Ep. 357: Ali Hussain - Rethinking Finance Stacks in the Age of AI
Welcome back to Count Me In! I'm your host Adam Larson and today I'm joined by Ali Hussain, Founder and CEO of Tabs. In this episode, we explore the evolving world of accounting and finance technology, from what drew Ali to the field to the recent advances that are transforming how finance teams operate. You'll hear us discuss why so many legacy systems are falling behind, how AI is more than just a flashy feature, and the real world potential of autonomous agents in finance. Ali shares his perspective on navigating industry change, what finance leaders can learn from sales and engineering, and how organizations can create a culture that embraces innovation without losing sight of people.
Adam Larson:So if you're curious about future of finance, the role of technology, and how to lead teams through rapid transformation, you won't want to miss this conversation. Let's dive right in. Well, Ali, thank you so much for coming on the podcast. Really excited to have you here. And before we kind of get to the big ideas that we're going to talk about today, I wanted to walk back to the beginning a little bit.
Adam Larson:What drew you to the world of accounting and finance technology, and how did that pass eventually lead you to building something like TAPS?
Ali Hussain:Adam, great, great to see you. It's one of those things that I stumbled upon. If if I look back throughout my professional career, if I if I had imagined in the post AI era, I'd be building agents for finance and accounting. I don't think I would have imagined that a lot of it came from just being an operator. Think one of the things about being a founder is like trying to find a founder market fit and and I was fortunate enough to get a ton of operational experience, pre tabs.
Ali Hussain:And so I was a COO for seven years. And as a chief operating officer, you often oversee a lot of different enterprise applications, cloud, ERP, CRM. And it was my fixation with kind of the finance stack in ERP, where even before AI, I was like, woah, this is like way behind what product engineering teams are getting to work on even sales teams and marketing teams. And so for several years, I started to really get my hands dirty with tools like NetSuite, QuickBooks, and other kind of general ledgers. And what I noticed, Adam, was that most of them were starting to unbundle, meaning a lot of folks were leaving NetSuite for their accounts payable and spend to vendors like Ramp or Zip, or on payroll and benefits, Rippling and Gusto and Deal.
Ali Hussain:And so for me, it was like this just maniacal fixation on why the perhaps the most important part of the ERP, the flow of funds in, billing rev rec, was left behind. And so with that, when I realized I wanted to go try to build something bigger, better, faster, it was that fixation of just the pain I had felt trying to get those systems up and running that ultimately led to the founding of TAPS.
Adam Larson:And that's something you hear a lot where companies are trying to get people to stay with them because they're trying to upgrade their tools. They're saying, hey, we're adding this new feature, adding this new feature. We've partnered with this organization, so stay with us and also get an account with them so that you can stay there. It seems that it seems that that that seems to be the the message that a lot of these organizations are saying. But what are most people in accounting and finance missing when they actually look at what's happening right now?
Adam Larson:Because I feel like without even realizing people are starting to use AI because like this tool I'm using, oh, they have a new AI feature, but they're not understanding what exactly they're doing and how it's actually fitting into their their work stack.
Ali Hussain:Yeah. Adam, I think this is the whole problem with this suite, in particular finance and accounting. They were kind of sold two worlds. One is a single suite, and you buy everything from us. And in theory, that sounds great.
Ali Hussain:You have one vendor, everything works out. But people forget, like, finance is a very complicated, multi threaded, multi workflow, multi data primitive set of actions, many of which are deterministic. So there's a lot of complexity in different data and workflows and domain knowledge. And I think for many years, was oversold on the belief that they could get all of that from a single vendor. That said, I have a lot of for finance professionals because the alternative seems so confusing, meaning, okay, I don't buy this suite.
Ali Hussain:Do I really have the capacity and the bandwidth to go buy 50 different tools and try to Yeah. Work together? And so the thing I always say is like, there's actually a really good middle ground. Like, it's not sweet or nothing. And it's also not, hey, I need a bunch of engineering support and implementation support to pick and choose a bunch of point solutions.
Ali Hussain:I think it's in the world of AI getting a lot simpler, where you can go back to first principles and and what really matters is, in in our view, GL's the least important in some ways, that's just a repository of data. But what really matters is where are the core data primitives that you must operate as a finance professional. And for most companies, and look, it can vary a little bit if you have inventory or you have a ton of like unique domain specific needs. But at the end of the day, finance runs off of your vendor management, who do you have to pay out. It runs on your employee management, which is typically your biggest cost center.
