AI! Doesn’t it feel like those two letters are on the tip of everyone’s tongues these days? AI for FP&A specifically holds a lot of promise . . . but it’s important to differentiate between what it can do today, and what it will be able to do tomorrow.
No longer fodder for good (and bad) sci-fi alone, we’re now living in a world where AI is set to impact (probably) everything we do in the coming decade. From self-driving cars to coding to AI for hiring - everything is set to change. And ‘everything’ includes FP&A.
This article will help you understand:
- The future of AI powered FP&A processes and the benefits of AI for FP&A
- How FP&A professionals can leverage AI
- Why you shouldn’t feel the need to rush into AI
AI for FP&A: Automation, Online Analytical Processing, Limitless Potential
In 2022, nine percent of organizations were already making use of AI and/or machine learning, while 57% of organizations were planning on adopting the technology, according to a survey by FP&A Trends Survey 2022.
By the end of the decade, the AI market is expected to surpass $2.0 trillion, growing at a CAGR rate of 21.6%over the next seven years. Among those certain to be impacted: finance teams.
Gartner predicts that by 2025, 70% of organizations will use AI and ML to support data integrity verification and modeling tasks, but most importantly, help not only produce accurate financial data, but also explain its derivation.
To wit: that same Gartner report predicts that by 2027, 90% of descriptive “what happened” and diagnostic “how or why it happened” analytics in finance will be fully automated.
AI, in other words, is slated to augment FP&A tasks - analyzing and interpreting data, understanding not only what happened last quarter, but why it happened. Pattern recognition, running endless scenarios, etc, can be done by machines so that finance pros can focus all efforts on the high value work.
End result: higher level data interpretation that informs every important decision a business makes. That way your organization can better avoid making the same mistakes twice while reproducing your successes. AI-powered predictive analytics will similarly help identify future threats and opportunities while prescribing solutions to them.
Where AI Thrives
With an AI assistant, you’ll get more than just data automation - you’ll have all the heavy lifting of running models, finding correlations, etc. and suggesting conclusions done for you so you can focus on the high level analysis, interpretation and - critically - translating that into smarter business decisions, ready on a dime when you need it.
And, of course, since cloud FP&A software has already automated a lot of the repetitive tasks, you can expect that a more robust AI system will essentially be able to offload even more time-consuming tasks entirely from your FP&A team.
Not just saving you countless hours. Freeing up time. Uncovering new insights. Giving you the insights and the way to convey them to make sure every important decision your business makes is smart and fast (rough language, but you get the point…what you can do with the time. What investments to make, products to launch, markets to enter…screw bookkeeping:)
AI for FP&A, in other words, is going to reshape the industry.
‘Going’ being the operative word, however.
. . . But AI Isn't There Yet - You Have Time
The techies do love their buzzwords and have a chronic ‘next-big-thing mentality’ that can often lead to overhype, premature adoption, and wasted time and money.
Which isn’t to say that AI is a fad - it’s not!
But what that means is that companies that are hot to adopt AI for AI’s sake are likely to spend a whole lot of cash on systems that just aren’t able to meet their promise . . . yet.
The real competitive advantage of AI for FP&A comes in helping inform strategic decisions. This is accomplished by performing the modeling and analysis mentioned above, being able to tell the story behind the numbers.
While AI has the potential to perform robust predictive and prescriptive analytics(and in all likelihood will achieve it within a few years), right now it’s not quite there.
The Best Way to Use AI in 2023
AI has a lot of promise. But there is a potential drawback to doing too much too fast.
In an FP&A context, if AI returns a data point or a whole host of data points that are inaccurate, that could severely hinder the effectiveness of your FP&A and even hurt your company.
Accurate forecasts can only be made by accessing accurate historical data (garbage in, garbage out). If you can’t 100% trust your AI to return accurate data, then it could disrupt your entire FP&A process.
The good news is that AI isn’t doing this alone - it’s going to be supervised by AI pros (FP&A professionals like yourself who have taken the time to understand how best to deploy AI FP&A tools) for the foreseeable future, so the two in tandem will ensure that you get faster, more efficient FP&A . . . without any dire errors.
