting channel motion, the revenue operations, capabilities set to support those three motions and also business intelligence and analytics.

Questions

00:01:06] How would you characterise the AI-native accounting software market today? What has materially changed over the last 1-2 years? What is still more hype than reality?

AD [00:01:22]: Yes, sure. The AI accounting software market really splits into two branches, if you will. There are the general ledgers, of software platforms like Kick, Digits and Puzzle that go after small businesses. Then there are what I like to call ERP light, and that would be like Campfire, Rillet, DualEntry that go after also the small business segment as well but are able to move into mid-market too, and serve just larger, more complex businesses, and they're going after the ERP space, namely QuickBooks Enterprise and then obviously, NetSuite as well. In terms of what's more hype than reality, I would say several years ago, none of these platforms were really able to serve businesses with an accountant or an accounting firm in the loop today that they can absolutely serve some portion of businesses, especially where they've concentrated their efforts in software companies.

00:02:26] What would you characterise as the biggest change you've seen in the AI-native accounting software market in the last 1-2 years?

AD [00:02:41]: Adoption. Yes. I mean, I would say, ultimately, you've seen the ERP space, the ERP light space like Rillet and Campfire really break away from the accounting space in terms of adoption. Those companies have just been adopted much better, grown much better than Digits, Kick or Puzzle have.

00:03:00] Where do you think adoption is, as far as it concerns core AI-native accounting platforms such as Digits and Puzzle?

AD [00:03:14]: I mean in the grand scheme of things, these businesses are tiny compared to the incumbents. They have thousands of businesses on them compared to hundreds of thousands or millions of businesses. Like I said, they're quite small in size and scope compared to the incumbents and have a long way to go to catch up.

00:03:42] Looking at the SMB and mid-market accounting software landscape today, how much of the shift towards AI-native platforms is driving actual replacement of legacy systems such as QuickBooks Online or Xero vs being treated as add-ons or workflow overlays?

AD [00:04:00]: I look at all the competitors I just spoke about are, in fact, general ledgers. It's pretty uncommon to run Puzzle and QuickBooks at the same time, though occasionally, like that was seen. But everything that we just discussed here, none of them are really add-ons per se. They are like core ledgers, which means that your P&L, your general ledger, basically, all of your accounting work is going to be done in that platform instead of one of the legacy providers. I wouldn't say that's a guarantee 100% of the time, but more often than not, 98% of the time, 95% of the time, companies will just be using one of these upstarts.

00:04:42] If we talk about the smaller SMB segment specifically, what percentage of the addressable market there do you think has shifted to AI platforms vs legacy systems such as QuickBooks?

AD [00:05:03]: How are you defining SMB? Are you defining as like 100 employees and below or is a certain revenue band? How would you define?

PB [00:05:12]: You can define it the way you deem fit.

AD [00:05:18]: I think if you're defining SMB in a traditional sense, the penetration is well under 1%, like, a tiny fraction of the market. QuickBooks has seven million, eight million customers. Xero, probably around one million, high hundreds of thousands. Like I said, these platforms have thousands of customers at best. If you were to look at like Campfire and Rillet, it's possible that they have 3% or 4% penetration of the venture-backed SMB market. That would be an estimate, though.

PB [00:06:00]: SMB is the one segment where they must be considered more, right? These players must be considered more than in mid-market or enterprise, but the adoption is still very, very low even in SMB.

AD [00:06:12]: Yes. Campfire and Rillet are capable of serving mid-market software companies that have subscription revenue.

00:06:21] How should we think about the distinction between AI-enabled legacy accounting software and AI-native accounting platforms? Are there any capabilities that the AI-native platforms offer which could drive adoption and help customers shift from the legacy ones to these players?

AD [00:06:51]: In general, the engine behind these is really automation focused. I mean, on historical terms, accounting, it's like you put all your entries, you categorise them and then you hope that you can close a month, let's say, in September, you hope that you can close September's month by the end of October. With the automation that these platforms provide mostly in the form of categorisation or rules and things like that, they deliver what somebody or the industry calls a continuous close. You still would have somebody in finance or accounting saying, "Hey, this month is actually closed, but because there's enough categorisation, automation integrations in place, they deliver a continuous close." You're kind of like 90-95% done at any given time, which allows companies to then close the prior month within a handful of days vs a handful of weeks.

