‍

In this episode of ‪What Worked, Mike interviews Julian Rowlands, Founder and CEO of Cashboard, a platform that connects your company's financial data to become a single source of truth for FP&A teams. Julian talks about how his on-the-job CFO training led him to make the tool he always needed. Julian shares his insights about:

• Lessons learned from Xendit (Stripe of SE Asia), Spruce, and his ad-hoc CFO board of directors
• The future of finance jobs, from apprenticeship to the one-man team
• Advice on how to make the leap from CFO to founder

We'd love for you to connect with us:

• Mike Wu
• Tyler Rachal
• Julian Rowlands

‍

Transcript edited for clarity:

Mike Wu

Welcome to another episode of What Worked. We are very excited today to have with us Julian Rowlands, the founder and CEO of Cashboard. You can check out Cashboard at cashboard.co. Cashboard is a business insights and analytics platform for finance teams. And over there, they are using AI to automate a lot of the work that traditionally FP&A teams have done very manually. Without further ado, Julian Rowlands, thanks for coming on the pod.

‍

Julian Rowlands

Thanks for having me, Mike. Happy to be here.

‍

Mike Wu

Yeah, I'm very excited to hear and learn more about Cashboard and how you guys are helping finance teams today. But I want to dive into your background and kind of like what led you to starting Cashboard, helping automate kind of the FP&A team's workflows. Yeah, would you share a little bit about your background?

‍

Julian Rowlands

Yeah, thanks for asking. So I'm kind of an accidental CFO. Right out of undergrad, I wanted to do something fairly business-y and applied for a hundred jobs. Only got one offer and it was to do data systems type of consulting for hedge funds, working for a software company that basically just sold to financial institutions. I spent four years doing that kind of data systems related work, a lot of data warehouse stuff. Before going to business school, I went to U Chicago. And the entire time I was there, I sort of made fun of it for being the CFO factory. And then immediately after I became a CFO, which I thought was kind of funny in retrospect, and actually wound up loving it too. But when I was there, I actually wanted to do emerging markets work. 

‍

So spending time in developing countries, trying to help businesses grow in tough places. So did some work in Sub-Saharan Africa and Myanmar and Bangladesh too, and then stumbled upon a startup called Xendit in Southeast Asia. And back then they were just in Indonesia. Now they're all over the place. They're in the Philippines, they're in Vietnam, they're doing really well. And they were looking for a CFO. And it was just a good fit. They needed someone who understood data systems really well. They needed someone who would be comfortable working in a place like Jakarta and being based there for a while. And you could speak finance and speak VC. And so, yeah, a bit of an unorthodox path, but wound up being their CFO for a year and a half, helped them raise their 60 plus million series B from Accel, Kleiner, YC Growth, and some other people. And then came back to the US and back to the New York area. I'm from New Jersey originally, so back to the homeland. And was head of finance for a company called Spruce, which is now a part of Zillow. They're doing pretty well. And they're an API for title insurance. So that's kind how I fell into the CFO mix. Totally accidental, was trying to do emerging markets work and then found that the financing background plus the data systems background actually lent itself really well to the role.

‍

Mike Wu

Yeah. I can see the connection there, but I love the term you're coining accidental CFO. I wonder how many of you there are out there, but I imagine you're not alone. But could you, could you share with us, maybe for anyone who's watching this that's an aspiring CFO on that track, you stepped into that role, how did you figure out how to, how to be a CFO, based on your background and where you're coming from.

‍

Julian Rowlands

Yeah, I had to move pretty fast to try and figure that out. Great question. I had never worked in a finance org before I became a CFO. It was truly sort of an out of nowhere thing, not something I was expecting, but a really cool role came up with a company that I thought was incredible. And so I just decided to jump in and take a swing.

‍

I was super clear with Xendit when I was going through the interview process, like, hey, I've not done this before. I'm not a CPA. I've done some investing work, so I'm pretty comfortable with that lens, I've worked with startups, but this will be my first time. I never hid that. And they were very open to the build rather than buy experience, if that makes sense. Like we think Julian's smart. We think he can figure it out. Let's take a bet.

