White Paper. Not Just Knowledge, Know How! Artificial Intelligence for Finance!

Transcription

1 ` Not Just Knowledge, Know How! White Paper Artificial Intelligence for Finance! An exploration of the use of Artificial Intelligence (AI) in the management of Budgeting, Planning and Forecasting (BP&F) by the Financial Office. November 2017 Maximizing Human Potential Transformation ~ Centers of Excellence ~ Artificial Intelligence FirstAlign 227 W 4th Street, Suite 107, Charlotte, NC

2 Executive Summary FirstAlign ~ White Paper Financial success is largely dependent on where the market is heading. Artificial Intelligence or AI offers an unprecedented ability to improve the accuracy and predictability of Budgeting, Planning and Forecasting (BP&F). This is able to reduce the expensive and time intensive exercise of the annual budgeting process. The objective of using AI is the broader concept of using machines to carry out financial tasks in a way that we would consider as smart. Machine Learning or ML are the applications of AI in that we are able to give machines access to data and let them learn for themselves using quantitative techniques and thereby reduce estimating errors. AI solutions with machine learning algorithms continue to selfimprove as more data is fed into it. The fact is, the use of business intelligence, particularly through AI, to more effectively automate the BP&F process is growing more important as speed, flexibility and agility takes center stage. BP&F is a 3-step process for determining and detailing an organization's short and long-term financial goals. The process tends to be managed by the Finance Department under the auspices of the Chief Financial Officer (CFO). AI impacts BP&F in the following ways: Analyzes unstructured and structured data to reveal patterns, suggesting actions based on trends. Generates signals and indicators for monitoring financial scenarios to help give organic view of the financial risks that improve accuracy over time. Enables near real-time understanding of financial objectives, facilitating changes in direction where required. Enables financial risk assessment through algorithms to make more accurate predictions of market movement. The objective of using Artificial intelligence (AI) is to better streamline financial managerial tasks and increase the accuracy of forecasting with quantitative predictions, through self-improving systems, thus reducing budget, planning and forecasting errors. AI solutions can more effectively automate the planning process is growing more important as speed, flexibility and agility take center stage. Key business driver data often lies outside of the usual suspect systems. To take advantage organizations need to collect, monitor and react to data in near real-time. Overall innovation is changing current BP&F methods, with more organizations adopting AI solutions to increase their financial accuracy. A shift towards perpetual, scenario based BP&F models in using AI is a directional step change ; that helps to eliminate ceaseless manual processes that have financial implications on daily operational realities and direction. Manual solutions won't capture dynamic operational and financial complexities in a timely fashion. As AI becomes adopted more into the mainstream, financial forecasting and modeling will become neither reactive nor proactive, instead near real-time and accurately predictive. For more information, please visit to discover ways we can help FirstAlign. All Rights Reserved. P a g e 1

3 Table of Contents FirstAlign ~ White Paper EXECUTIVE SUMMARY 1 TABLE OF CONTENTS 2 INTRODUCTION 2 WHAT IS BUDGETING, PLANNING AND FORECASTING? 2 UNDERSTANDING BP&F AND AI 2 FINANCIAL MODELING 4 WHY AI FOR FINANCE? 5 USE CASES 6 OBSERVATIONS & CONCLUSIONS 7 REFERENCES 8 ABOUT US 8 Introduction This white paper is an exploration of the possibilities for Artificial Intelligence (AI) in rethinking the development of enterprise finance. We explore the current thought around Finance, in particular Budgeting, Planning and Forecasting (BP&F) and how AI can improve accuracy and outcomes. We investigate the focus of AI on the development of valuable automated solutions that require less human intervention. We look into finance through AI-powered business intelligence and analytics. This review of why and how AI fits a finance office, and how AI can introduce intelligent applications to identify and mitigate impacts on BP&F processes, before not after the fact, promotes financial management decisions and value. What is Budgeting, Planning and Forecasting? Budgeting, planning and forecasting (BP&F) is a three-step process for determining and detailing an organization's long and short-term financial goals. The process is usually managed by an organization's Finance Department under the auspices of the Chief Financial Officer (CFO). Planning: outlines the organization's future financial direction and expectations; example next year to five years. Budgeting: documents specifying expenditures and how the overall plan will be executed month to month, quarter to quarter, or year to year. Forecasting: uses accumulated historical data to predict financial outcomes for future months, quarters or years. Understanding BP&F and AI Today s business world is driven by customer demand. As patterns vary it becomes a challenge to develop accurate forecasts. The goals of BP&F are to reduce uncertainty and provide benchmarks for monitoring actual performance. Artificial Intelligence techniques are used to improve this accuracy, thereby enhancing the bottom line [1]. Artificial Intelligence is a technology tool. Planning is a business management task. AI is applied as a cognitive function to do tasks associated with human intelligence like "learning" and "problem solving". AI powered solutions can tackle difficult problems involving intuitive judgment or requiring the detection of data patterns which elude conventional analytical techniques FirstAlign. All Rights Reserved. P a g e 2

