If you would like to earn CPE credit for listening to the show, visit earmarkcpe .com slash FPA.
Download the app, take a short quiz, and get your CPE certificate.
Finally, if you enjoy listening to FPA today, please go to your podcast platform of choice, click the subscribe button, and leave a rating and review of the show.
And now, onto the show.
From Data Rails, this is FPA Today.
Welcome to FPA Today.
I'm your host, Glenn Hopper.
And joining us today is Anna Tiamina.
Anna is an accomplished finance leader with over 15 years of experience.
Anna has held key roles such as chief financial officer at SoftTech and is now the founder of Blend to Balance, where she combines her expertise as a fractional CFO with her passion for AI integration.
She has guided companies through complex budgeting cycles and financial transformations across multiple locations, all while leading distributed teams.
Anna is here to share her unique insights into the challenges of FPA, budgeting, and the evolving role of AI and finance.
Anna, welcome to the show.
Thank you for having me.
It's a pleasure to be here.
So, we're going to get to the AI stuff.
And it's funny because I always get very, you know, just with my background, I get very excited when we have guests on who are equally as excited about AI.
But I'm realizing as this goes on, that's getting to be pretty much every guest.
And if you're not excited about AI right now, I'm wondering what rock you're living under.
Yeah, a lot of people are not there yet.
So you are losing a bubble.
Sorry to say that. Yeah, well, you know, and that's actually and that'll be fun to talk about too, because, you know, as finance and accounting people, I think we are inherently risk averse.
So there's still a lot of people out there who are thinking, I'm not ready for this and all that, but it's changed.
A lot of people. Yes.
All right, well, we're going to get to all that.
But I guess before we do, walk me through your career journey, and how you came to the place where you are now, where you decided to go out on your own and do the fractional CFO and AI integration consultant piece.
Yeah, well, it's actually not that easy to remember, but I've been a full time CFO, pretty much classic CFO role.
And I spent five years in a huge pharmaceutical organization, a corporation with offices in like 150 countries, a huge organization.
And they moved me between the roles, because this is what corporations do.
And then I switched to a much smaller company that was doing it consulting and software hardware development.
So a much more technical company, but also much smaller.
Both of the companies had international setup.
So I was involved in a lot of inter -country activities in both of my roles.
And I spent another five years in this IT company.
And early this year in 2024, they laid me off actually, as a part of a big restructuring, and as a reflection of what a lot of companies in the tech sector are now doing.
Somehow I got other companies contact me and my ex company is actually one of my first clients as a fractional CFO.
This is a very common story in a fractional executive story, right?
Somehow I naturally started to get companies contact me.
And I started this fractional CFO business, which I thought would be just something in between the full -time roles.
Since I got some time capacity during this period, I thought, what should I learn that would increase my value on the job market?
And the job market now is not really easy as a lot of people know, right?
And instead of getting another certification, I thought, okay, what's hot in there?
And AI was really, really hot.
And in this tech company, there was a lot of innovation around.
So I was very open to looking at ways to innovate.
But this is so hard to get into something new when you are in a full -time executive role.
You don't have enough time to even start looking at these new things.
And AI space is so dynamic, you need to block some time to get there.
So I had this time, I started to read, to watch YouTube videos, whatever.
And then I started to experiment in my fractional CFO work.
And I learned that the only way to actually start using AI is to start using AI.
Like learn by doing, there is nothing else.
And I'm a part of a couple of CFO organizations, and I started to talk about what I'm doing there.
And people get interested.
And they start saying, okay, why don't you do a webinar for us?
Why don't you get a group of people and tell them what you're doing and how you're doing that.
And eventually, my business grew like a second leg, where I do the fractional CFO work now.
And I don't want to drop it.
But I also do the webinars, I do the workshops with the companies, helping them to start implementing AI.
Because this is very, very difficult to do on your own if you are in a full -time role.
So this is where I am now.
The idea is that I will continue this both parts, because I think there is a lot of synergy between them.
But we'll see. And you write the AI stuff is very exciting.
But the classic CFO work is also something that inspires me a lot.
And I really love doing that.
Yeah, and I think what you'll find is you get into this more.
And as you take on more and more clients, there's going to be more people who are, you know, quickly become experts in AI, but that domain expertise that you developed through your finance career, that's what's going to make you stand out.
If somebody's looking to implement AI in finance, you want the finance person, not just the straight AI person.
So I think you're in a great spot.
Yeah, this is what I think.
You're right. And also, finance people are different.
They're much more conservative, they are risk -averse.
And you need to keep all the things in mind and not just offer them solutions that will not work for them.
So typically, I think we could spend the whole episode talking about AI.
But you and I were talking before the show about we're in budget season right now.
I think everybody's so busy, we actually see a little bit of a decline in listenership, because everybody's working on budget.
But I thought it would be fun to actually go through and do some budget war stories.