‍

And so the first thing I did when I joined Xendit as their CFO is I basically went to them and said, hey, who are the CFOs that are in your network who I can talk to, who can help me get smart on this really, really fast. And also did the same in my own network with my more limited pool of CFOs. And so one of the people that helped me, I spent a lot of time with this guy named Cam Hyzer, who was the CFO of ZoomInfo for a while and actually IPOed them in the first half of 2020, which is pretty wild. He's been an amazing mentor and then Xendit through the Y Combinator network. They basically said, hey YC, who do you know who's a good CFO, who Julian can talk to? And in the fall of 2018, they put me in touch with the CFO of OpenAI, the CFO of Segment, the CFO of Brex, and a couple other folks. All pretty wild names to have been in touch with, but all those people basically took a ton of time. I remember walking through downtown San Francisco with the then CFO of OpenAI, Brad Lightcap, who's now their COO. And all of these people really spent a ton of time just helping me understand exactly what mattered, what to invest in, what not to invest in right then, how to build a team and how to prepare the org to continue to scale. So I felt really lucky to be able to talk to those people and just try and learn from the best as fast as I could.

‍

Mike Wu

Yeah, that's amazing. It sounds like you built a very quickly, like a personal board of directors, like CFO board of directors that you could reach out to and ask questions too. That's amazing.

‍

Julian Rowlands

Exactly. I'm still in touch with a lot of these folks. I've traded emails with them. A lot of these people in the last year, they've been really amazing.

‍

Mike Wu

I want to talk about Xendit because I feel like in the States, it's kind of a lesser known name, but I know in like you mentioned in Asia, it's like, you named some of the investors on the cap table. Did Xendit come out of YC as well? 

‍

Julian Rowlands

Yeah.

‍

Mike Wu

Yeah. And so very much what we would consider a blue chip startup that had a great, story introductory to it. Is there one thing about Xendit that you remember that you guys did fantastically that you carry over to your next roles?

‍

Julian Rowlands

Yeah, so Xendit, it's an incredible business. Maybe I'll take a second to talk about them because no one in the U S really knows who they are unless you're tracking top YC companies and stuff like that. But they're basically Stripe for Southeast Asia, helping merchants accept and move money across all the local payment rails and countries like Indonesia, the Philippines, Malaysia, Vietnam, Thailand, Singapore. It's been a while since I was there, I left at the end of 2019 so I'm sure they're doing a lot more cool stuff now. But they're really incredible. They've grown a ton and are just doing pretty wild things. And the founders there are unbelievable, like Moses, Tessa, Juan, Bo. And Bo is actually an angel investor in Cashboard. So thanks Bo, if you're listening to this for your support. 

‍

Mike Wu

Shout out.

‍

Julian Rowlands

Shout out. So I think the number one thing that I learned from Xendit is just build things that people want, right? And that's classic YC advice, right? People just say that a thousand times, but if you have product market fit, that's the only thing that matters. If you don't have that, you can build all the foundations you want, but your business is not gonna explode in the positive way. And if people are ripping your product out of your hands, then everything else, like you have the opportunity to figure all this stuff out. So I think that's the number one thing.

‍

They moved to Indonesia and basically started this business at a time when it was so hard to move money. And there was this burgeoning class of startups in Indonesia that was trying to do a lot of fintech stuff. And just had very little opportunity to actually make the money transition from their customers to their own accounts or vice versa. Xendit came in and basically built a very simple API, like the classic Stripe story, right. Built a very simple API for accepting and moving money. And all of a sudden things just exploded. 

‍

So when I showed up, they had at that time probably 70 or 80 employees. The developers were working out of the basement of the mansion in South Jakarta. Sometimes the basement would flood. I wasn't there for that specific incident, but I've seen the videos. And then by the time I left a year and a half later, there were 250 people and last I checked there were 700-800 or so. They raised a couple hundred million dollars from Insight and Coatue at the end of 2021. Just fell totally by accident to a pretty incredible story.

‍

Mike Wu

Yeah, that's so cool. So you got to see what a really strong founding team looked like and what product market fit felt like. You saw that like the product being ripped out of the hands of Xendit  into the customers, you know, into the customer's use cases. So that's probably very valuable experience and things to have experienced as you, as you're building a Cashboard. Can I ask you the same question about Spruce? I don't want to skip over Spruce, but was there one thing, you know, that you find inspiring about Spruce or one learning that you took away from that experience that you're applying at Cashboard today.