Ali Hussain:How do you run payroll benefits and compliance? And it runs on how do you manage your customer relationships to drive revenue, billing, and how do you drive compliance on those customer relationships, which is commonly tied to rev rec and reporting. And so if you take that first principle, that also ties to a lot of domain knowledge around accounting. Do I do billing? Do I do accounts payable?
Ali Hussain:Do I do revenue accounting? And so our belief is in this new paradigm, you don't have to go in either extreme direction. There's this happy middle ground, which is a good way to start to think about the stack. And I think if you're not doing it that way, the data doesn't actually empower the things you need from an AI standpoint. And you're left with a bunch of cool party tricks, but not technologies that are built around the appropriate kind of first principles.
Adam Larson:I think you you really hit the nail on the head there where, you know, a lot of times we're just left with a bunch of real cool party tricks like, hey, this can do this, but we don't know actually how to apply it to our to our tech stack, but it's really cool though. Look. And I think we I think so many times organizations get caught in that. What are some what are some tools or some ideas that we can say, hey, so we don't get lost in the party tricks?
Ali Hussain:Yeah. The number one rule I have is the only way AI and ultimately agents can work is if they have ultimate data context. And so if working with a tool, and this is and again, this is why I think it's so hard for legacy systems to adapt is party tricks are built downstream off of cool kind of features versus like a really core way in which the technology works in a different way. And so the way to avoid the party trick pitfall is to focus less on the features downstream. Because, yeah, any legacy org or any new org can get some thoughtful product folks to put some cool new dunning features or Cash App features or reconciliation features.
Ali Hussain:But what ultimately really matters is, is there a data moat that my vendor is building that not only gives me much more power, but allows the technology and ultimately the agent to have the same aptitude and latitude that I as a human would have and even more because it can work around the clock. It can have access to data in real time much faster than a human can process. And so Adam, for me, the best tools create some form of context mode. And so example, something like tabs, like we create context modes around all of your contracts and your product usage and your audit opinions. These are things that pre AI native tools would never be able to understand in full context.
Ali Hussain:They'd be highly tied to structured objects that came from like integrations and APIs versus a completely new paradigm in how data is understood. So that is actually the most important thing I look for is less about the feature and function, but how is the tool completely reimagined in terms of how it operates as a new technology?
Adam Larson:So let's talk a little bit about, agents because, you know, there's a difference between AI as like a feature within your software or AI as an agent that can actually do some of the work. So maybe we can paint a picture of like what that looks like in practice.
Ali Hussain:Yeah. The best way I describe it is does it start to emulate what a human on your team would look like? And so what that means is a few things. One is, how does it work with you? And to me, I think there's a lot of fatigue around.
Ali Hussain:I'm not gonna open up 40 different chatbots. It's hard enough for me to just be in Clot or Slack or email. For me to be in a QuickBooks box and a Ramp box and a bill.com box and a ADP box and a tabs box. It just doesn't make any sense. And so for me, the easiest sniff test is, does this person or agent work with me like another human on my team?
Ali Hussain:And I think if they start to pop up in email or Slack or even phone call, which is one of my hot takes, sooner or rather than later, like that starts to feel like an agent. The second is can the agent handle tasks autonomously, but also in ambiguous use cases, meaning it's not just simple y equals x equals z, but there's like the ability to handle ambiguity that requires deeper thinking. Like I need to go read the contract, or I need to go look at prior emails to understand, or I need to understand product usage to understand if this is a real churn risk. Like the things that an actual human would do that we've never seen out of a system of record is a second litmus test. And then the third is it has the ability to also know when to ask you questions.
Ali Hussain:And it's not just completely rogue. I think that last piece of like understanding memories and preferences and knowing when they may need to come back to Adam or Ali to just get clarity or direction, to me actually demonstrates like this deeper level of comfort for you to actually be able to give up one and two. And so those are some of the signs that I think the agents have arrived, which which I'm pretty bullish is like, I mean, our our our users are starting to experience those today. But I think we're gonna see real leaps and bounds ahead of differential and products this year, because there's a lot of noise to your point earlier, Adam, around AI native and AI AI party trick. We're about to see certain magical workflows where we're like, woah, like that's very different than what I can vibe code or what I see with my legacy systems that are now all AI.
