Christian Martinez, Finance Analytics Manager at Kraft Heinz is an in-demand Conference Speaker and specialist in AI in FP&A who teaches a top-rated course, with (previous guest) Nicolas Boucher. In this episode discover the secrets about game changing uses of AI and Python for your FP&A career.
“We’re in the same era as when Excel was first invented,” says Martinez “There were people still using calculators and pens while others shifted to Excel and dramatically improved performance.”
In this episode:
- Why Python and Excel together at long last as a “game changer” for FP&A
- “Explainable” AI in FP&A
- How AI is improving overall budgets and forecasting
- Can non-data science people jump right in with AI?
- Getting comfortable being uncomfortable
- His path to a “boutique” course teaching practical application of AI in FP&A
- Why the best way to fully learn something is to teach it
- The most awesome uses of AI in FP&A
- How Gen AI is going to change FP&A in 2024
- Waterfall charts in Excel
Follow Christian Martinez (LinkedIn): https://www.linkedin.com/in/christianmartinezthefinancialfox/
Links
FREE COURSE – PYTHON FOR FP&A AND FINANCE. Curated by Christian Martinez
Advanced ChatGPT for Finance course by Christian Martinez and Nicolas Boucher
Paul Barnhurst:
Hello everyone. Welcome to FP&A Today, I am your host, Paul Barnhurst, aka the FP&A Guy. FP&A Today is brought to you by Datarails, the financial planning and analysis platform for Excel users. Every week we welcome a leader from the world of financial planning and analysis. Today we are delighted to be joined by Christian Martinez. Christian, welcome to the show. Hi,
Christian Martinez:
Paul. Thank you. Thanks for having me.
Paul Barnhurst:
No, we’re really excited to, to have you. And just a little bit of a backstory before I need to do an introduction, as Christian was originally scheduled to join us for the AI LinkedIn live episode, but he had to travel, and so he wasn’t able to join us, and so I asked him if he would join me for an episode. He was really excited to do that. So today we’re gonna have the opportunity to chat one-on-one, but just a little bit about him. He’s coming to us from Amsterdam. He earned his bachelor’s degree in international business and his master’s in banking and finance. And he has worked on multiple continents. If I have this right, you’ve worked in the North American continent, I think in Mexico, Australia, and also in Europe, right?
Christian Martinez:
Yep. Correct.
Paul Barnhurst:
So quite a quite a traveled man. So we’re gonna go ahead and start here with one of our favorite questions we like to ask everybody who’s dealt with FP&A, what has been the most challenging or kind of worst budgeting experience you’ve ever had in your career?
Christian Martinez:
Yeah, that’s a very good question. So I think that one of the, I use like woorseones, it was also one of my first ones. So I basically like started in a, in a company like mid budget process. So it was like my first budget in that, that company. And then I was not very clear on like how lengthy was the process what to do, things like that. So then it started to be like very chaotic for me at the beginning. Luckily like because of the support, like of, of a manager and of the teams and so on, everything went well. And then in the future I saw that like if people join like mid budget or, or things like that, then to be like extra supportive with them because, basically it was very difficult to, to navigate, all the complexities of it.
Paul Barnhurst:
A hundred percent agree. Whenever you join in the middle of a process, especially budget, it can be chaos if you, if things aren’t really, really clear what’s expected and what’s needed. So what, what’s your key takeaway or learning from that experience?
Christian Martinez:
Yeah, so I, for I guess like the main one was to, whenever you are in a new environment, try to absorb as much knowledge as possible, but at the same time to allow yourself to be like, uncomfortable with it. So it’s like, it’s okay not to know everything from the very beginning and basically to just like have that, mindset of like, I’m here, like learning and basically trying to absorb as much as possible.
Paul Barnhurst:
I really do like that idea you mentioned of you know, being willing to be uncomfortable. I think sometimes we, we want all the answers and it’s really hard when we don’t know things. And so, and sometimes you just have to remind yourself it’ll come with time. Yeah,
Christian Martinez:
Definitely,
Paul Barnhurst:
Definitely good learning there. So why don’t you go ahead and just take us through your background. Tell us a little bit about yourself, what, you know, what you’re doing today and, and how you ended up where you’re at.
Christian Martinez:
Yeah, definitely. So as you’re saying, I’m originally from Mexico, so I did my bachelor’s in there and then started working in logistics. Then global trade that took me to a couple of travels. So I lived in France in Paris, then in Geneva, in Switzerland, and then in Bratislava in Slovakia. And then after that after a couple of of years in my career, I decided I wanted to do a master’s degree. So I saw this program in Monash business school in Melbourne, in Australia that basically it was called like bank and finance, but it had the opportunity to give me a huge component of the master’s of data science and of data analytics. So I ended up doing a hybrid degree between finance and data science.
Paul Barnhurst:
Okay.
Christian Martinez:
After that I started working in the, let’s say, startup ecosystem in, in Australia. So like for a startup company and then doing a couple of like financial analysis where I send them sales and a lot of like different other things in between. I eventually even started my own startup as well in there. And after a couple of years I joined Kraft Heinz. So basically that one was five years ago. And that took me from FP&A then to finance controllership in one of our factories. And then now more recently to, to Amsterdam to be leading the finance automation and all finance analytics team basically for apac EMEA and Latin America.