And I do want to get to those because I know you'll find a lot of us shaking our head and saying, Oh, yeah, I've been there.
I remember that. But let's back up a little bit before that.
And one of the things that our listeners are always interested in is hearing.
So I know you were at the at the huge company before, and then the smaller IT company, but sort of hearing how fp &a teams are set up, and how you manage and how the responsibilities were.
So either looking at your last company, or even when you were at the larger company, maybe comparing and contrasting.
Tell us a little bit about the teams that you ran.
Yeah, it's not easy to compare them because the big company and the small company are like two different worlds if it comes to fp &a right.
But I realized that I wasn't this remote set up long before everybody switched to that with just the different that everybody was coming to the offices and was jumping on the calls from the offices instead of doing it from home as everybody does now.
So I live in this remote set up for 10 years already, maybe a little bit more.
It is interesting to like in in this big company, I was first in a market CFO role, and then they switched me to a cluster CFO.
So I was responsible for several markets.
And it gave me such an interesting perspective on how the cluster people or fp &a teams look at the market and how the market looks at the cluster.
So like how the budgeting process is organized, how it is communicated.
When you are just at the very bottom of this process, you sometimes get irritated like, why didn't they set up something night before it had to be submitted?
But then you realized that there are like five people over the people that you are talking to.
And they don't necessarily have control over this process.
And they are doing their best to make this process as comfortable for you as it can.
But it's not always possible.
Right. Another thing is that I learned that it is very important to mind the cultural differences in a international team.
Right. And sometimes you will get a team that is very diligent, follows all the deadlines, you can always rely on them.
But sometimes you will get a team that needs five reminders before they do something that switch their computers 6pm sharp, and you will never get to them until nine in the morning.
And you really have to keep that in mind and to change your communication style.
And you do these reminders because like, what else is there to do?
Right. So that was a lot of learning for me in switching between the roles.
And in a smaller company, I had a little bit different challenges.
So the company was relatively small, but international, we had five different offices, right?
I couldn't afford having a separate finance or FPNA person in every of the offices.
So the way we handle that, every person on the team was wearing multiple hats.
So it was either an administrative person doing some office management and doing some FPNA on top.
Or we hired someone from finance, but then we would add something else to their role.
And it was not easy to balance these two parts.
People were not always happy with having to switch between something that they prefer doing maybe more or something else.
Or if we would, for example, acquire an office, and I would inherit a team there, the team would not necessarily understand how even finance works.
So I would inherit a team of non -finance people and I would have to put some financial tasks on their table, right?
So then we had to teach them how to handle the numbers, what balance shape is, all this stuff.
So all ways of trying to balance complexity and small size of the company.
It's funny. My brain just skipped and I had a war story of my own that I remembered.
And I normally don't tell these stories, but this was at a small company.
And when you were talking about picking up the team that finance may not be their first responsibility, it was myself and a controller.
And we had someone for AP and someone for AR.
And I really needed a senior accountant or a staff accountant.
This is a company that's $12 or $15 million a year in revenue.
So very, this is the amount of work we're doing.
I need to get a staff accountant in the house.
He said, well, you are in luck.
My sister is looking for a job.
And I said, oh, your sister's an accountant?
No, she's an executive assistant.
But it's all the same.
So I ended up getting someone with no accounting experience to be my senior accountant.
And she was great and we got there.
But that's tough when you inherit a team that this is not what they're doing.
And you're trying to manage that.
Yeah, I had more or less the same story.
One of my key people on the team managing a lot of accounting stuff is not an accountant.
She's great, but she needs really a lot of oversight in the accounting layer of what she's doing.
But she I think has a marketing background.
And she ended up doing accounting just because somebody asked her, do you want to take this additional responsibility?
And she said yes. And this is how she evolved into an accounting role, which is funny sometimes.
This is how it works.
Yep. So let's dive into some more war stories.
So we talked before the call.
And I know one of your stories and we can all relate to this.
I had a kickoff of an ERP implementation one time while I was on my summer vacation to the beach.
So that wasn't much of a vacation when I was on onboarding calls all day from the beach chair.
But one of yours was submitting reports during the family vacation and reformatting overnight.
Tell us about that story.
Yeah, so it's like, it's actually two different stories.
So one was, you know, when you are in this big corporate role, you try to understand when is a good time to take vacation because there are no people to back you up and you understand that you will take your laptop with you and still be responsible for the things that you're responsible for.
Right. So I was completely aware of that when I planned my vacation, but I have a family so I had to adjust to the schedules of everyone in the family.
And initially, this time that I walked for vacation was not supposed to be the time when we would be submitting the budgets, but the company was changing the software and something didn't work well.
So they pushed the deadline, so I found myself on a beach resort with my family and with a very, very poor internet connection.
And I kind of did everything I could before going there, but the submission of the reports still needed to be done during this week.