‍

Julian Rowlands

The story with Spruce is pretty similar, really incredible founding team, Patrick and Andrew, really, really smart guys who were building around their own pain points. I joined Spruce because it rhymed in a lot of ways with Xendit, like complex B2B, money moving infrastructure, with a lot of complexity that was being simplified through an API. And APIs are great for executing a lot of transactions at scale. And so I just felt there was something really powerful there. I had joined Spruce right before the pandemic started, I think like February 10th of 2020 was my first day in office. So was not in office that long. But yeah, that was a pretty wild ride and really, really smart team worked really, really hard. They're now a part of Zillow. But yeah, that was a great experience as well.

‍

Mike Wu

Yeah, very cool. And then let's jump to Cashboard. This is what I really want to dive into. And I'm most excited to learn more about. I gave a brief intro. Could you share with us what Cashboard is, including the vision, what's the vision here?

‍

Julian Rowlands

So Cashboard is a single source of truth for CFOs and finance teams that can show all the numbers you care about, all the metrics you care about from all your systems in a single place. And we have AI that can actually write reports for you and write analysis for you on your own data and deliver it on autopilot. The North Star of the business is inventing the solo CFO. That's obviously aspirational and meant to be, not quite we're going to go live on Mars like Elon's thing, but it's pretty far on the future. So I don't want anyone to think that that's where we are right now. But when I was a CFO, you have a couple of pain points, right? The main one is you don't have a single source of truth. You use a sales CRM, you have an accounting system, you have a budget, you have a payroll system. and so all the data you care about is spread out in a bunch of different places. You don't control any of those data systems. So all the data labels are different. Every system might have different names for every department.

‍

And so just to understand exactly what is going on in the business and how performance is tracking and sharing that data across the team is super, super hard without investing a ton of money in serious data infrastructure. And so when I was a CFO, like one of the things I noticed when I talked to all these incredibly smart people that I looked up to a ton when I was trying to get smart on  how CFOs operate is the most forward looking ones with the most resources, which is a critical thing, we're building pretty heavy duty data infrastructure for themselves. So they were using something like Fivetran to connect all the different data sources. They were piping all that data from NetSuite, Salesforce, payroll, everything into a data warehouse like Snowflake or Redshift. They were putting a BI tool on top like Looker or Tableau or something like that, or Power BI most commonly. And then they were paying a bunch of analysts to basically go in and look at the dashboards and write up narratives and share those with the business. So each department leader, each budget owner, each revenue center leader knew exactly what their team was spending and how their teams were forms of tracking. And that's an incredible destination to get to, but it's so expensive and it takes so long. 

‍

A thing that I mentioned earlier in the back story is I didn't have a background on FP&A teams or being in finance orgs or something like that. I had done data systems work. I had helped hedge funds conquer pretty similar challenges around getting all their data into a single place. And so I can naturally gravitate to that a lot. And so Cashboard is essentially an all-in-one tool that does all those things, including actually writing the analysis for you so we can actually help you avoid having to hire on that front. We can get people live in a couple of weeks with a single source of truth for all their data. And you can just immediately understand how the business is tracking. And you can have the system actually write analysis for you and send it to you whenever you want. I've been talking for a while, so I'll pause there.

‍

Mike Wu

No, that's amazing, the imagery is just very vivid for me. I start off with like you mentioned the solo CFO, I see a CFO and a lean team, which I think is something we're all trying to achieve. And then I see someone, traditionally someone on an FP&A team or many people at a company would be building like these pipes, the infrastructure to connect the data, to make it uniform and usable. And that's something that Cashboard is doing today. And then you also have the analysts, the people who are interpreting the data, drawing out the insights and telling the stories to the business units to make it useful and applicable to their decision making. And that's something you're bringing all together into one platform. Pretty awesome. 

‍

Julian Rowlands

Thanks Mike, that's a great summary.

‍

Mike Wu

 Tell us more about AI. I think to like, like non-technical people like myself, AI can mean something like image generation or just like automation really. I think for me, AI at Hireframe is just a lot of small automations here and there that are making us a little bit more productive. As you're building Cashboard and thinking about the roadmap, what are some of the maybe exciting things you're thinking about and applying AI to or where you guys are today in building some of those features out?