Paul Barnhurst:
Great. So you’ve had quite the traveled career. Which kinda leads me to a question. I’m curious, you’ve worked obviously in a lot of different roles,FP&A controllership automation analytics, you did the data science degree. How has having, you know, robust experiences around finance, I would say even small, obviously small company, large company, you know, all over the world in multiple roles, how has that helped you? How has that benefited you today?
Christian Martinez:
Yeah, so I guess like one of the things is that I really learned how to learn things. So it’s like, as I was mentioning, like how to really like absorb information of different domains on different like size of companies or different things in a quick way and to be able to like really use that knowledge. So as I was mentioning before, I always try to like be comfortable with being uncomfortable because I do think that that’s how you really grow your mindset and like your career. So basically trying to like learn as much possible of like different things. I remember even when I started to decide to study finance I knew that it was already like a complex topic and then a lot of things to dive in, into finance itself, but then I wanted to look like, what’s the future of finance? So then that’s why I was like, okay, I need to observe knowledge, not just about finance itself and like FP&A accounting and stuff like that, but also about like analytics and data science and all of these program language, SB intelligence tools and so on. So I guess that that’s the, the main takeaway.
Paul Barnhurst:
Yeah. So it leads me to a follow up question. As you were talking, you mentioned about the importance of getting comfortable being uncomfortable. Any advice you could offer others around that? What’s allowed you to get comfortable being uncomfortable? Because I think naturally it’s hard for most of us, most of us don’t just go into a situation and be like, oh, good, I’m uncomfortable. You know what I mean? <Laugh>. So how have you, how have you got to where you’re that way?
Christian Martinez:
Yeah, no, I think to be fair, it has been like practice. To be completely honest one of the main things like looking back is to, as I was mentioning, like really like having that mindset of like not be like super stressed or something like that, but like really like allow yourself like not knowing things and that’s okay. And then just like having that mindset of like, okay, I am learning, so I’m not going to know everything, but it’s okay. ’cause I think like one of, that’s one of the main things that a lot of people start thinking like, okay, no, I do want to be in control with everything, of everything or like, know everything or something. And that’s where they get like, without this like, being comfortable. Being comfortable.
Paul Barnhurst:
Yeah. So if I’m hearing you right, there’s two things I hear under there. First is, you know, practice, you gotta, you gotta practice allowing yourself to get comfortable in that situation. And the second one is mindset. You gotta focus on having the right mindset, having that growth mindset and being open to challenging situations that don’t come natural.
Christian Martinez:
Yeah, definitely.
Paul Barnhurst:
Shifting gears here a little bit, but recently you had talked about how Microsoft announced Python is coming to Excel. You know, you’d shared that on LinkedIn. So what excites you about that announcement? What, what made you decide to share that and what do you think will happen with that?
Christian Martinez:
Yeah, so ever since I like started learning a lot of these like data analytics tools, like seven, eight years ago I discovered Python, basically. Mm-Hmm, <affirmative>. And then back in the day in order to get Python up and running on your computer, you have to like, it’s open source, so it’s all for everyone, but you need to install one thing and other thing, and then eventually get up to these like very black screen where you didn’t know really what to do. Then you start learning it, and then you start typing code and so on, so on. But it was basically very daunting all of the environment to use Python. After learning it for, for a bit, like for a couple of years, I realized the power of Python for finance and for everyone, like that uses Excel really.
Christian Martinez:
So I decided to teach it. So I was teaching it all the way to people, but as well we tried to, to do it like a little bit, less daunting and a bit more user friendly and so on and so on you know, of my conference and workshops, but still a lot of people were saying like, okay, but it’s a separate thing that I need to learn as separate tools, separate environment, and so on and so on. And right now with this announcements of like using Python within Excel itself, now it’s just a function. So like how you can get Python in your Excel is literally just like equal PY and then you start in there using Python. So I think that that’s a very like, big thing, like really like a game changer for all of us, like in FP&A accounting and finance because it allows us to use like natively the tool that all of us are comfortable with. So Excel but use the power of this program language.
Paul Barnhurst:
Thank you for sharing and a little bit of your journey there. So just so our audience understands, I’m sure we have some people that are quite familiar with Python and others that really don’t know it. You mentioned Python’s a language, but why is it so valuable to finance people? What are some of the things Python allows you to do? Can you maybe share a little bit
Christian Martinez:
Bit? Yeah, definitely. So I always think that Python can allow you to do three, three things. Basically like improve the way you analyze data. Then make consolidations or making manual tasks faster or automated. And the third one is basically improving your ways of working as a whole. So in the first part on the, the analytics space with Python, you can create, let’s say better and more powerful visuals or better and more powerful like forecast, for example. Then on the second one with like automation, consolidation, all of those things a lot of times people inFP&A they struggle that they need to do the same task every day or week or a month or every year with budgets or, or something like that. And it’s very time consuming and it’s very manual. So with Python, then what you get is a way to automate those tasks and really relieve that time so that we can spend that time better elsewhere, either like business partnering or even if you are spending that time at 7:00 PM then you can shift yours and then actually spend time with your family, for example. So like, so many like, better things. And the third one, just like improving like ways of working. I do think like by learning Python or any other program language, it also shifts a little bit your mindset and then it allows you to understand data better, which at the same time it’ll improve the way you work and way you interact with your with your data.