And because the internet connection was poor and the company systems didn't work well, I had to really spend like six to seven hours a day just loading and reloading the reports, which was like, you never want to find yourself in this situation.
Kind of, I did everything I could to avoid that and my family was very understanding, but that was not a nice experience.
However, when I reflected on that and I thought, okay, going forward, what can I do not to ever get myself into that situation?
It's actually not possible because of the ways big corporations structure the finance teams with little to no backup in executive positions.
I never had a number two that would replace me in this situation.
So the teams are very lean, do more with less, all this stuff.
So it is very challenging to manage the work -life balance in this big corporation, although they often declare that they care about it.
It's often, in practice, it's not how it works.
And the second story is a kind of funny story.
So we did this like 72 pages presentation for a regional business review.
So it was like previous year budget, current year budget, all the story between.
And I submitted everything on time and everything looked great.
But the night before this review was planned to take place, someone from the region contacted me and they said, you know what we have in U standard that you need to now use dots instead of commas as decimal separators.
And it sounds like an easy task, but in the presentation, you have so many data formats that you really have to look through every piece of information.
What made it even more challenging, the regional CFO at that time was a very, very tough guy who wouldn't let us have any errors on the budget presentation.
And if you had even a small rounding error, he would stick it to your face.
So I had no chance to get any excuses on not doing that.
So yeah, I spent a night redoing and rechecking everything.
I just thought it was a two waste of time.
And when I got into the cluster role, I never did that.
You need to focus your team on value adding activities, not in checking if the dots are in the right place.
I think anyone who's been around for a while and loves to try to solve things in Excel, I was immediately trying to figure out the solution there of like, but then when you get to the number format, it's not like you can search for the comma because it's being put by the format.
So then I was like, well, could you concatenate?
I bet some of our VPs, I'm sure they'll say, Oh, here's how I would automate that.
I'm stumped here. I don't know, other than it's because of the formatting that happens in the spreadsheet itself.
Yeah, because it's a mixture of and the presentation itself was connected to the internal company system.
And then you had all the numbers as a part of the text in the presentation.
And sometimes we even had like the pictures like screenshots of graphs included.
So on the screenshots, we also had to go back and change it in Excel and then redo the screenshots.
You know, like maybe there would be a good solution, but you would never find it inside like three hours.
So when you are under a tight deadline, you just go with the classic ways to solve things and hope that you get there.
You don't experiment.
Yeah, I'm somehow picturing on the screen with that old that liquid paper, the whiteout stuff, taking the bottom of the comma off and turning it into a dime.
But imagine it was like maybe 15 markets that had to do that.
So all finance team didn't get to sleep before a very important presentation day.
So, you know, you've been through those.
And I'm sure there's we all have like our, sort of our big wins, where you get through that and get out on the other side, but having gone through several of these cycles of budgeting, how would you say, and you know, maybe it's, I don't know if it's more based on time and technology changing, or if it's
the size of company where you've been, but how in the budgeting process change kind of over your career, especially with the growth of distributed teams and fast moving tech companies.
Have you seen differences from when you started to today?
Yeah, I don't think I'm in the place to really compare because I'm not in the corporation anymore.
And I wouldn't be surprised if I go back and I see that the processes are exactly as they were five years ago, honestly, because the corporation is so rigid, they, it is difficult to change when you have 150 countries right in your structure.
But what I am able to see is that a lot of companies are now trying to implement some higher level of agility in the budgeting.
So in the corporation, we would do two budgeting cycles in the year, right?
And it more or less worked.
Now the companies want to do quarterly, some companies are doing monthly.
In the tech company, I had monthly recalibrations on the budget going pretty deeply into the layers, right?
So it's not like the latest estimate that you would do on the top line on bottom line and just skip everything in between.
We would go into the hiring updates, we would go into like what we buy, what we purchase, software, whatever else, because the environment is so rapidly changing.
And the expectations of the leadership team is for the CFO to be able to keep up with this change.
And they don't want you to do that coming to the every department every month and asking them questions.
Everybody gets irritated by that.
So you got to come up with ideas how to automate that, how to make it work on its own.
And AI actually helps there a lot, but the demand for this higher density of budget updates, I think it is there maybe in corporations as well, because I hear it from fellow.
What do you hear? You know, it's funny, I started out not in finance, but in the military.
And I did have in my department, I had a budget I had to manage.
And I would think about the budget every year in the military did it very differently.
It was basically back then anyway, maybe they've gotten smarter since then, it's been a million years ago, but they you basically had to be sure that you used all of your budget by the end of the year, otherwise they would take it away from you.
So people just buying stuff they didn't need at the end of the year.
And this is still happening.
This is still happening.
And how stupid it is.
It is not bringing value to the company.
But I don't think it's changing.
I still hear like if you talk to salespeople, they are at the end of the year.