‍

Julian Rowlands

Yeah, so I'll talk a bit about what I think inventing the solo CFO looks like and then how AI can fit that goal. I think it's important to stay focused on what you're trying to help people accomplish and then think about how AI can actually play a role in that. I'll stay very focused on FP&A, which is understanding the business and just making sure it's performing well and driving good decision making across the organization, right? Keeping it healthy and keeping it strategically aligned with the right things for the business. So an FP&A function is really in the business of like, I should always pause. My mom recently Googled FP&A and I realized I needed to stop using acronyms so much. 

‍

Mike Wu

The Mom test.

‍

Julian Rowlands

Yeah. So FP&A is financial planning and analysis. The analysis means how's the business doing. And the planning means how should we drive the business forward to achieve our strategic goals. Which is another way of saying, understanding the data, pulling all the data into a single place and just understanding if things are going well or not, at a pretty granular level if you can. And the planning piece is actually building a budget, building a forecast. And if accounting is money you've moved in the past, a forecast or a budget is your guess of what future money movements will be. Which employees will we pay, which customers we plan on signing on, what vendors will we bring on, what locations will we expand to and stuff like that. 

‍

We've made a choice in the past that's been maybe a little heretical in terms of building FP&A software, which is to completely ignore the P. There's a ton of people out there that are trying to bring financial modeling into the cloud or try to bring planning into the cloud. And I don't think anyone's actually doing a great job of giving people that single source of truth, right? Replacing that heavy duty infrastructure that CFO has been building for themselves for a while. And the more market research we did, the more we talked to people, it was obvious that there are great ways to consolidate your budget across tools like Anaplan or Adaptive or newer things like Runway or Mosaic or something like that. But no one was really creating a replacement for like that heavy duty data infrastructure, right? 

‍

I've talked to ton of CFOs who have something like Adaptive spun up for the budgeting or forecasting workflow, but we're still building that Fivetran to Redshift to post Power BI with Alteryx in there or something for the data labels. And so we decided to stay laser focused on the piece that we felt like no one was doing a great job on, building true heavy hitter data infrastructure for CFOs that they don't need to talk to a data engineer to have set up. They can have it live in weeks for pennies on the dollar compared to what they would have to spend to build themselves. So that's where we started out. And then adding the AI piece in to actually write the analysis for you was a natural extension of that. But the next piece is to enable the solo CFO. That's helping people spin up a budget or a forecast very, very quickly. And so that's one of the things that's coming up on our roadmap. I think that should be out later this year. 

‍

And my dream is budgeting, it takes months. And so much of it is just like data chores, right? Like taking a bunch of templates that you've worked with your department leaders to build for who they're going to hire next year and rolling them all up into one big template for the whole company. Then you show that to the board. The board shockingly says make more money and spend less money. You go back to everyone, you help everyone figure out how to modify those templates and redo that thing.

‍

The conversations are critical, having thoughtful decision makers in the room to actually drive that forward and advance the thinking of the business and the strategy of the business. AI cannot and should not do that. But all the data chores around consolidating data, splitting it back out, writing up analysis and explanations for what has to change and why, when you get some feedback, that's something that I think AI is pretty capable of. 

‍

So to answer your question, we've automated the deep insights and the analysis. Next up is the budgeting and forecasting and helping people build that really, really quickly and enabling the humans to just be strategic, without having to spend a ton of hours on data workflows and data stores. And yeah, that's the next big thing.

‍

Mike Wu

Yeah, super interesting. And appreciate you, sharing a realistic view of how you're using AI in the products today and what that might look like in the future. I'm personally curious, I'm invested in staffing and people, so I'm actually curious, what does this mean for someone that's coming up the FP&A ranks, or maybe go further up the pipeline or down the pipeline to university students studying finance and they want to a coveted financial analyst role at Disney or something like that. Any thoughts around that?

‍

Julian Rowlands

Yeah. So maybe I can answer that by talking about what I think finance orgs will look like, FP&A orgs will look like in particular. Talking about the roles I see on those teams and then talking about what I think will make those roles successful. I think we're going to a place where… there's two types of hiring for finance teams, right? There's decision-making, there's execution. And when I was at Spruce, I joined when there were about 70 employees, when I left there were about 450. I very rarely ran out of time for decision-making, right? And so we got to 350 employees and I said, okay, I don't have time to be in all the meetings and to represent FP&A well. So I hired a director of FP&A who was great and who was able to fill that role and help be the voice of FP&A throughout the org when I didn't have time, because I was in other meetings. 