4 As the backbone of the enterprise, the Finance Office has been leading the charge in leveraging machine learning and artificial intelligence to deliver real-time insights, inform decision making and drive efficiency across the enterprise. Therefore Finance, and in particular BP&F, is one of the business units that sees the impacts of these technologies on day-to-day activities. This includes everything from automating payments to calculating risk to maintaining records [2]. Budgeting, Planning & Forecasting Financial forecasting is an estimation of future financial outcomes for an organization, industry, or country using historical data. Forecasting methods utilize the following: Sales information including orders, channels, backlogs, and inventory. Payroll, rates and headcounts. Financial budgets and actuals (both current and historical). Production assumptions (L/OH, New Lines, Vendors). Insurance costs, medical, workman s comp, and causality. Corporate costs, technology, infrastructure, facilities, rebates, and pensions. Market information, competitor intelligence, etc. Difficulties of such data points are their time to compile and process, notwithstanding their inevitable historic nature. This often prevents real time data being available when decisions are required to be made. Budgeting and Forecasting Approaches: Towards the later part of each year organizational leaders begin setting business objectives. At the top-level an organization may focus on margin, revenue, profit, and/ or volume depending on market conditions for the upcoming year. In doing so, a top-level algorithm (ex: 4% +$.05) can be established for the entire organization or individual business lines. BP&F processes and timelines are realigned alongside business rules to accomplish the upcoming year s planning and budgeting cycle. Figure 1: BP&F Lifecycle Planning and budget meeting schedules are established, tasked with compiling assumptions and gathering the necessary data. This culminates in the notification of regional and individual management teams that the budgeting process has begun. Months of actuals and previous year s data are loaded into financial systems that tend to end with the current year s last month of budgets. First pass reviews allow for adjustments before final review and publication of the plan. AI forecasting adds continued quantitative and qualitative analysis, or a combination of both to the planning and budgeting process. Quantitative forecasting or objective analysis can follow both a bottom up and top down approach. Qualitative forecast or managerial analysis, however tends to be a top down perspective. AI quantitative forecasting is developed using the following techniques [4] : 2017 FirstAlign. All Rights Reserved. P a g e 3

5 Time Series: The future tends to look and behave like the past. This technique takes snapshots of a time sequence along successive, equally spaced points, to predict future values or outcomes based on that previously observed. Relational: The future is dependent on the direction of a variety of factors. For example, new housing declines might be a function of interest rates and higher unemployment conditions. Financial Modeling Financial models tend to use information from one or more data sources. This is incorporated into applications, that serve the needs of many departments and lines of business when doing budgeting and planning. An Organization s Chart of Accounts (COA) lists all of the accounts used and which defines item classes or categorizations for which money or the equivalent is spent or received. Figure 2: Forecasting Process COA is used to organize the finances, thus segregating expenditures, revenue, assets and liabilities in order to give management a better understanding of the financial health of the organization. Each account in the COA has a unique identifier and is associated with real financial data via data entry or batch processing loaded daily, weekly or monthly. Additional snapshots of the same financial data are loaded into the Artificial Intelligence platform. BP&F Financial Models BP&F models are mechanisms for the presentation of information from one or more data sources that serve the needs of multiple organizational groups. These models are designed based the BP&F functional requirements defined by the organization. Example models for financial management are: Sales Planning Model Profit and Loss Model Auto-Adjusting-Rolling- Forecast Models Break Back for Planning Model OPEX Model (All Cost Centers) Alignment and synchronization of sales planning and budgeting functions. Includes the ability to update budgets and forecasts quickly for more effective sales planning, revenue projections and strategic initiatives. Helps estimate revenues, costs and expenses over specific time periods. Enables better management and profit generation by amending revenue projections and costs, published into P&L Statements. A strategy used to continuously re-forecast as conditions change. Financial actuals are loaded, re-adjusting for better future accuracy. Automated planning with calculations for automation of budget processes (proportional, relative proportional, equal, percent change, growth, straight line). It utilizes top-down approaches when beginning the budgeting process by allowing Break Backs. The break back scheme executes at any point by focusing downwards with re-adjustments from the bottom-up. Handles operational expenditure thus helping the budgeting and planning of the ongoing running costs. This consists of Expense Cubes that store various costs by Categories from the various Departments and Lines of Business FirstAlign. All Rights Reserved. P a g e 4