PB [00:07:54]: Is this differentiation something that customers care enough about? Given the low penetration, do you expect (talking over each other 08.03).

AD [00:08:05]: It's the primary. I mean, there are a couple of things that drive adoption. One is just a modernised version of what is considered legacy software. Two is the efficiency and the continuous close. I say the third piece is just more AI workflows, mostly in the form of reporting embedded into the product.

00:08:32] Industry reports estimate that the broader AI and accounting market is expanding at a CAGR of 30%-plus. Where specifically within the software stack is that spend being captured first? Is it core general ledgers, vertical-specific extensions or any other workflow add-ons?

AD [00:08:52]: Yes, it's a good question. I would guess it's spread across those three pieces, like vertical-specific extensions or even just extensions of the workflows in general could be pretty broad. I mean, there are all kinds of AR/AP, AI-powered AR/AP reconciliation tools like tabs, for instance. Then on the accounting workflow side of things, we've seen tools like basis and accrual focus specifically on the accounting firm, the mid-market accounting firm workflow, delivering accounting agents. I would argue actually more revenue has probably gone in that direction than has actually gone to the AI ledger space. I mean, accounting technology is pretty broad, especially when you start to bring in the accounting services market. PE-backed accounting firms and mid-market enterprise institutional accounting firms, everything up to the Big Four. There's a lot of market value to go after in each of these categories is getting some adoption.

00:10:09] Which players do you think will be at the forefront of capitalising on this opportunity based on the aforementioned categories and the ones that you expect to see higher adoption?

AD [00:10:21]: I would say workflow reconciliations, to be honest. They're not add-ons per se. They're more like they're really replacing the core workflow outside of the general ledger.

PB [00:10:30]: Which AI-native players do you think will see the benefits of this?

AD [00:10:41]: Basis, I think, is the most interesting one. Followed by Tabs, I would say. Tabs is more in the finance world, though. There are a handful of companies that are doing pretty well in the AR/AP reconciliation space going up against tools like Bill.

00:11:08] The industry expects the broader AI and accounting market to see 30%-plus growth. Do you agree with that industry estimate? How much growth would you expect for each of the segments within this, given the overall 30% expectation?

AD [00:11:34]: It's a good question. I would guess that 50% of it, if I'm defining workflow reconciliation add-ons the right way, I would guess that 50% of the revenue is actually going there and probably 25%, 25% to general ledgers and we'll see the other category, vertical-specific extensions. I think vertical-specific extensions is like the weakest proposition of all three. I honestly can't think of a single vertical-specific extension off the top of my head.

00:12:16] The underlying accounting software market is growing much more moderately than the industry estimate of 30%-plus growth for the broader AI and accounting market. How do you reconcile the differing growth profiles of the AI aspect and the underlying software?

AD [00:12:37]: Again, look, the accounting software and the accounting services market are quite large. They're some of the larger software and services market out there. Those grow essentially at the rate that the economy grows more or less. The tools that we're discussing here are eating up some of the budget within that space. Yes, they're like eating up the budget of the legacy players. They're eating up the budget of the other workflow tools, many of which are very much legacy software that accounting firms have been using.

00:13:21] Would you say that the expected growth of AI in the space would include the AI offerings of the legacy players as well, or would it be purely AI-native players?

AD [00:13:41]: I think it's probably towards AI-native players. I would say that Xero is doing basically nothing in AI. Bill.com is doing basically nothing in AI. NetSuite is basically doing nothing in AI and QuickBooks talks about delivering AI products, but hasn't really delivered all that much yet. So all that adoption is going to upstarts.

00:14:12] Traditional CPA [certified public accountant] and accounting firms are going through a wave of consolidation backed by PE funding lately. How is the surge of PE capital and consolidation into CPA firms and accounting firms impacting the adoption of AI-native accounting platforms? How is this shift changing how players target and sell to the newly aggregated practices?

