Educational MNC Does its Homework and Passes Cash Forecasting Test

Pearson has operations in 200 countries with 160 million users. This means granular insight into its cash position is vital.

Educational publisher and edtech specialist Pearson has mastered how to improve its cash predictions thanks to new AI-powered software, which also helped it win the Gold trophy at the TMI Awards 2023 in the Best AI Deployment category.

Pearson has developed its processes using a new cloud-based cash forecasting solution that uses TIS’ cash forecasting software-as-a-service (SaaS) option to back-test its model to identify gaps or inconsistencies in its predictions. It then adjusts its logic for subsequent forecasts, thereby increasing accuracy over time.

The actual vs. forecast analysis is driven by an AI model that learns as it goes along, enabling Pearson to increase accuracy and efficiency when managing its cash forecasting. The complexity of managing cash across so many currencies and collating information from large numbers of stakeholders previously limited the team’s ability to analyse and respond to the data.

The most obvious benefit has been in improved consistency of group operating cash flow, which now consistently achieves 100%, or close to, as a percentage of operating profit. This comes from tighter working capital understanding and management and has resulted in improved efficiency in FX conversion, lower interest costs, and lower levels of gross debt.

Across-the-board visibility

Getting its cash position right ahead of time, so that funds are in the correct place to oil supply chains, meet payroll, pay for acquisitions or to optimise liquidity options is essential for Pearson, as with any firm. But it’s not easy to do this efficiently when operating in 200 countries with a 160 million-strong user base and a $4.7bn turnover.

The platform gives Pearson visibility of all entities managed by the central treasury team, allowing for multicurrency forecasts and to include the impact of all expected treasury settlements. This means the company can make robust plans for its liquidity and FX needs.

Not all cash flows are suitable for AI, either due to small sample sizes (dividends or acquisitions for example) or due to inherent unpredictability. Nonetheless, the AI model helps the team to better understand where efforts would be beneficial to change processes to make the timing of receipts and payments more predictable and those benefits can be quantified.

Relevant financial data can easily be pulled on to the SaaS platform with AI functionality to run analyses on primary cash flow drivers. Data is continuously reviewed and analysed to ensure that it is properly categorised, as this forms the basis of the AI-based forecast. The models used for prediction are also interrogated to ensure that they are consistent with how the business expects to perform in the future and adapted as appropriate.

In addition, reports can now be more easily shared by the treasury with other departments and stakeholders thanks to the cloud-based SaaS functionality, which also means collaboration is easier. Elsewhere, this functionality means the platform can handle frequent updates as well as being scalable for the future.

Overcoming hurdles, managing growth

Gaining granular insight into the company’s cash position is extremely valuable and the team has worked to progressively develop its visibility and understanding of its cash flows. The TIS system complements existing Pearson systems by offering one designed to provide the insights needed by the treasury team. Granular data from the AI can – and does – positively impact cash flow dynamics under the new solution. 

Interestingly, the TIS system has shown its worth as Pearson’s business has changed. The ability to segment by business, currency and flow type means that it is relatively easy to test hypotheses about growth in receipts from new products, the impact of cancelled exams, as happened during the pandemic, and acquisitions. The flexibility of the new solution can handle reformulating its models better because data inputs are faster and more granular. It does also mean the system has coped well with post-pandemic challenges such as rising interest rates, global conflicts, and increased volatility, which all make good cash forecasting more vital than ever.