‍

But I ran out of time for execution all the time, right? Like downloading CSVs and turning them to charts quite frequently and things like that. I think the future of FP&A is decision makers only, right? And so the solo CFO, we're quite a ways from that, but a world where the execution can be handled by smart AI enabled tools that are actually sitting on top of your data flows and have purposeful workflows that match exactly what finance teams need. And so I think the future of FP&A teams is really like just those decision makers with people making sure that the data pipelines are running really well. 

‍

And so it kind of takes us to a place where like, I have some hot takes on FP&A apprenticeships, which is, I just think they're closer to hazing than really like a true like growth opportunity, right? If you're an engineer…

‍

Mike Wu

Why is that? Say more about that. I'm not familiar, I came up through the investment banking route. So there's quite a bit of hazing, like professional hazing in that path. But tell me more about the junior FP&A role.

‍

Julian Rowlands

Yeah, maybe I’ll ask a question on the iBanking side. How many times did someone say like, hey, rearrange these logos on this pitch deck or something like that.

‍

Mike Wu

Yeah, it's dozens of times a day, probably. It's like, Hey, stay up. I'm going to send you some comments on this deck before I go to sleep tonight from a managing director. So you're sitting by your desk waiting for the email to come in and it's like, Hey, we're going to swap out these logos, move these things around and change these colors. That was like the type of work we had to do sometimes.

‍

Julian Rowlands

Yeah, and how much of that feels transferable to being an MD?

‍

Mike Wu

Yeah, none of it. Exactly. It's like none of it. Like the MD roles are commercial, strategic, making those decisions, decision-making for sure. And yet that execution work was what you were evaluated on at that level, but not transferable if you wanted to climb the ranks essentially.

‍

Julian Rowlands

So FP&A is very, very similar where if you move through the FP&A ranks and you become a CFO, you get to be in the boardroom, you get to be helping the founders and the execs make really strategic decisions. You are one of the most important people in the entire business and your impact is so, so massive. But an entry level FP&A job historically has been pretty similar to that banking analyst role you described. A lot of times it's, hey, I need to see what the latest chart showing number of customers and amount of revenue is. Go to our budget and go to our accounting system and go to our sales CRM and download CSVs and clean up all the data labels because you don't have control over those. So maybe the customers all have different names in every system. And then do all this manual work to just refresh this chart manually.

‍

And that's not a real apprenticeship, right? That's just giving someone a bunch of crappy data chores to manually slog through. And if you compare that to engineers, right? Like when you're a junior engineer, your first job is pretty similar to doing what a senior engineer does. It's just scoped really tightly. Someone's giving you more of a breakdown of exactly what to execute, but it looks quite similar to the work you're going to grow into. And I think FP&A deserves the same thing.

‍

So we're kind of circling around this a little bit where if AI can do a lot of that execution work, what does the future apprenticeship look like? I don't have a great sense here. If I could predict the future, I'd be buying lottery tickets right now. But sales and revenue organizations have a role called rev ops and they're very data driven. And the job of rev ops is to make sure all the systems talk to each other to make sure the data labels are really good, make sure things are connected really well and really enable that revenue and go to market machine to thrive. If AI is going to do work like automate a ton of execution work in the CFO world, you're going to need your data to be really well structured. You're going to need your data labels to be matched. You're going need all the systems to talk to each other really well. You're to need someone who owns the prompts and owns the exact way that you your company interacts with the models and your data to drive outcomes. 

‍

And so I think we're going to see the future of apprenticeships like early roles in FP&A orgs be something a bit closer to that. What that in turn connects to in terms of what you should be studying. So I went to Rutgers for undergrad and there were a lot of people at Rutgers who were studying both finance and things like management information systems. I don't know how common that is at Ivy league schools or something, but schools like Rutgers, that was pretty common. I think that's going to remain a really safe bet, right? Understanding your left joins and your right joins and stuff like that and how APIs work and how data flows knit together, understanding what a primary key is. know SQL, understand databases and run towards that. Don't run away from that and don't look for a tool to completely solve that for you. You should really, really understand the data super well.