AD [00:14:34]: I think in theory, like this is a benefit for the new platforms. In practice, the newer platforms actually aren't prepared to take on books of business of this size. It will be interesting to watch how PE firms that are doing accounting firm roll-ups evolve over time because they'll be comparing QuickBooks and/or Xero plus workflow automation tools on top of those existing ledgers vs replacing the core ledger. What I saw when I was at Puzzle is that these PE-backed accounting firms were not replacing the core ledger because they serve many different types of businesses, many of which are service businesses vs software businesses. The modern ledgers that were built here, Puzzle, Digits are really meant to serve small venture-backed startups. They really weren't a good fit for the vast majority of the businesses that these PE-backed accounting firm roll-ups are serving themselves. I was not seeing much adoption going towards next-gen ledgers. Now that might change in the next two, three, four years if some of these tools can survive that timeframe or we see better solutions in that timeframe.

PB [00:15:58]: Right now, they're not scalable to be able to cater to these aggregated practises.

AD [00:16:05]: Yes, that's right. You can't put 500, 1,000, 1,500 businesses of various kinds and sizes on Puzzle and expect a firm-wide adoption. That's a huge part, too, is training accountants that are very familiar with QuickBooks and in lesser cases, Xero, but still that tool as well, trying to get them to move different types of businesses to Puzzle or Digits is a huge challenge that will wipe out all the margins of a PE firm.

00:16:37] How does the value proposition differ between selling to an accounting firm and selling directly to a business or a CFO? For an accounting firm, would the accountants who will need to adapt to the new platforms be the decision makers? Would that be a barrier to adoption then?

AD [00:16:59]: Not usually the decision makers. Usually, it's the firm owner or somebody who manages that book of business, that's the decision-maker. I would say that the accountants or the practice managers themselves are absolutely stakeholders. Unless you have them on board, the firm will not be moving over to a new system of record like Puzzle or Digits.

00:17:20] There seems to be a shift from AI being used for individual tasks to AI taking ownership of the entire accounting workflow. How real is that shift on the ground today? Where are we on that journey?

AD [00:17:41]: Look, it's fairly real. It's also business-to-business dependent. It's like if you have a small software company, your expenses are pretty standard. It's like software expenses and food, maybe some travel. Very easy to categorise those expenses and then add them to the right accounts and a general ledger in those types of situations. Yes, you can see AI taking ownership of pretty much the entire accounting workflow with the exception of officialising a month-end close. If you're in a completely different type of business, you run a dental practice or you run a construction company, it's a totally different scenario in which case you won't see AI taking ownership of the entire accounting workflow. You might see 60-70%, maybe 80% as categorisation improves over time.

00:18:40] Are there any macro tailwinds for the AI-native accounting platform space currently? There's an ongoing shortage of certified CPAs and other accountants. How is that impacting accounting workflows and adoption of these software systems?

AD [00:18:56]: I don't think it is. I'm not sure that I buy this argument anymore because if you look at CPA forums and things like that, what you see basically are people getting fired right and left and also it being extremely difficult for accountants coming out of school to get jobs. I think that the CPA shortage is just a story that these companies are telling to drive hype, but I don't think that's a dynamic in practice in the market.

00:19:29] We discussed the key differentiator for AI-native accounting platforms vs the legacy ones. From an end-user economics perspective, what does ROI look like for accounting firms or small businesses adopting an AI-native ledger? Are they seeing a meaningful reduction in billable hours per client? Is the value showing in terms of increased capacity to handle more accounts per headcount? What is the quantifiable ROI like?

AD [00:20:02]: Yes, it's a good question. It really depends on the scenario, when we were doing this analysis at Puzzle, we were seeing maybe 10-20% improvement in efficiency just due to the categorisation. The categorisation in many ways is the easiest part. Then from there, it comes down to what's the business' revenue model? Are they on accrual books, things like that? Depending on the complexity of the business, it could be either worse, a wash or slightly better. I would say in the best possible circumstances, you're looking at maybe a 10-20% improvement in efficiency, which is meaningful for the bottom line of accounting firm.

00:20:58] How often do the AI-native accounting platforms end up in RFPs [requests for proposal] against legacy vendors? What do the discussions during such RFPs look like? What are customers most bothered about or focused on?