Ali Hussain:Like this is this is a new type of paradigm and how I work with technology.
Adam Larson:Well, and I can imagine it's a little scary to give some of those roles over to the agent saying, hey, I want you to do this and you have to trust that it'll come back like the things you were saying. It'll come back and ask you questions. But, you know, what is it? What does it take when you're when you're having conversations with people to say, hey, try this out and give it a shot? Like, how what are those conversations like when you're talking to finance and accounting professionals?
Ali Hussain:Yeah. I think we're first, I'm always amazed. Like, finance and accounting, which has historically been, in my opinion, left behind in terms of technological innovation and is often seen as a function that is cautious and risk averse. I I would bet for many companies now, especially in the last six to twelve months, finance and accounting is the most AI forward of any function in many companies. They are stewards of doing more with less.
Ali Hussain:They are stewards of keeping their even if companies are growing, teams a little bit more leaner. And they're some of the most heavy users outside of engineering of tokens to try to get more insights and reporting. That said, on the actual agent autonomous side, rightfully, finance and accounting has a higher bar because they can't get 80% of the way there. They have to get there in the high nineties. And so when I when I look at ways to get them comfortable, there typically needs to still be a workflow that allows the human on the team to have approval processes so they're not driving the call police autonomously.
Ali Hussain:They're still touching the steering wheel, and they want systems, particularly still feels like software systems to give them audit trails of what happened. And then the last add to it is I'm seeing a level of comfort on more low labor, low cost workflows like billing. I think there's still trust to be built to do things like rev rec and month end close and things that are much higher higher risk. But on things like billing that they're already outsourcing, very tactical, I've been amazed, Adam, by how quickly they've been excited to move to the next wave.
Adam Larson:So, you think about AI agents and where they're heading the next couple years, which roles in finance and accounting are most likely to look different and which ones do you think will become even more valuable as these continue to evolve?
Ali Hussain:Yeah. I think a lot of the tactical roles, very honest, have been hard to recruit for anyway because of the lack of CPAs in The US, the higher cost of CPAs, and also their ability to guide what type of jobs they want to do. I think the more tactical entry level jobs like the billing department or the general ledger ERP admin type role, like internal services and maintenance type jobs, I do think are going to be more challenged in the coming years. I think the types of roles that'll be a bigger question are the more cerebral roles, revenue accounting, FP and A lead, etcetera. I think that is an open question.
Ali Hussain:But I do think those roles are gonna have to reimagine themselves quite a bit, be a lot more AI forward, and will generally be leader departments, at least in house. I think the areas where we're gonna need more folks are people who are highly AI forward. And beyond being domain experts in accounting or FP and A or strategic finance or capital markets, I think they're gonna also just have to be good agent administrators. So as you think about promotions, you go and historically have always, I'm a good people manager. I'm a good domain expert.
Ali Hussain:I think a lot of that will evolve on how good am I understanding my system stack and can I run it that is both able to use vendors that bring agents to extend my team? And then on top of that, how do I become pretty good at building on top, so that I can get the data to the agents or I can run my own reporting or dashboards in like research lab tools like Quad and others. And so I actually think that is getting very real and we're seeing some really emerging finance leads get really good at.
Adam Larson:Yeah. It's almost like you'll have a new role in the future of like agent manager, you know, where you're managing a bunch of agents as opposed to people and that that's gonna that's gonna make the workplace look a lot different.
Ali Hussain:100. We're already here's where I do think finance has lagged a little bit. We're seeing it a lot in the sales and marketing front. Like we at Tabs have a go to market engineer. Their job is to purely administer agents for our BD team, our sales team, etcetera.
Ali Hussain:I've yet to see full on the equivalent in finance and accounting in terms of a named hosted role, but I I wouldn't be surprised if that doesn't start to emerge in the second half of this year. Yeah.
Adam Larson:Now, when you and I were first chatting about having this conversation, you mentioned to me that you're a realist. And I was curious about, you know, what conversations are accounting and finance professionals like having? What conversations should they be having internally right now? Because most of them are probably avoiding.