‍

Mike Wu

That's super interesting. Obviously you've been, you're in it and you've been very thoughtful about this, but you mentioned RevOps and the connection to FP&A. I think that's really interesting. We actually did a research study on RevOps roles. This is probably 12, 18 months ago now, but one of the key insights from that research was that leadership roles in RevOps, director and above, the trend was the job requirements for that role were more accounting and finance related and less sales related. So I think maybe two, three years ago, what we saw was rev ops hires were coming from sales departments or had a background in sales and more and more now it's actually like accounting and finance. And I think we were at some sort of inflection point or we just saw that trend moving. And so I think that is an interesting take on where junior FP&A roles will go. I think that's also better for junior FP&A or junior finance folks to be closer to revenue or to have a straight line, direct impact on revenue versus being confined or in a finance department, quite literally in a cubicle bullpen where you're just focused on the execution work and, less proximity to revenue in the commercial side of things. 

‍

Julian Rowlands

I totally agree.

‍

Mike Wu

I think it's actually optimistic.

‍

Julian Rowlands

Yeah, a hundred percent. I think roles are gonna get more impactful, more interesting, more strategic. I'm curious where you're taking this Mike, people talk about how they think AI is gonna take people's jobs. I think that there's so much room for the money that people are spending on FP&A teams to go on more strategic roles and things like that and allow people to actually drive the business forward instead of doing data chores. And I think people will take that and run with it, right? I don't know if we're necessarily going to see leaner teams. I think we're going to see massive productivity boosts and we're going to see more and more people being hired into these roles as companies understand just how valuable finance roles can be.

‍

Mike Wu

Yeah, I generally agree with that. I think what we're going to be able to do is everyone is going to be more on the decision making side of things, even at a junior level, than on the execution side of things, I think we're already seeing that. When I speak to my colleagues that work in investment banking and the team that I used to work on, what is happening is that like the first year analyst role, which I did for my first year out of college, you did all of that execution work, all of the grunt work, whether it was moving logos, building financial models in Excel. Now things have changed over the past 15 years and that role is less about building the actual model, but taking the model that was built through a tool or through someone on an international overseas team and interpreting it and actually drawing out the insights and manipulating it, changing assumptions, making those like qualitative decisions to make the model very useful in some sort of analysis and structuring the analysis and that sort of thing. So yeah, I think we're already seeing it. So I agree with you. I think we're gonna see more jobs and just all jobs eventually will be decision-making roles, which I think is very good for us, because it'll mean roles will be more fulfilling, you'll have more impact. And I think that's also gonna require a change in apprenticeships and internships and even university level education. I think it should be focused a lot more on the so what, the insights and less so on the mechanics of the work because a lot of that will already be done.

‍

Julian Rowlands

Absolutely. Yeah, I totally agree. And to use the example of investment banking, it's not like, investment banks love making more money and, and spending less money. And they have not, to my knowledge, I'm curious if this is the case, but I don't think they've decreased the sizes of their analyst class or anything like that, other teams, right? They're just trying to do more and more and more. 

‍

Mike Wu 

No, I don't think that's changing.

‍

Julian Rowlands

Yeah. So a good example.

‍

Mike Wu

No, I don't think that's changing. Yeah. So I'm surprised people, yeah, there's still a lot of interested students that want to go into investment banking. This is really interesting. Anything else you want to share about Cashboard, before we shift gears here a little bit?

‍

Julian Rowlands

Yeah, maybe just a quick plug. If you're a mid-sized business, if you're doing 15 plus million in top line each year, and if you are desperate for a single source of truth, where can see all your data really easily and you want to be able to make good decisions effortlessly across the organization. Cashboard can give you a single source of truth connected to your accounting, sales, payroll, et cetera data, show you exactly what's happening across every single slice of the org. And our AI can actually run the workload that used to take a couple of FP&A analysts days to produce. It can write and distribute analysis throughout your organization automatically. 

‍

Mike Wu

Love that.

‍

Julian Rowlands

If you want to be able to make good decisions really, really fast at scale, give us a call.