Paul Barnhurst:
Thank you. That was a great explanation. I appreciate, you know, the three points you made there about helping with analysis, automation and this improving the way you work with data. It’s something new. You learn as you learn new tools, you’ll learn new ways to work with data that you might not be able to do otherwise. So with that being said, how do you think it will change the way most people work in Excel? Do you see it having a big difference in how finance people work in Excel?
Christian Martinez:
Yeah, so like, I do think that with all of these new tools, like it could be like Python, Excel, or even like ChatGPT and something like that. Mm-Hmm. <affirmative>, it could be. It, it’s going to be like a set of people that know and understand and embrace these tools that will allow them to really grow their careers and also improve the way they work on things and other type of people that they don’t want to embrace all of these new tools and they want to, remain on, let’s say just Excel without Python or without ChatGPT and so on. And basically they’ll be like a little bit more like okay, maybe it’s stay in the market, but they want to be able to do all of these like powerful things. I was not present back then, but I imagine right now we’re in the same era as whenever Excel was first invented.
Christian Martinez:
You can do accounting or financial planning in a piece of paper numbers and so on. And then came up and with a calculator, then became Excel. And I imagine there were a lot of people that they did not wanted to use Excel, and they still needed like piece of papers, calculators, pens, and so on to do the analysis. And again the people that shifted and that they embraced this, new technology on, on Excel, I’m sure like they improve the way they do work a lot more rather than the people that stay, with the calculators.
Paul Barnhurst:
Yeah. You know, it’s like anything you have, you have early adopters, you have the general population, then you always kind of have some laggards. Yep. Whatever it might be. Whether it’s ChatGPT, Python, Excel, computers, you know, I remember the early days of Excel, and I tell this story, and I think I may have even told it in a prior episode, but I really like it, is I was taking a VBA course at when I was at American Express, they were teaching some basic VBA. And remember the guy gets up there and he tells us the importance of automation. He goes, I had started my career, it was like my first day at Amex. I sat down and this guy was showing me a process, you know, and he had his 10 key calculator, the old 10 keys to the side with the little, you know, paper ribbon.
Paul Barnhurst:
And then he had his computer and he had a spreadsheet open, and he typed in a bunch of numbers, one row after another, like, you know, 20 numbers. Then he went and added them all up on his 10 key and took that number and copied it to the bottom row. And the guy that was with him said, I’m pretty sure it will add those together for you. And the guy goes, I don’t know about that, but this is how I do it. And just moved on. Like, I’m not even gonna try to learn that. It was just like, it was just one of those where it’s like, you know how much time you could save yourself by just listening for a minute. And so I think sometimes we all do that in different ways where it’s like we just listen to others or try something. We could save a lot of ourselves a lot of time, but it takes that right mindset and that willingness to get out of our comfort zone, as you mentioned.
Paul Barnhurst:
So, you know, mentioned earlier ChatGPT and just technology in general. So I know you’re a big proponent of technology. I know in your current role you do a lot with that. So, you know, where does that passion come from? Have you always been interested in, is it something that just kind of came when you did your data science degree or how have you developed such a passion for utilizing technology?
Christian Martinez:
Yeah, no, it’s a very good question and it actually I was thinking about that and I feel like it, I have always liked technology just and like futuristic things. But one of the main things of why I was interested in technology within finance is because I had a, let’s say, passion or interest on doing things better than what it was done before. So I started looking of ways to do it better and it was like, yes, you could you would need to increase your knowledge. And as well, like I tried to do that as well, but I found out like with certain tools that maybe they are not like fully applied, like back then in finance then I could do things even like faster or better than the normal regular waste. So that’s why it’s like started that let’s say curiosity of like understanding all of these tools.
Christian Martinez:
Actually, my first one was like in Excel was VBA as well. So I like learned a little bit of that and I was like, okay, yes, it can improve the way you do certain things. Then my second one, which I didn’t continue that much, but it was SaaS, so like SAS mm-Hmm. <Affirmative>. And that was basically like, they told us at school and then still like they told us like, oh yeah, you can do this, you can do that, and so on. But then I found out that the licenses were very, very expensive.
Paul Barnhurst:
Yes.
Christian Martinez:
Then someone in my like more like data science classes, they told me about like, oh, there’s this thing called like open source, languages. And they told me about R and they told me about Python and so on. So I was like okay, I’ll learn both. So I learned a little bit about, then someone told me about SQL, I learned that as well. And so on and so on. So I guess like every tool that I discovered I found like, okay, it could be just to improve the way people do things. And that’s why like, I guess the, the passion came of really improving the, the way people like handle data and handle things and problems.
