LEADING KPIS

The first four themes above address user experience and adoption issues vs. technology and data considerations—how to integrate algorithmic forecasting into new ways of working and the most salient, people-centric challenges.

Human-centered design focuses on taking a holistic approach to developing a forecasting capability by incorporating individual requirements of the key user groups and the interactions across them. In this approach, user groups, also known as personas, are not defined by their function or business unit but by their business objectives. For example, while a vice president and an analyst may both sit in FP&A, their goals will be different. The former will need the means to toggle effectively between scenarios that model different business decisions, while the latter will seek the functionality to build those scenarios, assign values, set relevant thresholds, and complete their work accordingly.

For that reason, promoting successful adoption of a solution depends on asking two vital questions: Did we account for the right user groups or personas when designing the capability, or were the personas defined too broadly? And did we successfully connect the dots between each user group to create a comprehensive solution?

Adoption challenges could signal a missed or misaligned persona during the initial design. It’s critical to define meaningful personas through comprehensive discovery and research. Doing so involves working closely with users to understand their role within the forecast process, intended outputs, and interdependencies with other groups. The personas then become the basis of the solution design, creating a unique experience for each user group while addressing any cross-persona interactions.

A persona-agnostic design approach for an algorithmic forecasting solution can lead to an overly complex user interface. If every feature and function is available to all, users may not be able to navigate the solution easily and may be inclined to prepare their forecast outside of it. As a result, aggregation can’t be conducted systematically without the proper inputs. Even if only one stakeholder group is directly affected, rejection of the capability could cause a chain reaction, leading to widespread failed adoption.

Another consequence could be a lack of governance around sequencing, handoffs, and workflow visibility, elements that are critical to collaboration and aggregation within the forecast process. For example, without proper user groups in the workflow, tax rates could be applied to an aggregate forecast while adjustments are made at a lower level, resulting in unnecessary reconciliation issues and the loss of a key feature of the forecasting process. As algorithmic forecasting introduces a new way of working, adding persona-based embedded workflow into a forecasting solution can also support change management and training, which aids in the adoption of the solution.

Understanding the affected user groups and accounting for any personas that might have been missed along the way can be a powerful mechanism to improve the user experience and build a comprehensive forecast using the algorithmic platform.

Capability Building

Algorithmic forecasting capabilities can unlock benefits such as improved productivity—enabling a more detailed and granular forecast than before and generating more accurate outcomes without driving additional effort to produce it. But while these solutions serve as a central component of the forecasting process, they are not intended to replace the process entirely. In fact, deploying a forecasting solution presents an opportunity to rethink current processes and take advantage of the automation it enables.

However, ensuring the solution fits into an organization’s forecast approach is of critical importance. Because a top-down forecast typically doesn’t require the same granularity and complexity as a bottom-up forecast, each requires different considerations. Adoption challenges can often be traced back to a disconnect between the algorithmic solution and the broader forecast process. Introducing complex models and large datasets without the mechanisms to understand them can create a “black box” perception, whereby the machine outputs seem opaque and unexplainable.

Focusing on such elements as divisions, segments, and regions that drive most of the activity or volume for the business may be best suited for the algorithmic forecast, the better to maximize its impact while reducing the noise created by superfluous data. Considering how best to leverage the forecasting solution requires understanding the rationale behind the data collection: Are you including immaterial line items or products in the forecast? For example, are you forecasting for revenue or inconsequential meeting expenses within your time and expense line? The solution isn’t intended to consider every line item but rather to forecast the line items that matter most to business planning.

To validate the reliability of the forecast at selected intersections, continuous self-checks can inform the right level of granularity without compromising accuracy. Exception-based reporting allows users to conduct regression analysis and compare the outputs from the algorithmic forecasting system to the established historical or base trends to detect changes and signals. This feature enables finance business partners to home in on the most relevant drivers, improving accuracy and boosting confidence in the results.

Algorithmic forecasting is most effective for areas with the highest materiality. New intersections can be introduced and coverage expanded as the process matures. However, sharpening the focus to the key business components can be a quick win for enhancing transparency and driving stronger adoption of the solution.

Opportunities for improvement, meanwhile, when coupled with effective change management, can foster meaningful collaboration between users and the algorithmic forecasting solution.