So everybody is trying to push sales because the companies have budgets to spend.
And why the companies have budgets to spend because if I save this budget, my regional head will never let me use this budget in the next year.
When maybe I needed more, they will just say, Okay, you have savings here next year, bring more savings, please.
Right. And the same goes with revenue.
If you over deliver revenue, then the next year you have to over deliver more.
This is why all the companies are playing games with revenue recognition, pushing it over backwards depending on what they want to get.
I think I don't have an answer how it should change.
But I hear a lot of frustration in finance team across, you know, industry because you want to bring some value in your role, not just to manipulate the numbers to what that's like a dance.
We're all dancing. I mean, I know exactly what you're talking about with when a deal closes on revenue and, you know, the whole revenue recognition game to try to hit your numbers, but not overshoot them too far and all that.
But, you know, seeing the behavior that that drove in people and trying to like, you want people to be frugal.
You don't want them to be buying extra office supplies just because they're trying to figure out how to how to blow their budget at the end of the years, they don't get taken away.
So and it was early in my career, my in my first finance leadership roles trying to get people, you know, it's like, look, here's your run rate, we can, I'll give you your GL you can look at what your department did, you know, what by region by city, whatever.
But we're going to do a bottom up budget.
And it would drive them crazy.
They would say, you mean I've got to start and fill out what I'm going to do all year.
And yes, you can look back at the GL.
But we're going to we're going to knock that away.
So, you know, you end up in that kind of hybrid where it's somewhere, you know, top down, you know, you've got your goals, and then you're going to meet in the middle.
But yeah, the biggest thing now is is the technology side of it, how it's getting a lot easier because the way that things are tracked.
And but it's also, it's the same thing.
Smaller companies don't have access to the better, you know, ERPs and software systems out there.
So yeah, but it's, I don't know, it's, it's been a couple years since I've been doing consulting.
It's been a couple years since I've had to go through budget season.
And it's probably like an accountant with tax season.
Like when you're in the middle of it, it's, it's not fun, but getting on the other side and having a budget and especially when you watch how your, how your budget does compared to actuals the next year, when you feel like you really nailed it.
It's like, that's one of those small victories in financial, man, we got that.
I agree. But you mentioned technology and technology does help in a lot of things, but I don't think it's going to help with this dancing in politics.
That is a big part of the budgeting process still in many organizations, right?
So it can, technology can make the analysis easier, like presentations, like writing maybe reports.
Yes, you can leverage it for sure.
I don't see a way it would change this like take it to leave it approach or the necessity to manipulate numbers.
And I had to do that.
There was no way I could get out of that because my market was expecting me to look up for their interests.
And it all eventually goes to how people are identified, right?
So bonuses depend on what targets you agree on, which makes the target discussion not a finance or data discussion, it makes it a power game and politics.
Yeah, yeah. I mean, I'm sure it was slightly different at the massive corporation versus the smaller one.
But were there lessons that you picked up along the way of how to maneuver the politics through budget season?
Anything, any tips you could share with our listeners?
Like ideally, I would not have to share any tips.
Ideally, I would like to find a way to get out of this whole situation.
But because I was in this bold situation when I was at the bottom of this whole process, and when I was somewhere in the middle between the region and the countries, I can say that actually, CFOs should double down on politics, unfortunately, right?
Because when I was in this cluster role, I realized that if I get a stretch from the region, I'm looking for the easiest market to put it on.
Like I would look for a country that would push back as little as they can.
And this punishes the countries that actually respect me, that trust that when I give them a stretch on the budget, I actually have an idea how they're going to get there.
But I didn't always have this idea, I would just be challenged by the region.
And I had to put it somewhere between my teams.
So when I got back to a bigger country role, I would always push back as hard as I could, which made it difficult for everyone in the process, right?
It's like a lot of unnecessary actions.
Unfortunately, things I'm telling the aspiring CFOs is it's not that important how you handle data.
It is important how you build relations and how you navigate these politics in your organization.
This is much more important than how you handle your numbers.
That's the reality.
I hope it's going to change someday, but I don't see it changing right now.
And what's interesting when you say that is, as we look at all the sort of AI and automation and what we're doing with data, and the parts that AI and automation can replace versus that very human element of our job, the politics, while maybe, I think some people probably enjoy it, but it can be a difficult
part. But if you are able to navigate the politics in a company at budget time, and befriend the right people, and don't back down to the right people and all, it's a big part of our job.
It is. Yes. But I can actually share a story of how I see AI changing that.
It's an interesting one.
And this is like current budgeting season.
So I'm helping a company in the budgeting process.
And they have this approach of top -down and bottom -up approaches meeting somewhere in the middle.
And there is always a lot of politics.
The top -down tries to stretch.
The bottom -up tries to be as conservative as possible.
But this year, we did this AI exercise.
