Revenue forecasting is one of the most important activities in FP&A. Forecast accuracy affects budgeting, hiring plans, investment decisions, cash planning, and executive reporting. Most finance teams already forecast revenue using historical performance, growth assumptions, sales inputs, and management guidance. The challenge is that many of these processes still depend heavily on spreadsheets and manual adjustments.
That approach can work, but it becomes harder to manage as the business grows. Revenue may have seasonal patterns, changing growth rates, unusual periods, or business events that are not easy to capture with a simple percentage increase.
Prophet provides one way to bring a more structured forecasting approach into the process. When used alongside Workday Adaptive Planning, it can help finance teams generate a data-driven forecast, review forecast ranges, and refresh the output as new actuals become available.
The Problem with Traditional Revenue Forecasting
Many revenue forecasts start with a simple method. A team may take last year’s revenue and apply a growth percentage. Another team may average recent months and extend that trend forward. Others may build spreadsheet formulas that adjust revenue manually by quarter, region, product, or business unit.
These methods are easy to understand, but they often miss changes in the underlying pattern. Revenue does not always grow evenly. Some businesses have stronger Q4 performance. Some have slow summer months. Some experience step changes from product launches, pricing changes, large renewals, or customer churn.
When those patterns are handled manually, the forecast can become inconsistent. It may depend more on individual judgment than on a repeatable process.
A better forecasting process should use historical data, identify recurring patterns, allow business overrides, and give finance teams a clear way to explain the result.
What Prophet Does
Prophet is an open-source time-series forecasting library designed for business forecasting use cases. It works well when the data has a date field, a historical value, and patterns such as trend or seasonality.
For revenue forecasting, Prophet looks at historical revenue and separates the forecast into understandable components. These may include long-term trend, seasonal movement, special events, and forecast uncertainty.
This is useful for FP&A because finance users do not only need a forecast number. They also need to understand why the forecast moved and how much uncertainty exists around it.
Prophet does not replace business planning. It creates a statistical forecast that finance teams can review, compare, and adjust based on business knowledge.
Building the Revenue Forecast
The basic input structure is simple. Prophet requires a date column and a revenue value column. Historical revenue is loaded into the model, and the model uses that history to estimate future revenue.
Once the model is trained, it can generate forecast values for future months, quarters, or years. The forecast can then be compared against actual results, reviewed for reasonableness, and used as a baseline in the planning process.
This is where the approach becomes useful for FP&A. Instead of starting with a blank forecast or a flat growth percentage, the team starts with a forecast that reflects historical trend and seasonality.
The forecast is still not final. Finance users should review it against pipeline, bookings, pricing decisions, customer behavior, and management expectations.
Understanding the Forecast Output
The output from Prophet usually includes more than one number. For each forecast period, the model can provide a base forecast, a lower forecast range, and an upper forecast range.
The base forecast represents the expected value based on the historical pattern. The lower range gives a more conservative view. The upper range gives a higher-side view.
This structure fits naturally with FP&A scenario planning. Many organizations already maintain base, downside, and upside versions in Workday Adaptive Planning. Prophet’s forecast ranges can support that process by providing a starting point for each scenario.
The important point is that these ranges should not be treated as approved scenarios by themselves. They are model-generated ranges. Finance teams still need to review whether the ranges make sense for the business.
Connecting the Forecast to Workday Adaptive Planning
Once the forecast is generated, the output can be loaded into Workday Adaptive Planning. The forecast values may be loaded into a forecast version, a scenario version, or a reporting cube depending on the design.
A common pattern is to load the base forecast into the main forecast version and load the lower and upper ranges into downside and upside versions. This allows users to compare the forecast scenarios inside Adaptive Planning using existing reports, dashboards, and planning sheets.
The integration design should be clear. The team needs to define the source actuals, the forecast grain, the planning dimensions, the version strategy, and the load process.
For example, if revenue is planned by product, region, customer segment, and month, then the forecast output should align to that same level of detail where possible. If Prophet is only forecasting total revenue, then the team needs to decide whether the forecast is used at a summary level or allocated back to planning dimensions.
Rolling Forecast Use Case
Prophet can also support a rolling forecast process. As new actual revenue is available, the historical dataset can be refreshed and the forecast can be regenerated.
This fits well with monthly FP&A cycles. Actuals are loaded, the model is refreshed, the new forecast is generated, and the output is loaded into Workday Adaptive Planning for review.
The value is not only speed. The value is consistency. A repeatable forecast process helps reduce manual rework and gives finance teams a stable baseline to review each cycle.
Business adjustments can still be made in Adaptive Planning after the forecast is loaded. That is usually the right design. The model provides the baseline. Finance and business owners provide the final planning judgment.
Benefits for FP&A Teams
Using Prophet with Workday Adaptive Planning can reduce manual forecasting effort and provide a more structured starting point for revenue planning. It can help identify revenue trends, capture seasonality, produce forecast ranges, and support scenario planning.
It can also improve forecast conversations. Instead of only debating a spreadsheet assumption, the team can review the historical trend, forecast range, and business drivers together.
This is especially useful when leadership wants to understand not only what the forecast is, but why the forecast changed.
Practical Design Considerations
A Prophet forecast is only useful if the source data is reliable. The historical revenue data should be clean, complete, and aligned to the same planning grain used in Adaptive Planning.
The team should also review one-time events. Large customer wins, unusual revenue adjustments, acquisitions, divestitures, or accounting changes can distort the historical pattern. These items may need to be adjusted or handled separately before the forecast is generated.
Forecast ownership is also important. The output should have a clear review process. Finance users should know which values came from the model, which values were adjusted by the business, and which version is used for reporting.
The model should also be tested against prior periods. Before using it in a live planning process, finance should compare forecast output against known actuals and review whether the results are reasonable.
Final Observation
Revenue forecasting is a strong use case for a structured forecasting model because the business problem is clear, the data usually exists, and the output can be reviewed directly by FP&A.
Prophet can help generate a baseline revenue forecast using historical trend and seasonality. Workday Adaptive Planning can then be used to review, adjust, report, and manage that forecast as part of the normal planning process.
The right design is not to let the model own the forecast. The right design is to use the model as a starting point, bring the output into Adaptive Planning, and let finance teams apply business judgment before the forecast is finalized.
For organizations looking to improve revenue planning, this is a practical place to start: clean historical revenue data, a repeatable forecast method, clear version design, and a review process owned by FP&A.