Ali Hussain:Yeah. So I think there's like three or four things that my kind of pragmatic realism around this. One is for many finance and accounting folks, they're not gonna completely rip out the world, meaning they have been on NetSuite for seven years. They're under audit with NetSuite. Like, they're planning to go through some type of M and A event or a public offering.
Ali Hussain:And so my first hot take on realism is I don't expect you to blow up the entire world, but I do think there are ways you can be leaner and less reliant on a single stack. So like, I don't believe in the suite anymore. I think you, like, you can put in whatever GL you want, but you can build a lot of modernization by moving your AP spend, your payroll and benefits, your order to cash rev rec to modern stacks that can augment what you already have and move you quite a bit in the future. So that's number one. Number two is there's some amazing ways you can vibe code and self build and MCP into the research labs like Claude, etcetera.
Ali Hussain:But I think people forget finance, again, can't be 70 to 80% right. But a lot of it's also deterministic in which LMs are not perfect at. And so anything that is a workflow or a compliance based, my realism is you should buy versus build. And I think it's never been a better time to be a consumer as a finance professional. There are many options.
Ali Hussain:Pricing is more pragmatic. There's a lot less cost to implement and onboard. And so I'm a big realist on, yes, there are some places should be building and understanding how to use Quad and other tools. But the core workflows are still going to be a buy. And it's very hard to build anything that understands determinism in contracts or touches payments and fintech or ties to gap and six zero six compliance and all of those things or touches workflows like email and dunning and AP portals.
Ali Hussain:And then the third part of the realism, which I think is my biggest take is I don't I actually believe the head of finance, the CFO controller is more important than ever. Meaning like, I am not a person that believes in the zero person finance team. I'm not a person that believes finance and accounting professionals go away. I just think the teams will look different and the winners now. And I think we see the same thing in the engineering side.
Ali Hussain:Like some of the most technical engineers that used to be the most sought after are being augmented with a new type of curious engineer who is adaptive and understanding where to put their calories and where to augment amazing LM technologies to do more and deeper and spend a lot more strategic time on the whiteboard with their team, etcetera. And so I don't think the finance and accounting team as a whole goes away, but I think it looks a lot different. Generally, yes, it is going to be leaner with less FTEs and drive a lot more impact. And I think that's just a reality that most people have now kind of succumbed to in the last in the last twelve to eighteen months.
Adam Larson:So, you know, you've mentioned a few different industries like sales, marketing, engineering, fields that are kind of more advanced in the AI field, however, doing more with it. What what can what can accounting and finance professionals learn from watching those industries go through this? And are there warning signs they should be paying attention to?
Ali Hussain:Yeah. I think the learning can just be what is the parts up? So I think there's a couple of things. I do think finance and accounting teams can set some good precedents, meaning like I think and truly, I've not yet seen finance accounting be the office of no. I do think they're learning how to provide each of the team more exploratory budgets to go able to augment their operations with AI.
Ali Hussain:I think the caution is I've yet to see someone effectively a 100% nail what that means in terms of huge human capital replacement. And so some of the warning is like, if we could drive more build this year, I'm yet to see someone say, oh, I need half the engineers. I think it's more how do I keep the same number of engineers, but I get a lot more out of them. And so if the two x engineer, which used to be a good engineer, can now be the five x because everyone can't just find a 10 x engineer. Like, generally, I'm yet to like want to see drastic cuts in like human capital.
Ali Hussain:It's just getting a lot out of those individuals, I think is the thing. I think the one thing finance and accounting still is seeing, and I see the common frustration with CFOs I talk to, is they actually, I would say most of them feel like their teams are catching up to sales, marketing, and I think everyone's trying to catch up to engineering. But some of the frustrations I hear are some of the other teams are not catching up, meaning HR, people ops are lagging behind. In some ways, legal depends on the company could be falling behind. In some cases, G and A functions or I hate to say it even sometimes product like is falling behind.
Ali Hussain:And I think that is more of the warning sign is like you may enable, but ensure that enablement is pretty cross functional. And I think that is where you can be a good steward to the org and try to ensure that the right people and the right enablement is going on so that some of the other teams also push the puck on using these technologies to make the org better.