‍

Mike Wu

Yeah. And if you don't want to ever type in a V lookup formula ever again, call Cashboard. 

‍

Julian Rowlands

Have I got the product for you

‍

Mike Wu

What about index match? You guys, you guys doing that too? 

‍

Julian Rowlands

I'm an XLOOKUP girl.

‍

Mike Wu

No, that's great. No, I'm very excited for Cashboard, what you guys are building. As we're wrapping up here, I have a couple of questions I want to ask you about more along this thread of like careers and to help people think through those things. But first, you're an ex CFO turned founder CEO. Any learnings or any advice to someone who might be a CFO today that's thinking about starting their own business?

‍

Julian Rowlands

This is such a great question. I always have such a hard time answering this because I feel like my story is so, there's happenstance, like accidental things or unexpected things that came about, from becoming a CFO at all to working for these startups that did really well and really turbocharged my career in unexpected, but really great ways. I think maybe a couple of pieces of advice, know who you really are, right? Like know what your strengths are. I'm pretty creative. I'm pretty scrappy. I work really hard. I'm not very organized. I have pretty bad ADHD. I was not a great student in high school. And so I think I knew that and that's part of why I wound up in the startup world to begin with, right? Startups win by figuring out exactly what matters and focusing relentlessly on that and letting everything else burn. If it's not going to actually move the needle for the business, just let it be on fire. And that's a very uncomfortable place for a lot of people, especially a lot of CFOs. And I'm just a weird person who really thrives in that and actually probably wouldn't do well as like a public company CFO. So I think knowing who you are and how your personality works and what you're good at and what you care about and and then in turn, like what you're not great at. I think that's pretty important.

‍

When it comes to actually making the jump to starting a startup, people ask me sometimes, how do you guys get into Y Combinator? I don't think I would have gotten into Y Combinator if I had been building a startup that was not focused on a problem space I really understood. People talk about founder market fit, and I do not have founder market fit for a consumer focused marketplace or something like that, or an iPhone app or something like that.

‍

But as a former CFO who also has this data systems background, I do have founder market fit for like this one very specific narrow slice. And so I think that's the only reason we got into Y Combinator. The fact that we're focused maniacally on something that we had founder market fit for. So I think just really being thoughtful around that and making sure to build around something you know a ton about. Those are kind of the main things.

‍

Mike Wu

Yeah, awesome. That's great advice for any CFOs out there. Maybe thinking about starting a business and one last question here, Julian, for CFOs out there, really anyone, but maybe CFOs, cause that's part of our audience here that are trying to figure out AI. Any advice for actionable things we can be doing or they can be doing to figure it out?

‍

Julian Rowlands

Take a lot of demos, I think is the number one thing. This is moving really fast. Even if you're not interested, just say yes, the demo and let someone show you. And we're in the midst of a paradigm shift and it's kind of hard, all the things that were true a couple of years ago were the only way to make a decision at scale was to invest a million dollars in data infrastructure and throw execution-focused roles at it. A lot of things are shifting. More stuff is going to shift very, very soon as people keep building new and incredible things. And just take the demos. I think there is a lot of stuff that's real. And it's just worth continuing to keep an eye on it.

‍

I would say definitely understand, use ChatGPT, use those things, but don't assume that the caps on your experience using that are the same as the caps on what someone can roll into a product. If you roll an Excel file into ChatGPT and just say, hey, write up some analysis, it's going to get a lot of stuff wrong. The short version is a GPU is just not that good at math, whereas a CPU is really great at math. And you're asking a GPU to do a CPU task.

‍

What I see great products doing really well is they're allocating GPU tasks to GPUs and CPU tasks for CPUs. And they're able to build their own distinct data transformation pipelines to actually have the products to get things right. That if you just drop a spreadsheet into chat GPT, it's going to get wrong. And so understand the tools, but also keep an eye out and take the demos. Yeah, I think, I think that's the main stuff.

‍

Mike Wu

Yeah, things are moving fast. That’s great advice. Thank you, Julian. And thank you for coming on the podcast. We really appreciate your time. Please, anyone listening, check out Cashboard, cashboard.co. They're doing some awesome stuff, as you probably just heard. And yeah, thanks again, Julian.

‍

Julian Rowlands

Thanks, Mike. Appreciate the time.

‍