AD [00:21:13]: I would say frequently. Usually, it comes down to depending on the size of the business, it's either the finance stakeholder, the VP of finance, the director of finance, driving the decision. I think if you have that type of individual in the loop where finance is either partially or entirely internally managed, you may see the company push for something that's a little bit more modern and user-friendly like a Rillet or a Campfire. In the scenario where businesses are using outsourced accounting firms, they tend to defer to what the accountant says and the accountant will tend to use the thing that they feel most comfortable with, which is nine times out of 10 QuickBooks. I think the short answer is all the time QuickBooks. Sorry, like Puzzle or Digits are finding themselves, and I wouldn't even necessarily call them RFPs, but in purchasing loops with QuickBooks and Xero, and there's all kinds of reasons why the businesses end up choosing the legacy player. It could be capability sets, it could be accounting firm familiarity with QuickBooks and thus they don't have to change their practices, things like that.

00:22:42] How would you segment the competitive landscape today? There are incumbent platforms, AI-native ledgers and a lot of different categories of vendors. There are players such as Campfire, which are more ERP [enterprise resource planning]. How should we segment the competitive landscape today to make more sense of it?

AD [00:23:06]: Like I was saying, I think there is accounting software. The companies, the start-ups that they put in that bucket would be Puzzle, Digits and Kick. There may be some new ones by now. Then you have the ERP space. Like I said, I like to call it ERP light. Then in that space, you see things like Rillet and Campfire, DualEntry, stuff like that. Then in the automation and workflow vendors, I mean, this one is quite broad. It can be everything from categorisation agents all the way to complex accounting workflow agentic replacements like Basis, for instance, that I mentioned before. I guess you could also put that in the AI agents sitting on top of the ERP, but then that's also accounting firm technology platforms as well. I would say the automation vendors, AI agents sitting on top of existing ERPs and accounting firm technology platforms they blend together quite a bit. Then the one that you may want to include like the category you may want to include or not include depending on what you see as being relevant here. Are the AR/AP platforms like Tab, basically the platforms that are going after the incumbents like Bill.com.

00:24:37] How are players in the AI-native accounting platform space positioning themselves relative to one another? For instance, who is going after the SMB start-up market vs building exclusively for accounting firm channel partners or any other characteristics that would make them stand out?

AD [00:25:00]: If you look at all the players today that I mentioned, you'll see that they've all started to directly work with accounting firms, like they all basically realised that they couldn't build businesses by going just after the SMB start-up market. That doesn't necessarily mean they build exclusively for accounting firm channel partners, but the vast majority of businesses that are doing accounting, which is the vast majority of businesses do it through a bookkeeper or an accounting firm. That means that the business owner isn't doing the work themselves. That's the accountant or accounting firm is very involved in the decision for the tech stack here, that business owners tend to be not that opinionated. Over the last couple of years here, like when Digits and Kick and even Puzzle were all started, none of them were actively targeting the accounting firm. Puzzle was the first to actively target the accounting firm space, followed by Kick, followed by Digits and now they're all working as closely as possible with accounting firms, but with varying degrees of success.

PB [00:26:19]: The accounting firm channel is the more important distribution model. In terms of mix, what percentage of their book of business comes from accounting firms vs direct businesses?

AD [00:26:37]: It tends to be a small mix that actually comes from the accounting firms. It's more about what percentage of the businesses on the platforms use accounting firms.

PB [00:26:48]: Right.

AD [00:26:50]: That makes sense. Maybe 75% of the overall book of business comes direct from businesses themselves and maybe 25% comes from accounting firms, but it would be reasonable to expect that 60-70% of all the businesses on the platform are using an accountant or an accounting firm.

00:27:13] How are the players positioned in the AI-native accounting software market today in terms of the market share or perception that they currently hold? Is there a way to understand how Puzzle, Digits, Kick or any other similar players are positioned in the market today?

AD [00:27:52]: Between the three of them. What's the market share between the three of them?

PB [00:27:58]: Yes.

AD [00:28:00]: It's probably about the same.

PB [00:28:05]: Within the SMB segment, of course, I mean, just the segment that is relevant to them.