Paul Barnhurst:
It sounds like there’s two things that I hear there. One, you’ve always had a passion for kind of futuristic and technology and where things are going, but your passion to learn these things is really fueled by trying to automate and streamline processes, whether it be SQL, V-B-A, Python R, whatever it might be. It was a lot of, it was about, okay, this will make me more efficient. I’ll be able to save time and focus on what’s important.
Christian Martinez:
Yep, definitely. And apart from that, that just, automation as well, like better at like analytics as well. Mm-Hmm. Because I saw like, let’s say forecast, like if you do like a run rate, okay, it’s a forecast, but then I discovered this let’s say more advanced way of forecasting with like machine learning and all of these like algorithms and so on. So I really like was like very excited about like, okay, in this way you can really do like a better forecast, let’s say. Or even of like, okay, just like with more like descriptive analytics, okay. With business intelligence tools such as Tableaux power BI , you can really see what’s happening with your data and then you can advance to a more like prescriptive analytics of like, okay, now you have, you understand what’s happening. Then how can you make it happen a better future for your company or for your like like problem that you’re trying to solve.
Paul Barnhurst:
Yeah. Love via point on analytics, right? Often in FP&A we get stuck on the descriptive. We get a little bit into the diagnostic, but we often miss the predictive and prescriptive. And the predictive and prescriptive often take machine learning data science tools that go beyond what some of us may know or what we’re used to. You know, I’m a big believer that using statistical modeling, I haven’t done, you know, algorithms or machine learning, but a lot of times trying to use statistics saying, okay, is there a regression here or what, a couple variables, things like that. I remember, you know, one of the companies, we had a, we had a tool called jump, which was a very simplified version of SaaS that they, they put out and we used that for a lot of analytics and it definitely made a difference. So, question for you, and I’m gonna go back to the manual work for a second, then we’ll talk more about analytics. But I’m curious to get your perspective, because I think there’s a debate there between how much of, you know, the manual work that goes on in finance today is due to lack of tools. We don’t have the proper tools to do it, because I’ve asked that before in post, and most of the time most people say it’s due to tools. Are you think most of it’s due to skillset and mindset, kind of based on your experience? What have you seen in that area?
Christian Martinez:
Yeah, no, no, that’s a very good question and I actually will divide it in three and Okay, perfect. Have like 33.3% <laugh>. The first one, it’s, for sure as you mentioned, like tools like without like tools, like you cannot do like, let’s say like automatic process and so on. And it can be any tool, like it can be Excel, it can be Python, or it can be like a more advanced, like Microsoft Azure, for example. Then you mentioned as well like skills and mindset that it’s also really important. And then even if you have the best tools, if you do not know how to operate, and if you do not want to operate it, then it won’t work as well. But then the third one that I will mention as well, it’s basically the process itself. Because a lot of times you do not need a better, the best tool and then you do not need the skills to handle that best tool or anything.
Christian Martinez:
But if you have a better process, then it can already be streamlined a lot, which is, I guess like combined with what you mentioned, like mindset. So like that mindset process, I would put like on the third part of like really having that hunger for improving the way you do things every day. If you have that, then it’s like way more important than let’s say like just having the right tool and just having the, the right expertise. But having said that, I would put like those three things, let’s say for like people to, to really get to, to having
Paul Barnhurst:
Great, great point. I like how you’ve mentioned process because you know, people have the right mindset and you’re constantly striving to improve processes you’re gonna develop your skillset even if you don’t have the tools, you can get a lot of things done. You know, I worked for a company where our tool set lacked severely and I had to do a ton of stuff in Power Query that really should have been done in a database, and it was able to make a lot of difference in the company because I continually challenged myself and tried to figure out how to do it. ’cause It’s like, I can’t do this all manual, you know, this just doesn’t make sense. And so I had to find ways to automate it and it, you know, it allowed me to make progress. Could have we done more with the right tools? For sure.
Paul Barnhurst:
Could it have saved me a lot of time? Yes. But I think the biggest thing it taught me was with the right mindset, you can do what we call a digital transformation. You could start that transformation journey and over time, hopefully you get the right tools. And what I’ve learned is tools is the enabler. If you’re not fixing the process, the tools aren’t gonna solve your problem. One of my favorite quotes, and it’s by a consulting firm, Hasker or something like that, he says old processes plus new technology equals expensive, broken old processes. You know, <laugh> and I, I just love that because it’s the reminder that you talked about, you really emphasized so well there is the importance of the process and challenging the process at every step.
Christian Martinez:
Definitely.
Paul Barnhurst:
So I know you’ve been a big proponent of generative AI. I know you’ve talked about that quite a bit. I believe you have some articles and things, you know, what, what have you seen that’s made you such a big proponent? What is it that had you jumping on the bandwagon so early so to speak?
Christian Martinez:
Yeah, so like, I guess like as I was saying with like my, like data science background, I have always experimented on ways mostly to use machine learning as what was mentioning, like, like part of the whole umbrella of like artificial intelligence and Sure.
