Skip to main content
Home
The Online Resource for Modern FP&A Professionals
Please register to receive the latest FP&A news, updates and tips Login

Main menu

  • Home
  • FP&A Insights
    • FP&A Trends Digest
    • FP&A Trends Research
    • FP&A Trends Insight Paper
    • FP&A Trends Survey
    • Short Videos
    • Our Contributors
  • FP&A Events
    • International FP&A Board
    • FP&A Trends Webinars
    • Digital FP&A Circles
  • AI/ML Committee
    • Introduction
    • Members
    • Resources
    • Meetings
  • FP&A Tools
    • FP&A Trends Maturity Model
  • About Us
    • Company Policy
    • Privacy Policy
    • Editorial Guidelines
    • Our Ambassadors
    • Our Sponsors & Partners
    • Contact Us
image
A Human-Machine Learning Loop Holds the Key to Improving Budgeting and Forecasting Agility
September 5, 2023

By Michael Lashinsky, Director Research at Gartner, and Randeep Rathindran, Distinguished VP Research at Gartner

FP&A Tags
AI/ML FP&A
Planning and Budgeting
Agile FP&A
Modelling and Forecasting

Human-Machine-Learning-Loop-Main

Recent events – such as war, new sanctions, supply chain shocks, a global pandemic, and high inflation – have demanded a level of agility that typical budgeting and forecasting processes have been unable to meet.

There’s no guarantee that current levels of volatility will subside anytime soon, and a majority of organisations are suffering from low planning agility that significantly hinders their ability to take advantage of opportunities or mitigate downside risks.

In fact, Gartner research finds that low-agility organisations capture, on average, 45% less financial upside of each opportunity and mitigate 36% less financial downside of each risk than high-agility organisations.

Moreover, commonly cited ways to improve financial planning processes, such as more frequent forecasting and greater use of operational data, do not drive financial planning agility on their own because they fail to improve the underlying planning methodology.

Digitally enabled planning methodologies drive the biggest impact on agility where, in contrast, relying on spreadsheets is a major drag on organisational agility (see Figure 1).

 

Gartner-Research-Figure-1

Figure 1: Impact of Financial Planning Transformation Actions on Planning Agility

Human-Machine Learning Loop

Planning digitalisation is critical to improve planning agility. Digitalisation matters in terms of not only the sophistication of planning tools but also how FP&A staff interact with and use them.

Specifically, a human-machine learning loop — collaborative and continuous learning between humans and Artificial Intelligence (AI) about assumptions, performance drivers and new forms of data — drives planning agility by increasing the organisation’s confidence and ability to respond to changes in the business environment.

One of the reasons for low planning agility is that too much planning time and effort are wasted in transferring data across systems, fixing errors, debating data quality, and determining a starting point for planning. Moving forecasting and budgeting out of error-prone, delay-causing spreadsheets into synchronised planning systems that automate data transfers and assumptions updates significantly improves planning agility. Many organisations realise this and are actively upgrading planning technology.

However, even if organisations improve the speed and quality of data transfers via synchronised planning systems, they will not see full agility benefits without three things:

  • a comprehensive understanding of the environment; 
  • how the opportunities, risks and drivers of business performance may be changing;
  • and the range of associated actions it is possible to take.

This is where a human-machine learning loop is critical. It creates collaborative and continuous learning between humans and AI about assumptions, performance drivers and new forms of data to establish early warning indicators, decision triggers and situational recommendations. This essentially turns FP&A improvements into an organisational capability that drives agile response.

A significant driver of a human-machine learning loop working well is an understanding of the respective strengths of humans and machines in financial planning (see Figure 2).

 

Gartner-Research-Figure-2

Figure 2: The Human-Machine Learning Loop in Financial Planning

Don’t Just Pick Low Hanging Fruit

Finance organisations often tend to rush for quick ROI in projects in order to show value. While nothing is wrong with that in principle, the biggest payoffs of a major digital transformation are unlikely to come in the form of quick wins alone. Think of implementing a new ERP system, for example, that can take years to come to fruition. A similar mindset will be required to make the biggest gains in forecasting and budgeting agility with Artificial Intelligence/Machine Learning (ML). Even putting the right team together could take a year or more.

Going back to the relative strengths and weaknesses of humans and machines, start with hard-to-diagnose problems that really have the potential to leverage the unique strengths of ML. In the early stages of ML deployment in financial planning, organisations typically make a few common mistakes that can stall momentum. These include overfitting models to historical performance, running ML models on purely financial data and undertraining ML models to secure quick wins when using them in areas that human strengths adequately serve.

To avoid these mistakes, firstly, use an iterative approach to deployment — that ensures the transformation uses a large amount of data and that humans become familiar with machine strengths and weaknesses for their organisation to have proper training and governance. It also allows for easier ML expansion to other financial planning processes, such as on-demand Scenario Analysis, as well as driver and planning assumption updates.

Secondly, use ML in areas it is best suited to! Most FP&A leaders believe the best places to start ML are in areas where they are more confident of success and the consequences of failure appear low: low-materiality, well-understood areas, frequently within the finance function. However, low business materiality reduces buy-in from the rest of the organisation, and well-understood areas make it hard for ML to generate novel insights. This is because traditional methods already deliver sufficient insights and diagnosis.


Create a Community of ML Users to Maintain Momentum

A key challenge in implementing the human-machine learning loop is getting stakeholders and users comfortable with a new division of responsibility and, in addition, how to use best and trust new tools’ outputs. Progressive FP&A organisations focus on user-level engagement and training to improve the quality of ML-based planning models and strengthen the human-machine learning loop.