AD [00:28:09]: Yes, I'd be willing to bet that they have very similar customer accounts. Of the market as a whole, it's negligible, right? It doesn't even scratch the surface.

PB [00:28:25]: Do these players also compete with Campfire and the ERP light players that you mentioned?

AD [00:28:32]: Not really.

PB [00:28:34]: Right. Got it.

AD [00:28:36]: Not in practice.

00:28:37] What do you think about players from adjacent markets that are entering the accounting workflow market, for instance, Mercury and Ramp are aggressively expanding into core accounting and ledger workflow lately?

AD [00:28:48]: Yes. That's going to be slightly different use cases, right, like Mercury entering the accounting workflow is going to be a challenge for tools like Puzzle and Digits and Kick, whose like some of their primary value propositions is honestly being the best accounting software for Mercury, and they're doing that historically through having direct integrations vs connecting to other bank accounts through Plaid. Yes, it's logical for Mercury and for Ramp to try and eat into the value of the accounting motion, eat into the value of the general ledger. Now for Mercury, that's a lot more valuable for small businesses. For Ramp, that's a little bit more valuable for mid-sized businesses. I think it's less of a concern for the ERP light space because it's just unreasonable and unrealistic to expect that Ramp is going to build a general ledger, though anything is possible in today's world. In the SMB space, many, many of those businesses, their accounting needs just really aren't that sophisticated. You could argue that QuickBooks, Xero, Puzzle, Digits is actually unnecessary. It could be quite competitive for the AI-native accounting software upstarts.

00:30:21] Over the medium term, what market share do you think players such as Mercury and Ramp can garner in the AI-native accounting software space? How much of a threat are they to the upstarts?

AD [00:30:40]: I would expect that Mercury already has more customers than Puzzle or Digits on their accounting software.

PB [00:30:47]: Okay. All right. Do you see them as a material threat to the upstarts then?

AD [00:30:55]: Oh, absolutely.

00:31:00] How do you evaluate unit economics and monetisation models in the AI-native accounting software space? What are the different pricing and revenue models that work in this space right now?

AD [00:31:10]: It's basically all subscription. In addition to subscription, they tend to err on the side of, I guess, utilisation-based pricing model. More transactions categorized, and a little bit capability oriented as well. They basically scale up subscription costs as complexity, whether that's sheer transaction volume or just like overall complexity of the business scales as well.

PB [00:31:48]: Would it be more or less the same across players?

AD [00:31:57]: I'm not sure whether the players are using specific models, but the biggest challenge with the accounting software space vs the ERP space is it's extremely, I guess, commoditised. Xero and QuickBooks are relatively low-cost products in general that offer lots of functionality. Most of the work is done by accounting firms, as I mentioned before. It's very difficult for a customer to associate a lot of value with the accounting platform itself. There isn't much in the way of pricing model innovation that I see being available to these companies.

00:32:50] Puzzle launched AI Close as part of its broader AI Suite in September 2026. What do you think the company is trying to become strategically? What do you think are its most differentiated capabilities today?

AD [00:33:18]: Originally, Puzzle wanted to be in the ERP light space, but they made some key missteps. In order to be in the ERP light space, you need to have multi-currency and multi-entity and Puzzle failed to deliver that, and they failed to estimate the importance of that as you go upmarket. They found themselves in the accounting software space competing directly with QuickBooks and choosing QuickBooks as the incumbent that they wanted to disrupt. They focused on having enough like ledger-based functionality to get adoption within accounting firms. Then since then, they've really just been trying to continue to build on that functionality and drive efficiency for accounting firms so that they can peel more adoption away from QuickBooks because accounting firms are 80-90% concentrated on QuickBooks. That was up until they went through some interesting, I quote they fractured. Some of the team went to Accrual to do, I'm not sure what, to be honest. Then some of the team stayed working on Puzzle where it looks like they've refocused on small businesses and started to offer services as well. I think the short answer here is that Puzzle has no idea what it's trying to become strategically, and it's just trying to survive at this point.

00:35:00] What do you think is the likely future for Puzzle? Do you expect the remainder of the company to get acquired as well?

AD [00:35:19]: No, I don't expect it to be acquired.

00:35:21] Is there something that you think Puzzle could do differently to scale up or to get more market share?