So we analyzed a lot of data and we came up with AI generated forecasts.
And having AI didn't make this possible.
It was possible to do it before.
But it made this exercise much easier.
So you don't need a data analyst, and you don't need to hire a team of people to actually get it done.
So we introduced this AI generated forecast.
And what it did to the budgeting process, it was a kind of like neutral input, which you never have in the budgeting process.
You always have someone protecting their interests.
And now you have like a robot who doesn't have emotions, who doesn't have any interest, and whose bonus doesn't depend on what it says.
So the budgeting discussion really shifted from, I need this number and I can only deliver this number to something like, okay, why does AI think we can get here and not here?
And then you look at the assumptions and you look at the model.
And the forecast was not perfect.
But it brought a very different perspective.
And I could see how the discussion shifted to something that to me would be much more valuable than just playing this power game.
Yeah, that's interesting that you say that with AI, because that was sort of the approach that I would take when I was dealing with department heads.
Sales and marketing was always the toughest.
But since they were looking at top line, and the company, maybe the CEO is saying, look, we need to increase revenue by 15 % this year, with a stretch of 25%, or whatever.
And then you'd go and look at the prior years, and all things being equal, you could do a forecast and say, look, all things being equal, this is our forecast, this is the 95 % probability of where it'll go, just on the statistical just launch.
No matter what we do, we're just launching this out.
And if you could show them the band of like, here's your max, median and minimum, here's the band, look where your forecast, look where you're either way high and you're way low, neither of these are justified.
But it was trying to do that same thing, where it's just, let's just be cold and analytical here.
But the thing is, then they were like, well, how can you say, not being stats people would say, well, how can you say what the 95 % probability is, you're just making, you know, you're making those numbers up.
But if you have the AI do it, and you have the forecast go out, it's like, look, this is where we are.
And then so you can, it's almost like it gives you someone to blame for the rationality that you're imposing on them.
So if our gross margin is x percent, and it has been x percent for the past five years, why do you think we're going to cut that in half?
You know, what, what are we doing?
And then, but that that meet in the middle of the top down and bottom up is if you have a program or a project, we're going to do something to do away with one of our cost of goods sold or whatever, then you can come up with it.
But it is what I do like about using AI or even just statistical modeling.
So if you step back in before AI, I would think of this as what a private equity person to do if I'm evaluating a company, I'm not going to listen out of the gates to what the CEO tells me a strategy is for how they're going to grow top line and how they're going to cut these costs.
I'm going to say, let me what are the numbers say, if we just continue the trends, and then you look at it, and AI kind of helps you do that because it's very easy.
And you don't even have to be in your cellwiz to do it.
If you're using generative AI, you just tell it, forecast this out based on the trend.
Yes, but then you can also ask, like what what assumptions did you use?
So you can challenge AI.
And you can say, and if it says, I'm assuming that everything will be the same as in the last three years, you can say, okay, this is not what we're expecting, we want to do this and that.
And it's very easy with the natural language just prompting to get there.
So you don't need to translate this into some data analytics talk and then ask the person to embed it in some model that somebody developed for you.
So the accessibility of this analysis is now much, much higher than it used to be.
And not only companies like Amazon can now use AI to do whatever analysis they need.
So that brings up a question.
Right now, people like you and I, and I'm sure a lot of our listeners too, are going out and they're playing around with chat GPT and Claude and lama and whichever model they're using.
And they're seeing how you can now you can run Monte Carlo simulations, you can do all this stuff within the inline window on these tools, but they are outside of our workflow.
So getting the output from them, you can do it, it just takes a lot of massage, you know, having an output as a CSV, and then you take the CSV and put it back into your Excel models.
And if you go back and run the same, you could do the exact same prompts and queries two different times and get different answers.
And that's, you know, you shouldn't obviously for numbers that are set in stone and are he done, obviously, that's a hallucination problem.
But even if you're doing the forecast, just like if you went to two different financial analysts, you might get different interpretations of it.
So I'm wondering, as you're working with companies now, and as you see it right now, for this budget season, and I know you touched on it already a little bit, but how would you recommend people right now in 2024 budgeting season for 2025 and beyond?
How could they use generative AI to help them today?
Well, that's, first of all, that's an excellent question.
And what you described is one of the most common challenges that I see.
I also see that there are a lot of finance professionals who are trying to bring some AI, but their companies are very much against that.
So, you know, if there is a strict policy, if you don't do anything, don't break the policy.
Try to talk to your security department, trying to talk to your IT department, but don't go against the policy.
It's not worth it. But my clients are mostly like small, smaller companies up to, I don't know, 20 million.
They don't usually have advanced security settings or advanced security departments.
And we are, and they are using also very basic software for budgeting, it would be Google shades or simple Excel, right.
So in this setup, it is very easy to add some charge to PTO or plot into that.
