How to automate cash forecasting processes

By Martin Gillespie | 10 December 2018

Cash forecasts are vital tools for all businesses. Depending on what the corporate objectives of the cash forecast are, they may vary in form, time horizon, and granularity. What unites them, however, is how frustratingly long the process can be if it is run entirely manually.

As a result, increasingly, CFOs and Finance Directors are looking for ways to improve the process through automation.

In this article we’ll review what cash forecasting automation is, how it can benefit treasury and finance teams, which parts of the process can be automated, and how full automation can be achieved.

Cash forecasting automation: an overview

Working with treasury and finance teams around the world, we’ve seen many cash forecasting processes with varying degrees of automation. These fall on a spectrum with some having “end-to-end” full automation, some with key parts of the process automated, and others are fully manual, labour intensive processes.

What benefits can automation bring?

The benefits of cash forecasting automation fall into three broad categories:

  • Increased efficiency
  • Increased accuracy
  • Reduced risk

Increased efficiency

Because of the range of data inputs, workflows, reporting, and communications exercises that feed a cash forecasting process, switching from a manual to an automated process can save a lot of time.

Even in the largest corporates, treasury teams are often small and overburdened. Removing a particularly heavy part of the workload can therefore free them up to focus on the strategic activities that add real value back to the business.

Increased accuracy

Increasing accuracy makes a cash forecasting a sharper tool. Automation enables robust accuracy analysis, where any issues can be identified and improvements made. Owing to the time savings brought by automation, this process can be repeated frequently, driving continuous improvements in forecast accuracy.

When the process is carefully designed and mapped from the outset, automation can leverage data already in the corporate technology infrastructure (ERP systems, TMS systems, bank systems, etc.) to make sure that the forecast is informed by all relevant data sources.

Reduced risk

Automating part or all of the process reduces the risk of human error, thereby improving confidence in the forecast. In addition, the use of specialised cash flow forecasting software enables thorough analysis to be performed quickly and easily. Therefore, any errors or discrepancies can be identified and investigated.

Which parts of the cash forecasting process can be automated?

To discuss how each of the elements of the cash forecasting process can be automated, we have broken the process into two parts, input data automation and workflow automation.

Input data automation

There are typically two sets of data that input into a cash forecasting process, forecast and actual. Automating the input of this data to the process encompasses both collection and classification (which can be achieved by either algorithmic or rules-based classification).

Forecast cash flow data

  • Accounts receivable / accounts payable (AR / AP). Forecasting AR and AP is a fundamental part of the cash forecasting process. Quite often, short-term AR and AP are already recorded in ERP or accounts systems. If this data is available, it is therefore beneficial to include it in the forecasting process. For more information on this element of the process, please read our post on increasing accounts receivable forecasting accuracy.

  • Treasury Management Systems (TMS). Depending on the system, and how it is being used, the TMS can also hold some of the required forecast cashflow data (such as financing or intercompany cashflows).

  • Budgeting / planning systems. Longer-term forecasted cashflows may be contained in budgeting or planning systems.

Actual cash flow data

  • Bank account data. Most commonly stored in MT940 or BAI2 files, automating bank account inputs means identifying and extracting the relevant information automatically.
  • ERP / TMS systems. In a similar way, actual cash flow data can be pulled automatically from these systems and fed directly into the cash forecasting process.

Workflow automation

Again, for the sake of clarity, we have separated the workflows that surround a cash forecasting process into two categories, input workflows and output workflows.

Input workflows

  • Automated workflow triggers and submission management. In an automated process, when a new submission cycle is due, an automated workflow sets up the new version, then triggers any required actions (such as reminder emails) and catalogues the version accordingly. With automated workflow triggers, these emails can be generated and sent within the system automatically. This automatic versioning enables an easier forecast vs forecast, and forecast vs actual analysis. In addition, approaching a deadline date, users can be prompted by further automated reminders.
  • Intercompany reconciliation. Intercompany reconciliation is often a very time consuming and laborious task for a company’s head office finance or treasury team. To automate this element of the process, the full process must be designed in a way that considers the automation of intercompany reconciliation up front. In the CashAnalytics system this is managed through a specialist counterparty driven intercompany tool.
  • Variance justification. Where users have submitted data that exceeds a predetermined threshold, an automated system can detect the variance and prohibit it from being submitted without a comment to explain the variance. In this instance, a user would be required to provide feedback on why the threshold had been exceeded before the submission could be made.

Output workflows

  • Reporting and analytics automation. Where the process is automated, scheduled reporting can be automatically generated from the system and circulated to key stakeholders as required. For example, this might consist of a weekly or daily cashflow report, bank account reports, or reports on funding requirements.
  • Real time feedback. When actual cash flows are updated (whether manually or collected automatically) a feedback notification can be issued automatically informing the user of any variation to the forecast. This enables quick and easy amendments to forecast figures, thereby improving accuracy.

How to achieve cash forecast automation

As discussed above, a cash forecasting process can be fully automated end-to-end, or it can be separated into its constituent parts, which can each be automated in isolation. Regardless, any aspect of the cash forecasting process is difficult if not impossible to automate without the use of specialist cash flow forecasting software.

About CashAnalytics

We have helped treasury and finance teams in multinational companies around the world automate their cash forecasting process. In most cases, we can fully integrate the system end-to-end, in six weeks. During this process, comprehensive project management with senior members of the CashAnalytics team ensures smooth collaboration across a company’s business units with minimal impact on day-to-day operations.

If you have any questions, or would like to view a demonstration of our software, please do not hesitate to contact us.

Become a bobsguide member to access the following

1. Unrestricted access to bobsguide
2. Send a proposal request
3. Insights delivered daily to your inbox
4. Career development