AD [00:35:31]: No, I don't think so. It's a very difficult space in general. Look, they started at the wrong time because yes, you could argue that they're kind of a modern version of software compared to QuickBooks. In the grand scheme of things, like their software is actually outdated in legacy itself, and it's very difficult to scale and very costly to scale. If I look at question number 15, Puzzle is not really differentiated at all anymore, especially now that the incumbents are starting to build some AI products. The team is split apart. There's no more capital coming into the company. I would say at this point, it's like try and survive as a lifestyle business.

00:36:20] What's your opinion on Digits? It's been pushing the concept of an AI-native ledger and has launched Agentic Close in June 2026. How would you compare its strategic positioning to Puzzle?

AD [00:36:34]: Look, these tools work basically the exact same way. They offer some form of continuous close in terms of ongoing categorisation. Basically, your ledger is always like 90% done. When Puzzle is talking about AI Close or Digits is talking about Agentic Close, all they're really saying is that we do an LLM pass over the general ledger as it stands before the end of the month, at the end of the month, and we allow the LLMs to flag what else needs to be done in order to close that month out, at which point a customer can probably just say, "Yes, I agree to all of this," and then the work gets done and then it's closed. Does that make sense? These are not super strong value propositions, right?

00:37:34] Is there any player that you think is doing anything differently or stands out in some way?

AD [00:37:45]: If any of them were to be acquired, it would actually be Kick. The reason why I would say it's going to be Kick is it's been the most efficient from a growth perspective. They're focused a little bit more on a specific niche and that specific niche is newer companies that are basically S corps and S corps care a lot about expenses and expense categorisation, especially as it relates to tax deduction. Kick has focused a decent amount on serving that niche, whereas Digits and Puzzle just went broad and tried to compete with QuickBooks, which is very, very difficult. Kick is a little bit differentiated compared to Digits and Puzzle, but at the same time, I'm not sure how well they're doing either.

00:38:46] You said that if any of the players were to be acquired, it would be Kick. Who do you have in mind as an acquirer? For which player do you think a company such as Kick would be a good acquisition target?

AD [00:38:57]: Like Mercury.

PB [00:38:58]: What's the rationale behind that?

AD [00:39:01]: Again, Mercury, they acquired a ledger several years ago. That was relatively undeveloped and they had a team, a product manager focused on that ledger and they worked on it for several years before releasing it to the public just recently, but it was in beta for multiple years. In a perfect world, I don't see any reason why the legacy ledgers would have any interest in buying these tools because you can't integrate them and they're not that strong, which means that it would be like a neobank that would be, I think, where these things would be like most attractive, maybe like Brex, Capital One, like to compete with Mercury, something like that, but Puzzle is out.

00:39:52] What do you think about Accrual? It acquired a lot of Puzzle and its employees and entered the space by way of this acquisition.

AD [00:40:08]: Accrual, I'm not entirely sure about. If I understand what they're doing specifically, it looks like they're focused on… It seems like they've almost actually pivoted a little bit they may be going after Basis. The sense that I got is that they wanted to focus on or at least initially, they were focused on lower-entry market like tax operations, think like tax workflows for H&R Block, companies like that, they seem to have like pivoted a little bit because now their website says tax, audit and advisory. You may believe that they're going after something like Basis and they're building more of a broad context layer and workflow tool for accounting firms. I think that this is a much better space to be in, honestly, than the ledger space, but it's hard to tell how they're doing. I wouldn't say it's a good signal that Puzzle employees went over there.

PB [00:41:21]: You mentioned that accounting workflow and reconciliation segment is the one that is going to see maximum growth. Basis is one of the players that's going to benefit because of that growth. Are there other players that are worth calling out in that segment?

AD [00:41:42]: I'm not sure about important, but there are other players in the segment, and you'll definitely see more coming soon.

00:41:47] Apart from Basis, which other players hold relatively important or strategic positions in the AI-native accounting software space?

