The integration of digital assistants into financial planning and analysis (FP&A) is enhancing efficiency, accuracy and strategic decision-making within organizations. There is a wealth of opportunity with gen AI digital assistants—from summarization to engaging with the finance data fabric.
In this blog, I am using TruQua’s FP&A gen AI tools to create “A/Rtemis”, a digital assistant geared toward a high-value area, accounts receivable.
Let’s start by looking at how a digital assistant like A/Rtemis can help us report, analyze and predict. With that context, we can then look at the larger topics on embedding gen AI into your FP&A processes.
In this example, I am using SAP Analytics Cloud as the finance reporting and analytics layer of choice, but these advantages are available on any finance platform. With the rapid evolution of large language models (LLMS) and composable architectures, I believe that loosely coupling your LLMs is essential to rapid time-to-value and long-term project success.
Interacting with gen AI assistant creative exploration
By leveraging large language models and code generation, gen AI digital assistants work well in both general exploration as well as data analysis. This capability encourages creativity and new ways of describing data.
Rather than be forced into describing the graph I think I want, I can look more broadly at my options. Here we see three diverse approaches to visualizing our data (with many more cropped out of our image): What’s a good graph to help understand our customer’s unpaid invoices?
Figure 1: Data visualization options
We can then have our digital assistant create one of these graphs (or a visualization that these suggestions help inspire). In this case, let’s go with the heat map. Though it was a later option, the heat map allows us to see trends quickly, both globally and by condition: Can you show the unpaid invoice age as a heat map?
Figure 2: Heat map visualization example created by digital assistant
We can even solicit suggestions for future dataset enrichment to support our analysis: What are some supplemental datasets that can help analyze invoice aging and likelihood of payment?
Figure 3: Dataset options
Whether independently or in collaboration with developers and data scientists, digital assistants for FP&A broaden our horizons for data interactions and analytics. Challenging the status quo for available reports and KPIs is essential for a modern FP&A team.
Simple modeling
Let’s examine a simple modeling interaction with our digital assistant. We start off with a critical A/R question—which customers are most likely to pay? By querying the underlying data structure, our digital assistant can help us design a prediction model, including suggesting supplementary datasets that can be used. Which customers are most likely to pay?
Figure 4: Example: Digital assistant modeling interaction
There are multiple directions that I can take in the modeling task and can even define my own calculated features. This approach provides an interesting alternative to Auto-ML tools or stock regressions.
Let’s add in an assumption about our data: Assume that all invoices are unpaid.
Figure 5: Data assumption examples
Let’s move forward without giving any more context. There are standard segmentations of invoices, but also insights across analysis: Proceed please.
Figure 6: Example of analysis results with FP&A digital assistant
In these example interactions, we have seen not only how digital assistants for FP&A can perform reporting, analytic and prediction tasks. We have also experienced how they can enhance our creativity by suggesting numerous options from a data and visualization perspective.
Enabling exploration
FP&A digital assistants overcome the constraints of analytics, reporting, graphing and predictive libraries within your FP&A platform. This flexibility gives a broader collection of functionalities for exploratory analysis and encourages creative thinking and deeper understanding of financial performance. By leveraging retrieval augmented generation (RAGs)—often called skills or functions—our digital assistants can integrate point-specific planning solutions like supply chain or headcount planning, maximizing their effectiveness.
FP&A considerations for gen AI—The gen AI analytic lifecycle
I believe that generative AI digital assistants can play a critical role in traditional FP&A development. Put another way, it’s critical to understand “Why wouldn’t you use gen AI as your only way to interact with financial data”?
For finance, generative AI is best to support rapid iteration of ideas, divergent ideation and data exploration. Let’s explore the advantages of incorporating analytic and predictive models assisted by generative AI. By pulling them back into the core FP&A system, businesses can enhance decision-making and operational efficiency.
Compute cost
The most advanced large language models have significant financial and environmental costs. Intelligently selecting the right sized model for an interaction can greatly reduce compute costs and environmental impact, as can traditional development techniques such as caching.
Providing the logs and outputs of a digital assistant interaction provides huge acceleration for developers on your FP&A platform, quickly bringing to bear new analytics to the entire organization.
Exceptions
Exceptions define finance analytics and LLMs—along with predictive tools from vendors—cannot reliably interpret those exceptions. Even the simple example here suggests several exceptions for how invoices are processed:
- Services versus products
- Intracompany trade
- Recurring versus discrete purchases
- Payment terms
Handling exception-based logic is a core tenet of finance analytics. Experienced developers and business users remain the most expedient way to ensure that those exceptions are handled accurately in reporting and analysis.
Speed
In many ways, LLMs are shockingly fast, but these examples are still generating code for execution and follow up text generation. Cutting out those steps and building queries optimized for your finance systems is considerably quicker for nonexploratory activities. We are also cutting out the data transfer and interim reads necessary to inform the LLM (or generated code) about the underlying data.
Taking advantage of native platform functionality
FP&A platforms are highly dynamic in terms of data interaction, including filtering, slicing and dicing, drill-down capabilities, integration with actuals, and variance analysis. By using digital assistant collaboration as a core part of their development pipeline, organizations can get the best of both worlds, combining rapid exploration with stability, speed and security.
When used as part of a cohesive strategy, generative AI is transforming human-system interaction in a positive way. Using FP&A digital assistants for creative exploration, rapid iteration and ad hoc analysis is a powerful tool for leaders to enable their business. That power is enhanced when used in concert with the strengths of traditional FP&A platforms.
This evolution is rapidly emerging; just as planning reflects an organization’s strategy, so does their usage of generative AI. If you’re interested in demos or would like to have a whiteboarding session on how to best use FP&A digital assistants, don’t hesitate to reach out through my LinkedIn profile. You can also ask questions in the following comments to continue the discussion.