6 OPEX Rules P & L Rules BP&F Calculation Rules and Parameters There are calculations and rules used by each of the BP&F models when doing budgeting and planning. Example formula groups for financial management are: Dimension Rules Disclose Rules Calculation and dimension rules for operating expenses that helps budgeting and planning of ongoing running costs. Various departmental expenses or the entire enterprise include wages, taxes, insurance, costs of sales, capital gains, profit income, etc. Calculation formulas and rules used in profit and loss models. These can consist of expenses, sales offset calculations, promo discounts, commissions, fees, recoveries, rebates, royalties, loss recovery, miscellaneous expense/ income, etc. Formulas and rules that link an individual or group of items into a cube or dimension. These can be administration accounting links for calculation formulas. Calculations for general accounting for regulatory reporting speed and cycle time. Why AI for Finance? No one sees the future with absolute clarity. During budgeting and planning cycles, organizations spend a huge amount of resources (people, time and technology) to complete these tasks [3]. Traditional planning tends to be two-phase activity: 1. Setting a vision for the future; and 2. Allocating resources. The inherent flaws in this ideal include a misconception that budgets and forecasts aim to predict the future, while their scope at times is setting targets. Artificial intelligence can analyze in near or real-time financial risk signals as well as estimate impacts of a variety of factors. This in turn increases the profit return rate of success and helps to optimize decision making. Commonly used AI systems in BP&F are: [4] Figure 3: Organizational BP&F Maturity and AI Automating Routine Decisions and Reports: Just 17% of time is spent on strategic activities, a deficit of automation being the major culprit for this lack of efficiency. AI, through machine learning predictive algorithms, can provide an additional edge in driving innovation within the accounting back office. Processes such as, procure-to-pay, order-to-cash, record-to-report, even closeto-disclose can be turned over in whole or part to AI; allowing professionals to focus on what really matters. Supplementing Human Process with AI Machine Support: In areas where processes cannot be fully trusted to a machine, certain tasks can. Pattern recognition within complex data sets can highlight potential discrepancies or abnormal activities that may require further human judgement and circumstantial reasoning. Risk Management and Fraud Detection: Machine learning has the ability to churn through large sets of data, transactions, identifying abnormal patterns or unique behaviors FirstAlign. All Rights Reserved. P a g e 5

7 AI is able to, in a matter of seconds; determine and analyze impacts of financial risks through adaptive learning techniques based on operations. Enhancing Dashboards: While a simple dashboard gives professionals access to the organization s health status with financial reports, an AI enhanced dashboard is able to better inform of more dynamic risks including recommendations. This allows financial professionals to better evaluate the past, and anticipate the future. Improve Forecasting and Reporting: AI takes budgeting, planning and forecasting to a higher level of accuracy, giving the financial professional more time to think tactically and strategically. Advanced Scenario Analysis: AI makes it possible to calculate the probability of financial scenarios before they occur, making it easier to test scenarios over a defined time period. It reduces time spent analyzing both internal and external data, improving availability across a broader range. Use Cases 1. Budgeting & Planning Challenge: Annual planning and budget cycles were costing manufacturing and distribution company millions of dollars each year. The BP&F process was taking 3 4 months, including a large percentage of management resources, to prepare, and establish the next year budgets. The budgeting process began with loading 9 months of current year actuals alongside the ending 3 months for the previous year. Mid-September usually begins the organizations planning and budgeting cycle, by mid-december there is a final review and publish the plan. Still after the new year the management team would still have to come back together to adjust for the last ending 3 months of actuals from the prior year, realign and publish the plan. The last quarter of the year is highly dependent on seasonal volume and highly effects overall profit and loss. Demand at times is difficult to forecast because it was strongly correlated with the strength or weakness of consumer discretionary spending. The company wanted to reduce the time senior management was spending on planning and budgeting, while at the same time increasing accuracy of their forecasting. Solution: An AI based BP&F platform with an interface for budgeting that incorporated Break Backs for planning. This allowed scenario development, with business intelligence reporting, to preparing possible outcomes including perpetual Auto Adjusting Rolling Forecasts as accruals came in from each line of business. It showed key financial indicators (signals) of what was most likely developing and allowed reaction using pre-defined plans of action and recommendations. Figure 4: Budgeting Modeling Using AI 2017 FirstAlign. All Rights Reserved. P a g e 6