AD [00:42:06]: Let's see here. There's one other that I had in my… I don't know. It's a good question, actually. There's a lot of stuff going on in the space. As companies have realised they really struggle to grow, you see a lot of changes going on in the space. Like Numeric, for instance, went from close management to building an ERP. You've got Docyt, FloQast. It looks like Bill.com has actually delivered agent as well. It sounds like Pilot has an AI agent as well. It's hard to say for sure, and then Stax as well. It's a little bit hard to say, honestly, what the future is going to hold here in terms of like who actually shows up as competitive in the space. In general, it's expensive and difficult to build these products. Partly, it just comes down to who has the most capital.

PB [00:43:14]: You mentioned that Basis is one of the players that you think will see strong growth. Could you elaborate?

AD [00:43:23]: Basis far and away has the most revenue of any of these platforms today by a very wide margin. From a revenue standpoint, they are outcompeting like Rillet and Campfire. Then Rillet is now pulling away from Campfire. Tabs is doing really well as well. Everything else is nascent.

00:43:47] What do you think about Docyt? It describes itself as an AI-native ERP combining accounting, payments and BI. At the same time, it is vertically focused as well. What is your opinion of Docyt?

AD [00:44:08]: I'm honestly not sure. They're in the hospitality industry. The hospitality industry is an interesting space. I really can't speak to Docyt in terms of what's working for them or not working for them. What I can say is that some of these larger but more bespoke places I guess, vertically considered markets like hospitality, construction also comes to mind, maybe supply chain, stuff like that. Specific integrations are integral to the accounting workflow and the legacy accounting software providers and then also the upstarts just cannot afford to invest in those industry-specific capability sets. It's hard to say, they may be doing well, like they also may be not getting that much traction.

00:45:22] In general, do you think verticalisation helps in the accounting workflow software space?

AD [00:45:32]: Yes. Look, it does and it doesn't. My sense here is that the software that's growing the best is the ones that are… I take a horizontal approach and out-execute the rest of the market. There's not going to be… I don't think we'll see a ton of competition in the verticals themselves because like, a, the software is hard and the market is a little bit hard. Over time, that might prove to be the right strategy, but I wouldn't say at all that it's obviously the right strategy.

00:46:08] Where do you see the strongest competitive moat developing in the accounting software world today, especially given AI integration?

AD [00:46:20]: I mean, once you're on a ledger, you don't really want to leave a ledger. Ledgers can definitely have pretty solid competitive moats. I'm really not sure about the AI workflow tools. I think that if you're bolting on to a ledger, that's a risky place to be in unless you're really in the centre of the accounting firm workflow and by the centre of the accounting workflow motion, I'm talking about client communications, document management, like basically, you're building an AI context layer on top of the general ledger that, I think, could be of value. Difficult to know whether or not anybody specifically is doing that with the exception of maybe Basis and maybe Accrual. I don't see that being the case for either of the ledgers or for Docyt.

00:47:19] Do you see any M&A deals that could make sense for companies such as Accrual or Basis at this point, since these are the bigger players that are growing?

AD [00:47:35]: Yes, possible. Hard to say. The model companies won't buy them. The ledgers won't buy them. It's like I'm not sure where they would go. PE firms might buy them. PE accounting firm roll-ups could be an interesting option. There are some pretty big legacy Thomson Reuters, for instance, there are some pretty big legacy software companies out there that have deep pockets, but I'm not sure that these tools would sell to them. I'm sceptical about the M&A opportunities for these tools.

PB [00:48:25]: Do you think they could be on the other end as in could they acquire some smaller companies to move beyond the accounting space into adjacent areas such as payments, expense management, FP&A, treasury or anything else? Do you think that makes sense for them as the next move?

AD [00:48:44]: No, I don't think so.

PB [00:48:46]: In terms of M&A, there's not much that you expect for these players.

AD [00:48:53]: I'm not expecting much M&A in the future here.

00:49:03] If we revisit the AI-native accounting software market in three years, what do you think would look fundamentally different from today? What aspects do you think might remain unchanged?