Christian Martinez:
So I knew already the power of like machine learning but then especially with this, latest cases of the journey t with the ChatGPT at the end of last year and the beginning of this year I really like how to, how they make it so easy to use for everyone. So when I discovered first time, like ChatGPT started like trying to use it and especially with those little use cases that they have on their like the, the web app let’s say. And I started like trying things. And from there I became really excited because everything that I was like, keep on trying, I, it just gave me like more and more ideas. So I guess like from the start of like January, February, March and so on, I was spending like almost all my evenings, like from seven to 10:00 PM just like playing with like ChatGPT and then trying to do like different things and then trying to like see what it can be done, why it cannot be done, and so on trying to do a bit of research of other people as well.
Christian Martinez:
And then eventually I decided like, okay, yeah, I do need to share this knowledge because I know that it’ll be very powerful in finance in FP&A and accounting. So I tried to do that from there because like through various means but it was basically because of that excitement on how powerful the tool was.
Paul Barnhurst:
Appreciate that. So stepping back for one second before ChatGPT, you mentioned you started on machine learning and it reminded me of something I wanted to share. So FP&A Trends, which I believe you’ve written some articles for Larissa, right? Yeah. they do an annual survey of fp and a professionals and they ask ’em to rate their budget and they break it into three categories, good to great average, basically useless or not, not good. And on average, this is last year, so 2022, they did one again this year, but the prior one, 39% of people said they had a good to great budget or forecast. I think 16% said they were useless, which makes me wonder why they were doing it. But that’s another story. And it got better. You know, if you had, if you did driver based, it got a little bit better, not much like 1%.
Paul Barnhurst:
If you added cloud, it was about 50% if you were using a cloud tool to help with the process. But those that were using machine learning, it jumped all the way to 61%. And I think only 4% basically said their plan wasn’t any good. And so you saw a huge improvement in the overall budget and forecast by utilizing technology with human judgment and human expertise. And so I just really like how you mentioned machine learning, because that’s always stuck with me and the value. That’s why I like seeing things like what Microsoft has done with releasing theirFINN their open algorithm they use for their finance department. You know, Facebook has done the same with profit. Yeah. So there’s a lot of things out there that, you know, with ai, getting back to what we talked about, it’ll be much easier for people to use some of those algorithms without have to necessarily understand the details, go through the schooling and, you know, under understand how it works, but to be able to look at it and say, okay, yeah, that makes sense, or no, I would adjust it here. Kind of what are your thoughts on that? Do you think, you know, with AI it will make, it will open up machine learning and that ability to, you know, use that data for those who aren’t maybe, you know, technically inclined, haven’t learned programming languages or algorithms or those type of things.
Christian Martinez:
Yeah, no, that’s a very good question. And I was actually hearing on one of the conferences in here in Amsterdam about like AI and technology and finance, that one of the main things that many people are trying to do, it’s explainable AI. So like it can be, let’s say these machine learning algorithms, but which one of them that are actually can be explained in simple terms of why they’re behaving like that to other people. There are some algorithms that they really cannot be explained, and it’s like a black box approach. Sure. will tell you like, okay, your budget is the next year your performance will be minus $10 million ebitda. And then if you just have that, it won’t be useful for anywhere. But if you really know like, okay, it will be minus $10 million because of driver A, driver B, driver C, and because of these conditions and so on, then you can understand, okay, what can I change so that it became better, let’s say. And then with all of these machine learning algorithms, the main thing, that I always like trying to get as well is like this explainability of the algorithm and then how to show it to people without a technical background, but that they will be able to understand what are the levers that they need to pull in order to do certain thing, whatever, like the actual algorithm is doing, either forecasting, budgeting or, or things like that.
Paul Barnhurst:
Got it. I appreciate that. Yeah, it’s, it’s exciting to watch and I love how you called it explainable, right? If you just say, Hey, here’s your forecast that’s down $10 million load it.Okay, that’s gonna, everybody in finance is gonna immediately go I don’t trust it. Yeah. Help me understand, right? Because I know finance people hate a black box and as a general rule, you know, you’ll pry excel out of our cold dead hands as a lot of people like to say, right? So having that explanation of, oh, hey, $3 million is due to this driver, four is this, three is this, and then, you know, providing context of saying, well, the following changes could help reduce that. Like you lost some salespeople if you hired, you could reduce that risk, you know, giving you some options or things to think about.
Christian Martinez:
Yep, definitely.
Paul Barnhurst:
Yeah, it’s, it’s amazing to think. So if we have someone in our audience that’s listening to this episode and they want some ideas on where to start with generative AI, what would you tell them? Where should they start?
Christian Martinez:
Yeah, so like, I guess like it really depends on what thing you want to do with generative. If you want to juice it, then there’s like one way. But if you want to build your own, there is another way in the terms of like using it. Like one of the things like it’s really if you try and open any tool, like either Google Bard or ChatGPT and so, so on and so on and then you start trying to do things, then you will get into one stage. Then basically the second part is like trying to get to okay, how to do things in FP&A in finance and accounting for your work. And then you’ll hit again another spot of like, okay, now I know those type of things. Then if you try to learn from other people that have done already that then you can understand really how to apply to your work or in a more, let’s say quick way.