And it already enhances the process.
And you don't need to invest a lot of effort and money into building some system.
So in this company is what I found out is that learning by doing and small steps, this is what works.
So when they start with a small project, and I usually like to point something that is really painful for everyone.
If the company, for example, has a lot of unstructured data or data in different formats, right, you can use AI to reformat stuff for you.
And it can do it pretty reliably.
You don't get two different answers when you ask the same question.
So maybe it starts with something that guarantees good results, and then share it inside the organization and expand the AI usage.
In bigger companies, CFOs alone can't get there.
They the only thing they can do they can scream on the top level.
Guys, we are behind.
Everybody is already doing that we are not, which is not true.
But you can try doing that.
Because there is no way that CFOs alone will be able to implement those policies set up everything that needs to be set up in order for AI to be embedded in the company processes.
At least this is what I hear.
And I hear a lot of frustration from the CFOs.
There were reports that are saying that people are using AI on the workplace, regardless of the company supporting that or not.
But I you can't bring something up during the budget decision and say you use AI for that.
If your company doesn't support AI.
FP &A today is brought to you by data rails, the world's number one FP &A solution.
Data rails is the artificial intelligence powered financial planning and analysis platform built for Excel users.
That's right, you can stay in Excel.
But instead of facing hell for every budget month and close or forecast, you can enjoy a paradise of data consolidation, advanced visualization, reporting and AI capabilities.
Plus game changing insights giving you instant answers and your story created in seconds.
Find out why more than 1000 finance teams use data rails to uncover their company's real story.
Don't replace Excel, embrace Excel.
Learn more at datarrails .com.
It seems like every couple months you see a new survey where the number of people who are using AI in the workplace is going up and up.
But at the same time, the number of companies who've gotten a handle on it and have figured out how to direct them has stayed pretty flat.
So you've got these employees out there who are figuring it out on their own.
Because they're doing it in this rogue way, and it's not systematic, and it's not being controlled by the company.
You've got these cyborgs who are enhanced humans doing this, but they're also potentially putting your data at risk.
They could be just loading it into regular cloud provided stuff outside of the protected environment.
Everybody's getting at sea level and managers.
Everybody's getting pressure from, whether it's investors, bosses, boards.
Everybody's saying, we need to lean into AI, but it's not really there yet.
It's not ready to us.
I work with clients very similar size to yours.
What I tell most of them is, I'm going to show you some cool things we can do.
I'm going to show you how we can use it.
I'm going to show you how you can use it safely.
We're going to work on your AI usage policy and make sure that everybody in the company understands.
We'll set up a system where you can train, rescale, upscale everyone to understand how this works.
But truthfully, depending on the industry, and I would say at least over 50 million in revenue.
Even then, depending on the industry, you may not have the team.
But if you have an internal security team and data science team or BI team, then you may end up building something internally you're using.
Otherwise, smaller than that, most companies are going to be at the mercy of they're going to wait for DataRail's AI Assistant, NetSuite's AI Assistant, whatever.
It's going to be in the SaaS tools that they're already using.
That's how they're going to use generative AI.
It's not going to be just going out and dumping it into publicly accessible chat GPT or whatever.
Well, you're right.
But when the smaller companies get access to this AI in a bigger software, they often don't know how to use that.
I'm trying to argue that companies should be using AI in safe cases, in safe environment, even if they don't use it at scale right now.
Because working with AI is a very, very different thing compared to rule -based automation that everybody is used to.
And it is very important for CFOs and executives to understand this difference, but also for everybody else who works on normal office tasks, because it is just so much different how you interact with that.
You ask the same question, you get two different answers.
People don't get it anywhere.
Finance teams don't get it anywhere where you ask the same question, and you get a different answer to tomorrow compared to yesterday with exactly the same input.
It is different. What I'm trying to bring into the company is this is the safe setup.
This is the safe data or safe documents that you can work with.
I think that paid team subscription for chat GPT is safe enough for the company to use it on some internal data, not social security numbers, but contracts.
Probably you could use contracts.
You could use it on company financials, even if it's not a public company.
It doesn't work like your balance sheet will be in the internet tomorrow.
So, again, I think the best way is to learn by doing and by starting small.
Even implementing it in some process, which is not critical, but painful, it gives the company the exposure to what it is.
Then the finance team is able to understand what other processes are a good target for AI implementation, and then you can add technology, you can bring in the team, and you can implement the software specialized in this specific use case.
It can be budgeting, it can be account payables processing, for example, or account receivables or expense reporting, depending on what's the most painful process in the org.
But if you don't expose people to AI before, I don't think you can successfully implement that.
That's what I've seen.
I've seen a lot of failures in implementation just because it's not easy.
The difference is too big between the classic tools and AI tools.
And you use it daily.
