Machine Learning and the Future of Intelligent Planning

When a business operates in a steady state, planning is straightforward. You can build forecasts and plans based on the previous two years’ actuals with reasonable confidence that the future you’ve mapped out will more or less come to pass.

In case you haven’t noticed, however, the business world has grown increasingly unsteady. Outdated approaches to planning (static, episodic, manual, and largely spreadsheet-based) no longer work; in fact, they can hurt you competitively by luring you into accepting false or outdated assumptions, and by dulling your ability to respond to revenue and demand shocks.

The pace of change was accelerating even before businesses were rocked by pandemic-related lockdowns, layoffs, and supply chain disruptions. And change will remain a fixture long after pandemic disruption is over.

Looking for Help in Intelligent Planning

Organizations everywhere are acknowledging that they must find a confident way forward—one that allows them to operate with agility. They are adopting modern best practices like continuous planning, which replaces static annual plans with rolling forecasts and constantly updated budgets using always-fresh actuals, and company-wide planning, which extends proven FP&A planning practices to other parts of the business, including HR and sales.

Those capabilities are already in use at agile organizations throughout the world. But CFOs and other leaders don’t expect business environments to become less challenging over time. They’re looking for what’s next. They’re looking for intelligent planning.

With intelligent planning, businesses harness data to work harder for them. This includes historical data, which shows where you’ve been, and forward-looking analytics that help you assess and anticipate where you’re going.

Developing a Business Navigation System

That forward-looking piece is becoming more important as the future grows inherently less predictable. Decision makers want to evaluate multiple options quickly. They want to know those options have been modeled not on instinct or guesswork, but on fresh financial and operational actuals and external data.

Understandably, they also seek accuracy. And here, they’re getting an assist from machine learning (ML) algorithms that ingest, process, and learn from data. You can think of ML as a prediction engine—and an enabling element of artificial intelligence. As these algorithms feed analytics and scenario-building systems, they learn from the outcomes of their predictions and forecasts, and that information helps create even more accurate predictions in the future. It’s a continuous feedback loop designed to improve over time.

This is why machine learning is much more than just some buzz phrase to drive clicks. It’s the underpinning of intelligent planning, which in turn is transforming planning into a business navigation system for operational agility.

The Scope of Intelligent Planning

You can see this in how Workday has advanced intelligent planning with machine learning in ways that give managers continuous guidance about next steps and possible outcomes. For instance:

Proactively surfacing anomalies in plans and planning data. This helps organizations identify potential errors in their planning data—errors that, if left undetected, could prove costly down the line. Because it’s powered by ML, it learns over time, especially when managers decide to either accept or ignore the anomalies that the system has flagged.

Automatically scouring data to detect potential outliers. Like anomalies, outlier data can send finance, sales, HR, or other managers down the wrong path and skew outlooks in an overly positive or negative direction. This ML-driven feature automatically converts data into insights, and allows finance and other business leaders to focus on managing exceptions—which in turn streamlines their own planning processes.

Already today, we’re seeing enterprises embrace these capabilities as they chart their own futures. (Our success in evolving intelligent planning is, we believe, a key reason Gartner named Workday a Leader in its 2020 Magic Quadrant for Cloud Financial Planning and Analysis Solutions.)

A future Driven by Innovation

But intelligent planning is continuously evolving, thanks in large part to machine learning, and more is in the works.

For instance, planners can expect to incorporate external data and macro indicators into models that allow managers to correlate their business (from headwinds to opportunities) with the market at large. And new intelligent forecasting capabilities will help sales and supply chain managers to anticipate and predict changes in demand for products or services. In sales, managers will be able to gain more control and insight over assigning and managing territories with new territory recommendation features that weigh various inputs to arrive at an optimal territory mix.

Workday solutions tap an intelligent data foundation encompassing high volumes of operational data (from enterprise applications and external systems) with financial performance information, so decision makers gain a 360-view of their business.

These capabilities may be next-level now, but before long they’ll become an expected element of enterprise-class planning. And we at Workday will push these technologies further. Because business is anything but steady, and those responsible for making their businesses agile need all the help they can get.

This blog was written by Matt Shore, chief product strategist at Workday.

Watch a Demo and See Workday Adaptive Planning in Action.