Christian Martinez:
So like instead of spending as I was mentioning, three hours of your evening every night during months, then you could spend like two hours of your evening with one person that have spent all of this time, and then they can tell you basically what to do. Or you can read articles, books, or something like that. Of people that are again, have spent all of the time and then they will be able to tell you what to do in terms of the practical application, then if you really want to know about them. The second part that I was mentioning of like coding your own AI, that one for sure is a more technical approach that you can do as well. But then I would recommend them to really learn a little bit more Microsoft Azure and more like programming.
Paul Barnhurst:
Sure. So it sounds like for you, learning how to use the tool, you know, there’s a couple things. One, just starting by experimenting, whether it be Bard, whether it be ChatGPT, whether it be Quad AI or any of any other tool out there. So many are using AI today. Start experimenting a little bit with it first, just kind of on a personal, just trying different things, then try to learn some work related things. But also another great way to learn I’m hearing is go ahead and read about and learn from others like yourself who have spent maybe 500 hours already using the tool and can save you some time. Just like with anything, there’s gonna be those people who have put in the work and will be able to allow you to get up to speed quicker.
Christian Martinez:
Yeah, definitely.
Paul Barnhurst:
Any, you mentioned, you know, reading books or resources, any good articles or books or maybe videos, anything you’ve seen that you really like around chat GPT?
Christian Martinez:
Yeah, so like I all also discovered like eight years ago, but like there’s this website called Medium that is basically for, for articles and blog and things like that, but a bit more it, it’s about everything but about like technology. It’s very, very useful. So it has been like my go-to source from about like eight years ago in all of these topics. Now I even became like a writer myself as well in there. But then I do think that that’s one of the very good websites to, to get to know about like your general I, or that’s in finance as a whole.
Paul Barnhurst:
Great. So we’ll have that in our show notes, you’ll send us that website. We’ll add that if there’s a couple articles you like, but I, I agree. That’s a good, good resource. So next I wanted to ask, I know you teach an advanced chat GPT course for finance professors, professionals. I believe you teach that with Nicholas Boucher, if I remember right? Sure. So how did the course come about? Tell me a little bit about the course and what people can expect from it.
Christian Martinez:
Yeah, definitely. So we both like Nicholas and myself were both like again putting all that work and then a lot of hours like researching and doing things like with the and then started to like share our knowledge mail on LinkedIn. So then one day we contacted each other on LinkedIn and say like, oh yeah, look, we have very good content on this. I have very good content on this, and so on and so on. And then how to really make it more accessible for more finance folks. So then we decided to create this course in, Maven and basically it’s a cohort based, so we do it every like one or two months. We get together around 20 to 30, like finance and professionals. Mm-Hmm, <affirmative>, we don’t want to do it like very like, we a lot of people in the same one because the way we teach is like, it’s very, very practical, but at the same time, like people come up with their own dataset, with their own problems and we solve them like right there in the course itself.
Christian Martinez:
So it’s like a bit more like let’s say boutique approach than your normal clinical learning course. And we really try to get to people to an level of expertise in JGPT so that they can really either automate their process in the work or improve their way. They do their normal things at work in finance.
Paul Barnhurst:
Great. So it sounds like a very hands-on type course. I mean, obviously you have your frameworks and the things you teach, but you want ’em to bring data sets. You want them to be practicing during the course so that they walk away with something they can take with them that will allow them to be more efficient at work.
Christian Martinez:
Definitely, yes.
Paul Barnhurst:
Awesome. Cool. Well, we’ll definitely put that in the show notes as well. What’s been your favorite thing teaching that course? What have you enjoyed most?
Christian Martinez:
The learning actually, because a lot of times I, I, I always believe that the best way to learn fully something is by teaching it <laugh> because there were never, I will, I’ll be like teaching this like so many like questions or even examples of like, oh, I have this problem at work, can we solve it? So we do it in two days and lot of times the first day we get that question and I be like, don’t sleep until I get like all the answers for those questions. And I’m like, because of that, I learned so much more because there are so many questions from either other parts of finance a bit more specific that I couldn’t post it myself. But then now that someone says it, then I really, really in until I get to know how to to do it.
Paul Barnhurst:
That’s great. And I’ve done that as well, gone down the rabbit hole. I get a question from someone, whether it’s Excel or whatever, and can you do this? It’s like, well, I’m sure you can, but I have no idea how. Let me see if I could figure that out. And then you send an email going, okay, here’s how you do that. Yeah, I, I know what you’re talking about. I’ve been there. Yeah, you wanna figure it out. And I’m a hundred percent in agreement with you. If you wanna learn a subject well or know if you know a subject well teach it, because that’s when you really realize, do you grasp the concepts where you can make ’em simple and help others understand them? So I’m curious, what’s the coolest thing you’ve seen done with generative AI? Like, what’s your thing that maybe just kind of has blown your mind or like, wow, that is just amazing. Anything stand out for you?