Adam Larson:So what does that look like changing your organizational culture from outside the finance and accounting from like the CEO down of saying, hey, we're going to be AI forward thinking. Because in some ways people are kind of like, wait, what does that mean for me? Does that mean my job's going to change? Like, how do you stay relevant and how do you change your culture to that AI? Because it's a it's a little scary because we're like, hey, we want to be better.
Adam Larson:We want to go faster. But am I going faster to moving myself out of a job?
Ali Hussain:Yeah. I think the smartest people generally have comfort that jobs are going to evolve. And and if the company is doing well and it's a good culture, you create a culture where the more you can make yourself redundant, the more strategic your role is going to get. And so if you are getting out of and enabling AI to completely get out of your job, there's always other things to do, whether more seniority, more strategic roles, etc. And I think we want to create that culture.
Ali Hussain:I think if you create a fear environment in which people think AI is a way to cost cut or bring jobs versus do more strategic work, it lays the wrong precedent to begin with. Some of that may be true. I'm not here to defend that. But let but but creating a culture of fear and pure efficiency on AI versus strategic ROI and upside, I think is the place you start. I think the second is as much as I think finance folks are great stewards, like a lot of this has to come top down.
Ali Hussain:The CEO, the the other CTO, other company is founder run, need to create an environment in which this is the culture at the company beyond engineering. Some of that's budgetary, like each department has a certain amount of exploratory AI tools they can go by. The second is like, if your entire org is not yet set up with full on access to Claude or OpenAI or other tools, like that is a huge miss. And I think that needs to change. Like, you have to remove the friction.
Ali Hussain:People shouldn't be having to self buy their tools to use AI in the workforce. Yeah. Third is, I'm always amazed by how much training gets missed. And there need to be a couple stewards within the company. I'm not saying it has to be a dedicated AI team, but there has to be you have to enable the people who are some of your best enablers of AI to get the podium and teach folks company wide.
Ali Hussain:Our view is we we originally believed it would be helpful to have like an AI center of excellence that trained all departments. We realized the nuances and the empathy for the day to day was getting lost there. And frankly, those AI teams have so much more to do just on the product side, is we've kind of kind of the equivalent of AI czars within each function. And look, like, it might be a dedicated role like we have in marketing and sales with the go to market engineer, but like in support or on the AE team, etcetera, there are people who have just learned this stuff much faster and understand how to apply it as a support lead or as an account executive. And their job is to spend about a third of their time teaching the rest of the AEs or support folks or building tools for them to do better.
Ali Hussain:That may mean a little bit of quota relief or it may mean less support tickets or some of those things. But those are ways we are kind of disseminating that purpose. And then the final one is I do think you just to push that standard with each leader. It could be your HR leader. You say, look, like, you're reporting, I will not look at Google Sheets anymore.
Ali Hussain:I need real time data out of a clawed dashboard. Or I often, in the middle, turn the conversation in my interviews and just say, hey, look, if you come on to lead finance or support or CS, how will you use use AI today? And what are some ways you would roll that out very quickly? And just ask it point blank. And if there's someone who hasn't been thinking about it, they're probably just not a good fit for your role in 2026.
Adam Larson:That's that's really interesting because I I I'm thinking like, you know, it's not all about displacement. Know, you can read the headlines in the news about these massive organizations laying off people and they're saying it's because of AI or not may or may not be because of AI, but I feel like roles are going to continue to evolve and maybe roles that didn't even exist before. Like you said, like there's the AIs are for this function and that function. Are there other roles you think that will be created that, like, because of these agents that simply just did not exist before?
Ali Hussain:I think we're gonna just see in most leadership profiles a different type of person. And it used to be a lot of what was the last three jobs you had and does it look like the job that I need you to do? And in my last couple searches, I have found that not to be true. I think people are generally looking for varied experiences, but the level of slope and just clarity around the moment we're in right now and how they want to come in and not only use AI, how they want to build teams, how they want to get access to data, how what type of company they want to build, what type of culture they want to build. And so I'm hesitant to say we're just going to see a bunch of net new roles.
Ali Hussain:But I think the profile of leaders is going to change. Many people who had the prior gold standard profile will adapt and they'll get there. But I think we'll see some nontraditional folks running sales teams, running finance teams, running AI teams. And it's going to be very interesting. It's going to look very different.