The norm is that the consumers play around with exciting new technologies while businesses wait and see where the real value can be accessed.
In this case, you have millions of people using ChatGPT for all sorts of tasks. But businesses would be best served by not getting caught up in the hype and instead waiting for tried and tested AI that can really drive value.
If you're eager to try, go for it. But you've got time to figure it out. First get automated. Integrated. Accurate, Trusted. First things first.
More to the point, AI at this current moment cannot provide in-depth analysis. Empirical analysis of bar graphs, charts, and other numbers? No problem. But qualitative analysis of the data? Forget it.
What AI today can do is tell you what happened (descriptive) - and tell it to you fast - and suggest reasons why it happened (diagnostic), but it’s still not quite there yet. And in terms of leveraging that data point to predict what’s coming (predictive) and recommending actions (proactive), that analysis still isn’t quite ready. And ultimately the real value will come from being able to suggest efficient ways to allocate resources, release products, create strategic workforce planning strategies, etc.
FP&A AI's Real Value and What to Do Now
The competitive advantage of AI comes in helping you make more informed decisions by analyzing past performances and data and predicting future states, risks, opportunities, etc.
AI will be able to provide:
- Comprehensive pattern recognition, allowing you to gain more insights into your data
- Long-term predictions and projections based on all manner of factors (market forces, geopolitics, central bank sentiments and past actions, etc.)
- The relationship between disparate numbers - uncovering patterns and connections that humans won’t find!
Right now, AI is good at completing simple call and response actions - input a task within its capabilities or one with a very black and white answer, and it’s great. This is fantastic for chatbots, for instance, that can answer questions, scour data near instantaneously, and otherwise do exactly what you need it to do when given the correct prompt. From there, you just need human validation of the data and the ability to contextualize it (which is where your FP&A team excels).
Some extant AI on the market today include:
- Vic.ai for invoice processing
- Indy for client billing
- Docyt for automating workflows
- MS Copilot being integrated into Teams and office
These are handy tools, but they’re still not analytical tools - yet. AI and machine learning (ML) are developing into powerful tools to hold in your toolbox, but there’s no need to rush into them, and the real value is still on the way.
MS Copilot, for instance, promises to help in four key areas with FP&A when it releases:
- Building ad hoc reports
- Creating and formatting data visualizations
- Determining correlations in data points (when prompted)
- Generating what-if scenarios and financial models
But we’ll still have to wait and see how this all looks when Copilot releases. Furthermore, you will still need to verify that the recommendations and analyses carried out by Copilot are sound.
Which brings us to the endgame: coupling talented, expert finance teams with AI so it can supplement and enhance their abilities, not replace them. This will save them hundreds of hours in rote tasks and ensure they generate more accurate and effective reporting.
FP&A experts training themselves on these technologies will help them adapt and give them an edge on rival companies (and finance professionals) by demonstrating that they are able to use these tools to produce value. What’s more, while AI is still in its nascent stages, by developing the skills required to use these tools now, your finance team will be more readily able to adapt to the more robust and powerful AI coming down the pike.
Modernizing Your FP&A in Advance of AI
AI promises to revolutionize the way we work - everyone, not just CFOs and FP&A professionals.
But before we encounter the singularity, there’s plenty you can do today that will help maximize the value of your FP&A activities.
With cloud FP&A software, you can benefit from machine learning and other AI-adjacent technology in addition to a variety of benefits, including:
- ERP integration
- SOC-2 security
- What-if scenario planning
These are powerful upgrades to your FP&A processes that can turn you and your team from data jockeys spending hours on data inputs and verification to value producers able to quickly assemble large volumes of financial data and perform in-depth analysis. This analysis in turn can produce prescriptive/predictive conclusions that AI simply isn’t able to match yet.
And the best part? Your team can figure out how to use these tools in 30 minutes, operating on a platform that looks and feels like Excel - but is better in every way.
What We're Up To
Limelight FP&A Cloud Software has long incorporated automation, machine learning, and other AI-adjacent technologies into our software.
Get in touch to learn more about how you can stay on the forefront of AI FP&A with our team.