I use it daily. And we tend to forget, but I still forget how it felt when I started six, seven months ago, and how many times I was frustrated and how many times I wanted to drop it because it doesn't work.
Okay, close it. Let's call back to Excel.
It is a new technology.
It is developing very rapidly.
Even what changed since I started since six, seven months ago, the models evolved significantly.
Now you don't only need to know AI, you also need to know how different models work, how you speak to them, how you interact with them, which use case is the best for which model.
So the complexity has increased significantly.
And the starting point is now much further, I would say, than it used to be.
So if you started a year ago, you would be in a pretty comfortable situation now.
If you are just starting today, there's a lot to digest.
Yeah. And I wonder, what do you think it means for the future of...
And you could swap out what I'm about to ask and plug in any profession, any sort of white collar thing.
But I'm thinking, because our focus is FP &A, if I'm an FP &A person and I am very good, I can do anything in Excel.
I'm Power BI even. I'm just super analyst.
I can use all this traditional software.
And I have defined myself in my career by my ability to build a really cool model, to use Power Query, to use SQL when I need to.
I can even use Python and Excel, all this cool stuff that is cool to FP &A, maybe nerdy to my kids probably.
But I'm very good at these specific tasks.
Now generative AI is coming along and it's saying that all the code that I used to be able to write can now be replaced with natural language and just interact.
So instead of having to learn how to write a SQL query, I can just use generative AI and ask the same question.
It returns my results.
And to your point, if you just tried to even just keep up with the different generative AI models that are out there, how does that make you better at your job if you know the difference between Gemini, Claude, Chat GPT?
That's almost a distraction because you're chasing this AI arms race.
What do you do, whether you're just starting your career or you're kind of mid -level or even senior level and now there's this rapidly advancing wave of technology.
And if you haven't started yet, this is kind of the million dollar question.
What should people be doing right now to use it today and to kind of prepare for the future?
What should they be looking at?
What are your thoughts on that?
Yeah, that's a good question.
I'm actually issuing a newsletter tomorrow dedicated to this exact question because I hear that a lot.
And you're right, it's not easy.
My framework that I kind of developed is start using something.
If you don't know what, I would say, okay, Chat GPT is the basis plus whatever you're using.
Maybe if it's a Microsoft suite company, then it would be co -pilot rather than Chat GPT.
But it's more or less the same.
And experimenting and learning by doing works much better than just doing a training.
So I did trainings before and when I get back to the company a month later and they ask, okay, did anybody start doing that?
And during the training, everybody's like, wow, wow, really?
Then you come back to them a month later and you ask, did you change anything?
No. Because nothing changes when you don't manage this change, first of all.
So you have to be really managing this change from the top.
And also, you got to let people experiment and push them towards this safe experimentation.
But right now, I would do five simple steps.
I would do a short training.
Google has pretty good training on basics of AI machine learning just to get the vocabulary.
I would identify a process that is manual, painful, and doesn't touch sensitive or confidential information.
Something like expense reporting, for example.
And I would try to implement a simple automation for this process based on the existing tools that the company is using and some AI.
And it can be very, very simple automation.
And then the people who will see the result, they would be very excited because normally, this is a game changer in this manual tasks.
And then you do a sequence of step by step implementing more processes, more AI in more processes.
And then your team gets some free time to experiment and to learn more.
So this is how I got to where I am with a lot of experimentation, a lot of hands -on experience, but also you need to dedicate time for that.
And I'm subscribed to a lot of newsletters.
I listen to podcasts.
I watch YouTube because there is no way you can navigate this space without dedicating time to learn.
Now, is it a waste of time?
Partially, yes. If you are in a full -time role, you don't have five hours a week to spend on that.
But the flip side is that if you're not doing it now, in six months, it's going to be much more difficult to start than it is now.
And now it is much more difficult than it was six months ago.
So the longer you wait, the more difficult you make it for yourself.
And even if you are an advanced Python person, you will find yourself in this situation.
But there are a lot of CFOs who never learned Python, but who can do the data analytics tasks now better than you do without even going to you.
So what's going to happen to your value on the market?
Yeah, it's an interesting time trying to figure that out because I, like you, leaned pretty heavily into this because I don't know exactly when the robots are going to come for my job.
But I felt like the more I knew about how the robots worked, the safer my job would be.
We'll see how long that lasts.
But yeah, I think for FP &A, for anyone right now, I say statistics first.
That's kind of the gateway into machine learning.
And then from machine learning, when you understand how that works, you understand how the generative AI models are working.
And it makes it easier to navigate this world.
But with natural language, you could learn everything you could about coding and all that.
And you could still, someone who never did before, could just type in a query and get an answer.
So it's an interesting time.
It is. But don't you think that knowing AI will be a requisite for a part of any job description?
And how soon do you think it's gonna be?
Every couple months, it seems like Indeed puts out a study.
I'd like to see where they are in job descriptions with AI.