Christian Martinez:
Yeah, so, so, so many things so far, but one of them, kind of new ones. So within ChatGPT if you get the premium subscription, you get this code interpreter tool Mm-Hmm, <affirmative> that’s basically run Python code actually. Yep. But then you can input some data of like, let’s say like some publicly available data. Yep. ’cause also never input private data, but publicly available data into their and Excel file. And then with a series of commands, you can say like, analyze this data, give me some descriptive statistics. Give me a bar chart that says like, okay, how many assets we have on this. Gimme things, gimme a forecast using linear regression and suggest me next steps. And then it will do everything automatically for you step by step if it doesn’t know how to do something. Or even if it has a problem, it solve it already.
Christian Martinez:
Again, like it tells you like, I encounter a problem on something, but then I’ll redo it and then redo it and redo it until it get every answer for you. Even as I was mentioning those next steps of like, nowthis suggested to do certain things and you can even say like, export all of this as a PDF. So then obviously once you check the answers, once you, you check if the analysis is sound, you can share it already,within just like two minutes of checking and typing that,prompted there. That’s I think one, one of the main things that really,was a very interesting,for me.
Paul Barnhurst:
Yeah. I, I love code interpreter. I’ve used it as well. I’ve used it to you know, analyze a public company, like give me the following ratios, create a horizontal analysis, you know, with showing one of the classes we have the things that we could do. And in one case I’d taken, you know, one company’s data off Wikipedia, just scraped it with Power Query. Yeah. Loaded it straight to code interpreter and had it summarizing a bunch of their data. And, you know, I think they were quite surprised. It was funny, one of the leaders goes, wow, that data’s available publicly was more worried about that than the fact that we could analyze it all. And I just kind of laugh, I’m like, got it off Wikipedia, you know, but it is, I’ve been amazed some of the things with code interpreter and the different tools out there.
Paul Barnhurst:
So definitely an exciting time. And I think for me, the thing that’s most exciting is to see how this will continue to help us be more efficient. I’ve heard it called the productivity revolution, and I really think that’s what it’s about, is being more productive and hopefully it will allow us to continue to eliminate more of that low hanging fruit. I know stuff you work on, having worked in automation and analytics and helping people be more effective so they’re not spending as much time cleaning data and prepping data and all those types of things so they can really focus on the insights and driving value for the business. So, you know, with that being said, how do you see ChatGPT and just generative of AI in general, changing FP&A moving forward? Like how do you see maybe over the next year or two and then say five years from now, how do you think it’s gonna kind of unfold if you were to guess?
Christian Martinez:
Yeah, so like, I guess like right now as, as I was mentioning there’s a lot of potential already that people can do, like in ChatGPT with the tool, like as of it’s, today it’s just that adoption curve, right?
Paul Barnhurst:
Yep.
Christian Martinez:
So then I do think that like the next year, for example, that adoption curve will be more like, okay, of people really using the tool and just like a very small people to be like more like laggers, like not using it for the same use cases that we have right now. So that’s one thing. But then the second one as well is like, I have seen this, let’s say beginning of kind of like decision intelligence, that one of the things that a lot of people like in FP&A do is like analyze different data and so on, then put it on different graphs and then present it to different stakeholders in a PowerPoint on some slides or some sort, and inputting some commentary in there to explain that this, decision intelligence can generate all of that for you, like automatically.
Christian Martinez:
So then you can spend more time like really understanding the business and knowing, again, that part that I was mentioning on the prescriptive analytics, like knowing how we can improve our performance, knowing which business unit we should like, really help, which category we should go to, and so on and so on. So I do think, like with them, let’s say all of this just like commentary making or variance analysis, or just like more like descriptive analytics, things that can generative AI can do for you, then, FP&A professionals will have more time for the other part.
Paul Barnhurst:
Got it. Well, it’s definitely exciting. I appreciate you sharing kind of where you see it going, and I know I look forward to seeing, you know, what additional tools and things continue to come, but even today there’s a lot we can do. And if you haven’t experimented, the biggest thing I tell people is experiment. Just start learning, start trying new things and seeing how it works. Don’t wait till it’s on your desk at work and you’re expected to use it to figure it out. Start now.
Christian Martinez:
Definitely.
Paul Barnhurst:
So, all right, so now we’re gonna move into our section. We call it the get to know You section. I got four questions I’m gonna ask you. You don’t get more than 30 seconds to answer each, each, or any of these questions, so keep it kind of brief and we’ll run through each of them. So the first one is, what is something interesting about you that not many people know? Something we wouldn’t find online? Something kind of that makes you unique. Unique,
Christian Martinez:
Yep. I guess like one thing that I was thinking of that is that I have done a lot of extreme sports, like paragliding, skydivingscuba diving and so on, mainly in Australia. But then the main interesting part is like, I did not enjoy them at all. So I tried them once. I didn’t wanted to try but I was like, not for me.
Paul Barnhurst:
So you didn’t like any of them,
Christian Martinez:
Any of them? <Laugh>
Paul Barnhurst:
I was gonna ask, which was your favorite, but you answered that. So just wanted to challenge
Christian Martinez:
Yourself. I in general, like more not extreme sports. Like I, I love football for example, that’s also not many
Paul Barnhurst:
Now, which football are we talking soccer, football, or American football?