AD [00:49:14]: I mean, QuickBooks and Xero will still own the vast majority of the market share. I think there's a good chance that Rillet will be claimed as the winner of the ERP light space. They'll become the real challenger for NetSuite. I think that you'll see a whole other crop of businesses like start-ups that go and tackle this space again because it's like a huge space with lots of opportunity that hasn't really been disrupted yet and companies will be doing it from a stronger position. What I mean by a stronger position is that companies will be able to start these businesses from scratch with agentic coding. They basically can start these businesses as AI-native businesses from scratch, whereas like Digits and Puzzle, I mean, Puzzle was started seven years ago. It's basically using what is already considered an outdated software stack and outdated processes, like it was never an AI-native company. Yes, I think you'll be discussing a new set of companies altogether. I think you'll see some players that have emerged as winners in the agentic workflow space as well.

00:50:55] Within a three-year timeframe, is there a lot of change that you expect in the competitive landscape of the space?

AD [00:51:01]: Yes, absolutely. Again, it depends, right, because people will take note of where companies have done well and where companies have not done well in the space. Like Puzzle, Digits, Kick, these companies have not done well, and you can tell by the lack of follow-on fundraising rounds. Basis, on the other hand, has done super well. If you zoom out a little bit, there are just much better places to focus by historical records, like looking at the broader AI and accounting landscape, and that would be, again, the ERP light space, accounting firm workflows and automations, things like that. The core ledger for small businesses hasn't proven to be a strong contender for software start-ups, but I think that you'll see another round of companies take a crack at it nonetheless.

00:52:00] What would define the winner in this space?

AD [00:52:19]: It's going to have to come from some form of innovation because if your plan is to just build a slightly better version on some fronts than QuickBooks, you're not going to see any adoption. I mean you're going against one of the most entrenched legacy software providers that's ever been out there. Yes, you need something more than just like, "Hey, we categorise a little bit better, we save you 10% of your time." You need like a fundamentally different approach here. I'm not sure what that looks like explicitly, but it could be a combination of software plus services. It could be a purely LLM-driven approach to accounting that connects natively to Claude and OpenAI or other models. It's got to be something better than, "Hey, we're 10% better than QuickBooks on one or two or three use cases."

PB [00:53:26]: Until then companies such as QuickBooks will continue holding most of the market even in the SMB space?

AD [00:53:36]: Yes. That's absolutely right. At the end of the day, the product works for almost any business out there. You have the entire accounting services market that knows how to use the tool, is willing to use the tool.

PB [00:53:54]: Within the ledger space, it's because the players today don't have the right software. They are dated. They are not what they're supposed to be. They're not truly AI-native. That is why we are not seeing a lot of adoption. If there are players that come up with a truly AI-native approach and tech stack, then there would be a shift from the legacy players to those AI-native ones in your opinion.

AD [00:54:17]: Yes. To put it in the most simple possible ways, you have to come up… You have to deliver something that's truly better than QuickBooks, like the Digits, Puzzles of the world, they haven't really delivered anything that's materially better than QuickBooks. I mean, of course, they'll say that, but in the reality, that's just not the case, and that's what the market tells us as well.

PB [00:54:42]: Is there scope for a truly AI-native player to flourish in the ledger segment as well?

AD [00:54:50]: I think so. I think there's still opportunity, but I would say, of all the software spaces that I've worked in, it's probably the single most difficult space. It's going to take like an excellent team that executes ruthlessly to challenge these incumbents. I wouldn't say it's impossible.

00:55:17] As of now, you see growth in the accounting workflow and reconciliation segment, including players such as Accrual and Basis, and there could be more players that enter that space. As far as the ledger space is concerned, players really don't offer enough differentiation for that transition to happen. Is that a fair summary?

AD [00:55:34]: Yes, that's absolutely right. Where there's pressure is on the accounting firm and also where there's value is on the accounting firm. It's like accounting firms are the ones with the biggest stake in a 10% to 20% or 30% improvement in their bottom line. A small business paying USD 200 or USD 300 a month or even USD 1,000 a month for accounting services, they don't get that benefit. They don't see the benefits in a tangible way of a 10% improvement or a 20% improvement in efficiency. They need to say something in order to move an accounting firm off of a system that they understand how to use like QuickBooks.

PB [00:56:16]: Thank you so much. That was the last question I had for you today. We can wrap up the Interview here. Thank you so much.

AD [00:56:29]: Okay. Thank you.

Transcription ends at 00:56:34 of the recorded material