I can tell you where they are.
They're not. Right now, they're not.
Because I'm doing this, and I'm in a lot of CFO groups, and I've talked to them a lot.
And AI is never there.
It's not there at all right now.
Even the companies who are using AI as a part of their business model, they haven't put AI as a requirement for finance people yet.
So it's interesting.
And there is a lot of requirements like advanced Excel user.
Yeah, Copilot is about to make that moot.
Yeah, right. It's well.
And there is already a way to add AI to Excel.
So you don't really need to be that advanced to do whatever you need to do.
Well, I'm sure we could talk about this all day.
But we are bumping up against time.
And I do want to get to a couple of questions we always throw out at the end.
Because we went straight into work and AI stuff.
So at the end, one is we always like to ask our guests, what's something that maybe not many people know about you?
Something we couldn't find just by googling you?
Any interesting personal tidbits you'd like to share?
Well, I'm actually a very open person.
So I don't hide stuff.
But what I don't necessarily share with everyone is that I took singing lessons.
So I like to organize, let's say, karaoke parties as a part of even my work bonding.
So whenever I would go and have to meet a new team, I would do karaoke with them.
And it's not that I do, you know, professionally, but I enjoy it.
I think it's a good way to bond and a good way to find stuff about your team that they don't necessarily share with the world.
Yeah, everything else.
I'm an open folk. That's funny.
I think if I tried to do karaoke with my team, they would all just immediately quit.
They would be like, if this is any indication of how anything else goes in your life, your singing voice is telling me that you should not be allowed in public.
I'm sometimes surprised with how great singers my team is.
And this is not what you see necessarily in the work environment because people tend to be closed and they don't want to share it with you.
So for me, it was like a door opener in many cases.
And yeah, but I can imagine the situation when people are not happy with this.
I'm trying to be cautious.
So far, it works. That's great.
All right. Now, now our our classic question that we I'm hoping that the producers are tracking this because there needs to be a tally.
And it's going to be interesting to see how it changes over time as CoPilot gets more and more integrated and as generative AIs out there.
But for now, what is your favorite Excel question or Excel function and why?
I was hoping you'd skip this question because I don't have a favorite Excel question and Excel function and CFOs shouldn't be necessarily dealing with Excel a lot.
Fair point. But I used to have a team of people who would do all the preparation for me and I would just press the button.
But I like people tables.
I think it's a very good tool when you need to structure the data.
And I do a lot of visualizations.
I'm not a visual person myself.
I like to look at the numbers.
But a lot of sales people, sales teams I work with, they don't see the way I do.
So I would plug in and make a simple visualization.
And this is how they see what's going on.
So they can see the trends.
They can see the like if it's growing, if it's declining.
So maybe this is what I use most frequently.
Yeah. And I totally get you on the Excel thing.
It's funny, the more I get removed from the day to day operations, the more I think of my early bosses who had their financial calculators and they would love to sit in meetings and calculate net present value on their calculators that nobody.
I'm talking old school stuff the way that they used to.
But yeah, I hear you on that.
And that's great. So all right.
Well, we're right at time here.
So before you go, how can our listeners connect with you, learn more, find out about Blend2Balance and all that?
Well, I have a website, blend2balance .com.
You're always welcome to visit it.
And you can set a call with me.
Again, I keep my calendar open.
I'm trying to very carefully balance the C full part and the AI part.
And I'm open to new clients in actually both directions.
Although, I mean, maybe I'll just switch to full time AI consultations at some point.
But I do think there is a lot of synergy and it is important to practice what you teach.
And this is why I think this is a good setup for me at least now.
And as you can hear from the name of the company, I'm a big fan of balance in all meanings, including balancing two not necessarily same activities.
So it works. I'm very active on LinkedIn.
It's very easy to find me there.
So go ahead and connect with me.
I'm issuing a weekly newsletter where I cover a lot of stuff that we talked about.
And I don't know, I'm just working on my tomorrow's issue.
And I'm trying to fit everything I want to tell.
And it's not easy. So maybe I'll switch to daily at some point, not now.
It's a lot of effort.
But there is a lot of information going.
There is a lot going on.
I'm always happy to connect to people who are also exploring AI and finance, finance and AI and all the combinations of the two.
And I love what you're doing, Glenn.
I'm very happy to hear your podcasts and to read your articles.
I think we're all going in the same direction.
And the more people are talking about it, the easier it is for us CFOs to get the understanding on leadership teams about why it is important, why companies should start if they haven't started yet and get more open maybe about AI.
Even some operations that deal with secure data, there are ways to set it up securely.
It's not a blocker.
It's just a thing to consider.
Absolutely. Well, keep up the great work.
Let's stay in touch.
We're kind of running down the same road at about the same pace here.
So we'll have plenty to talk about going forward.
Yeah. Thank you. Thank you for having me.