8 2. Forecasting Challenge: A consumer based company predicted their product price and profits margins of products using traditional financial data and negotiations with sales channels. They wanted to increase price, however historical sales and sales channels, were not aligning to better future pricing. This solely limited their outlook to past trend-based predictions, creating significant challenges in their forecasting of price change. This affected sales planning and future prices. Figure 5: Forecasting Perception on Prices. Solution: The company adopted AI forecasting into their BP&F process to help predict best pricing scenarios of their consumer products. This incorporated market information, competitor intelligence and ML algorithms to showcase consumer market willingness to purchase against market conditions. The AI analyzed historical time series data and estimated the impact of a variety of factors. With AI, the company gained confidence in their price adjustments. They were able to forecast sales and profit margins when increasing pricing, including monitoring outcomes when their sales channels advised them otherwise. Observations & Conclusions The objective of using Artificial intelligence (AI) is to better streamline financial managerial tasks. This in turn increases the accuracy of quantitative predictions for forecasting, thus reducing budget, planning and forecasting errors. AI solutions using machine learning algorithms self-improve over time as they are trained and more data is applied. The use of business intelligence, particularly through AI, to more effectively automate the planning process is growing more important as speed, flexibility and agility take center stage. Key business driver data often lies outside of the usual suspect systems that the organization uses and relies on. To take advantage, organizations will need to collect data, monitor and react to business drivers in near real-time. This is changing the way the Finance Office is budgeting, planning and forecasting (BP&F). More organizations are adopting AI BP&F solutions to increase the accuracy of their forecasts by improving data methods and expanding efforts to gather market intelligence on customers and competition. A move towards perpetual, scenario based planning and monitoring is a directional step change ; replacing current ceaseless planning cycle processes dependent on many manual tasks. Manual solutions won't capture dynamic operational and financial complexities in a timely fashion. As AI becomes more adopted into the mainstream, financial forecasting and modeling will become neither reactive nor proactive. Instead it will become near real-time and accurate predicting in using machine learning will be reality. For more information, please visit to discover ways we can help FirstAlign. All Rights Reserved. P a g e 7

9 References FirstAlign ~ White Paper [1] Artificial Intelligence has a Rising Impact on Financial Markets. [2] 7 Ways Artificial Intelligence and Machine Learning Will Impact the Finance Office. [3] Artificial Intelligence Applied to Business Planning. [4] Artificial Intelligence Techniques Enhance Business Forecasts. [5] Five ways Artificial Intelligence is Changing Tax. About Us NOT JUST KNOWLEDGE, KNOW HOW FirstAlign helps organizations innovate, transform, and lead. If you don't influence change, who will? FirstAlign enables you, your team and your organization to lead with targeted tools supported by a creative, seasoned and diverse group of professionals. We work across major industry sectors, public organizations, Fortune 2,000 and government authorities, utilizing startup mindsets in making complexity simple. Customers count on FirstAlign's strategic advisory services and target operating models, to help them transform uncertainty into opportunity. For more information, please visit IMPORTANT NOTICE: The information contained in this document represents the current view of FirstAlign with respect to the subject matter herein contained as of the date of the publication. FirstAlign makes no commitment to keep the information contained herein up to date and the information contained in this document is subject to change without notice. As FirstAlign solutions must respond to the changing market conditions, FirstAlign cannot guarantee the accuracy of any information presented after the date of publication. The document is presented for informational purposes only. FIRSTALIGN PROVIDES THIS PUBLICATION AS IS WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE FirstAlign. All Rights Reserved. P a g e 8