One way to promote organic knowledge sharing about building and using ML models in planning is to train a community of analysts across FP&A and the business and catalogue their application of ML into a use-case library. This perpetuates best practices sharing and encourages continued iteration and development of models around the organisation. 

FP&A leaders should reach out to other functional leaders using ML and coordinate to share best practices and build an organisation-wide community that can apply ML and share learnings.
 

The full text is available for registered users. Please register to view the rest of the article.
  • Log In
  • or
  • Register

Related articles

FP&A Trends White Paper 2020
FP&A Trends White Paper 2020: The Power of AI and Human Intelligence
November 12, 2021

To bring some perspective to this position of AI within the context of FP&A activities, the...

Read more
Connagh-Hopkins-AI-Article-Main
Could AI Take Over the Role of FP&A Professionals?
June 13, 2023

Is AI a friend or foe for FP&A Professionals? In this article, the author speculates if...

Read more
Michael-Nudelmann-AI-ML-Main
Is AI/ML the Magic Bullet for FP&A?
April 25, 2023

In this article, the author explains why AI/ML is not magic, even though some might think...

Read more
Gizelda-AI-Capabilities-Main
How to Increase Your Chances for Success When Implementing AI Capabilities in FP&A?
March 7, 2023

The implementation of AI capabilities in FP&A can be a complex and challenging process. From the...

Read more
machine-learning
Machine Learning in Financial Forecasting – A Possible Reality or a Relentless Trial and Error?
November 22, 2022

Today, many companies are intrigued about what ML can do for their company and are waiting...

Read more
fp&a
Combining Human and Artificial Intelligence in FP&A: Roche Case Study
June 25, 2021

Today, the tools that are being offered are smart enough that with a few inputs and financial...

Read more
+

Subscribe to
FP&A Trends Digest

We will regularly update you on the latest trends and developments in FP&A. Take the opportunity to have articles written by finance thought leaders delivered directly to your inbox; watch compelling webinars; connect with like-minded professionals; and become a part of our global community.

Create new account

image

Event Calendar

Pagination

  • Previous
  • May 2025
  • Next
Su Mo Tu We Th Fr Sa
27
28
29
30
1
2
3
 
 
 
 
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Transforming FP&A Together: Human & AI Synergy
 
18
19
20
21
22
23
24
Moving from FP&A to Extended Planning and Analysis (xP&A)
 
Five Critical Roles for Building a World-Class FP&A Team
 
25
26
27
28
29
30
31
FP&A Business Partnering and AI: A New Era
 
All events for the year

Future Meetings

BPAI
The FP&A Trends Webinar FP&A Business Partnering and AI: A New Era

May 28, 2025

The Face-to-Face London FP&A Board: Data Management & Analytics: Unlocking FP&A Value
The Face-to-Face London FP&A Board Mastering Data in FP&A: Smarter Analytics, Better Decisions

June 5, 2025

FP&A Trends Webinar The Evolving Role of FP&A: From Number Cruncher to Strategic Advisor
The FP&A Trends Webinar Making FP&A Teams Fit for the Future

June 11, 2025

The Face-to-Face New York FP&A Board
The Face-to-Face New York FP&A Board From Insight to Impact: FP&A Business Partnering in Action

June 17, 2025

FP&A Trends Webinar Practical Steps for FP&A Analytical Transformation.
The FP&A Trends Webinar Unlocking FP&A Analytical Transformation

June 18, 2025

The FP&A Trends Webinar How AI is shaping the future of Financial Planning & Analysis (FP&A)
The FP&A Trends Webinar Transforming FP&A with AI: Maturity, Impact, and Future Roles

June 25, 2025

The Face-to-Face Sydney FP&A Board
The Face-to-Face Sydney FP&A Board Modern Financial Planning and Analysis (FP&A): Latest Trends and Developments

June 26, 2025

The Face-to-Face Singapore FP&A Board: Modern Financial Planning and Analysis (FP&A): Latest Trends and Developments
The Face-to-Face Singapore FP&A Board Modern Financial Planning and Analysis (FP&A): Latest Trends and Developments

July 8, 2025

AI/ML FP&A
AI/ML FP&A
Data and Analytics
Data & Analytics
FP&A Case Studies
FP&A Case Studies
FP&A Research
FP&A Research
General
General
Integrated FP&A
Integrated FP&A
People and Culture
People and Culture
Process
Process
Technology
Technology

Please register to receive the latest FP&A news, updates and tips.

info@fpa-trends.com​

              

Foot menu

  • FP&A Insights
  • FP&A Board
  • FP&A Videos

Footer countries

  • Amsterdam
  • Austin
  • Boston
  • Brisbane
  • Brussels
  • Chicago
  • Copenhagen
  • Dubai
  • Frankfurt
  • Geneva
  • Helsinki
  • Hong Kong
  • Houston
  • Kuala Lumpur
  • London Board
  • London (Circle)
  • Melbourne
  • Miami
  • Milan
  • Munich
  • New York
  • Paris
  • Perth
  • Riyadh
  • San Francisco
  • Seattle
  • Shanghai
  • Singapore
  • Stockholm
  • Sydney
  • Tokyo
  • Toronto
  • Washington D.C.
  • Zurich

Copyright © 2025 fpa-trends.com. All rights reserved.

0