Christian Martinez:
No, like soccer, football.
Paul Barnhurst:
That’s what I figured. I just had to make sure <laugh>. Yeah. That’s funny. So if you could meet one person in the world, dead or alive, who would you meet and why?
Christian Martinez:
Yeah, so this one I guess like, it’s kinda a cliche for like people within my domain of like technology and futuristic and stuff like that, but not now, ever since I was like younger, I really was like looking up to Elon Musk because of all of these ideas of like how to improve the world, the world with like, okay, Tesla, and they’re like, okay, self-driving cars. But at the same time, like non-fuel cars and then like Solar City that was basically like solar panels and then SpaceX, they try to go Mars and so on and so on. So I would say like him, that would be like the main person I would really want to, to meet.
Paul Barnhurst:
That’s a, that’s a great one. He is definitely pushed forward technology and done, done a lot, so he’d be a fun one to have dinner with, so to speak. Right. So this next one we like to ask is just kind of curious to see what you say. So what is the last thing you Googled looked up on YouTube or asked generative AI about that was related to finance FP&A or Excel?
Christian Martinez:
Yeah, so I guess like, it’s similar to what we’re discussing, but the latest one, it’s how to get for everyone Python within Excel. So like right now it has been launched, it has been announced, but then it really depends on the computer that you have on the system that you have and so on and so on. So like really trying to dive more of like, okay, how to tell people to really be able to access it. Maybe for some of them now I have discovered it need to be buy a new computer. But then for a lot of them it’s can be like, oh yeah, you need to do this step, this step, and this step. So that was the, the last one that I was like Googling and getting the rabbit hole of finding out.
Paul Barnhurst:
I, I’m not surprised I could see you doing that. So this one is one we ask everybody. We’ve asked it since the beginning of the show. What is your favorite thing about Excel feature function?
Christian Martinez:
Yep. So I have two if that, if I may, but the first one is we’ll
Paul Barnhurst:
Give, we’ll give you two.
Christian Martinez:
It’s the automatic waterfall charts. So in finance there’s like these like wonderful charts that are basically start somewhere then up and down, up and down, and then end up somewhere. I remember when I first started with like work in finance, there are a lot of people working on this, very big file that they have, like, I feel like five or six different rows and different columns, and they’re telling me like, yes, the numbers need to be exactly in this way and then this calculation and so on, so on and so on to make this graph the wonderful, like charter graph. And then I discovered that like if you just put like the numbers in a row and then like, okay, generating like the workflow chart it was, was done automatically for you. Yes, you need to change a couple of things and totals and colors and so on.
Christian Martinez:
But it was basically that. So I think that was one of the, the my favorite features where it was like the launch, the second one just because it’s flexibility. It’s a very simple formula, but like SUMIF, so especially before Python, before Tableau, anything else, if you are really just working in Excel, what I normally use to do is like having all of my data in one tab as database, let’s say like kind of like database for whatever type of analysis I used to do. Yep. And then mainly with some functions, then I could generate like any type of like your, table I wanted or like graphs and so on in a very flexible way.
Paul Barnhurst:
Yeah, I, I agree with you both. Great. tools, the SUMIFs and the new waterfall chart. I remember when it used to not be automatic in Excel and you know, different ways would, people would create it. And I had one boss and it’s probably the best method I’ve seen, but it was complex, is it was using a scatter plot to build the waterfall. Using some error bars and things like that. And it actually gave you more control than the old, some of the old, you know, methods. But it wasn’t easy to follow. I remember reading the article the first time and I’m like, what in the world are they doing? But after a while using his file, I got the hang of it. So I, I was grateful when they came out with the waterfall chart because it made it so much easier for everybody.
Christian Martinez:
Yeah.
Paul Barnhurst:
Alright, so we’re heading into our wrap up here. Just have two questions left for you. So the first would be, what advice would you offer to someone starting a career in FP&A Today? So they’re just starting out. What’s the advice you’d give them?
Christian Martinez:
I would say like try to learn one new thing every week. That would my main thing. It could be analytics as I like, for example, but it could be as well like dive more into the business and knowing more about like, the categories that you’re like supporting with or, or learning about like the factories that you are in your business unit or learn about. Like, I dunno, like capital expenditures, whatever, but like if you really like learn one new thing like every week, like it really adds up.
Paul Barnhurst:
Great advice. So learn something every week. And so last question here. If someone wants to get a hold of you, if they wanna reach out to you, what’s the best way for them to do that?
Christian Martinez:
Yeah, definitely. So I think it’s LinkedIn. I’m very active in there. A lot of people have reached out to me in the past as well. I always try to be checking up like my messages and so on and trying to like help people. So yeah, feel free to, to reach out.
Paul Barnhurst:
Well, thank you for joining us on the show. We really appreciate it and we’ll mention that in the show notes and some of the other things we’ve talked about. I’ve enjoyed the discussion. I’m sure our audience will as well. So we’ll let you go and enjoy your evening as I know it’s nighttime over there. But thank you for carving out some time today, Christian, really appreciate it.
Christian Martinez:
Thanks so much for having me, Paul.