Ali Hussain:And I've already seen that poking up in particularly a lot of startups or venture backed companies, because I think they're often an early indicator of what's changing in market. And we're certainly seeing that.
Adam Larson:So I I'm I'm really curious because you're no. You're at the center like, you're a leading company at the center of this change. And when you and I first talked, you were saying that, you know, what you're doing as a leader has to evolve quarter by quarter. What keeps you grounded when that landscape is shifting so fast?
Ali Hussain:Yeah. One is just being incredibly open minded. That's like the number one thing. Like, I again, I mentioned earlier, I came from more of a COO background where you're so used to planning, thinking about annual plans, budgets, etcetera. I was at a breakfast not long ago with eight of the top AI CFOs, and none of them had more than a quarter planned out.
Ali Hussain:I think one had six months planned out. And more than half and I'm kind of planning at the quarter level now, but more than half, we're not planning more than a month out. And that to me is crazy. These are, like, much bigger. Some of these are much bigger businesses than even mine, well over a 100 revenue.
Ali Hussain:So that's number one. Number two is you have to give yourself some space every couple weeks to be like, what's changing and how am I adapting to that? Some of that is just playing anthropologists within the company, spending time with different teams and actually understanding how they're evolving. The second is you yourself. Like, I I spend a good amount of my weekends in the tools, understanding how I send up co work, how I get different types of access to HubSpot, how I change the way I send emails, etcetera.
Ali Hussain:Like, I think you have to give yourself space to do that because if you don't do it yourself, expecting others to do is incredibly hard. And I think the third one is like, you you gotta also just just also just like disappear from it all for a little bit too. Like, there's also some times where it's important to get back on the whiteboard and write with your hand and like it can also be so mentally that and so like finding some creative time and I think that comes back in one by just giving you that bag space.
Adam Larson:Yeah. Because if you don't, you could get lost in focus on the technology so much that you forget to think about the bigger picture to kind of shift your focus. How have you how have you as a leader adjusted yourself in leading people through this change? Because I know it's scary and you kind of have to be fast paced, but you also want to inspire employees, but also not want them to burn out. There's so many things you have to measure as you're leading folks in this in during this time.
Ali Hussain:Yeah. I think it's just communicate, communicate, communicate. I mean, we do an all hands every two weeks.
Adam Larson:Okay.
Ali Hussain:Create open forums for people to ask how I'm thinking about it, ensuring that as things change even every month, they're not just changing for me and my leadership team, but we're communicating that in terms of how that's going to change the profile. Also, being very honest where where there are some pockets that aren't moving. And I mean, even as we move from a series A to series B company, we have a lot of pressure to grow human capital, understanding where there are certain folks who this may not be the right place to evolve this fast every month, five days a week in person, and we have ways to move move as friends and and and find a different type of talent profile as the company evolves in this moment. And so so those are some of the things that I think have helped, but but it is something that is constantly evolving. Like, I'll give you an example.
Ali Hussain:Like, I used to meet with my executive team maybe once a week or once every two weeks. I now meet with them two to three times a week because I think when you get caught up in the motion, not taking a step back and like, a little bit of it's structured, but a lot of it's let's just come in, grab the whiteboard, and talk about what's evolving and how we're gonna address it. It may be macro tailwinds, things from our customers, launch of a new quad tool, competitive things, whatever that may be, we evolve and we talk about. I'll give you one really good real example. I sit down with my BD leader every couple weeks.
Ali Hussain:I'm like, hey, what's working? And the answer I could get last month is social, LinkedIn. It's certainly not email anymore in terms of cold emailing people. That's like dead. And it's been really interesting.
Ali Hussain:Like, it's like the most effective thing is humans calling again. It's like we're ten or fifteen years ago, cold calling. And it's like also being grounded in reality is like in some cases, AIs create so much noise. We have to go back to the most manual processes to kind of find our edge, and I think it's that open mindedness that that has been most helpful for me as a leader.
Adam Larson:That's amazing. Well, Ali, I really appreciate you coming on the podcast, sharing your insights with our audience, and I I hope they got as much out of it as I did. And I just thank you again for, sharing with our
Ali Hussain:I appreciate it. Thanks, Adam. Thanks for having me, and I'm looking forward to seeing you again.
Announcer: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.
