The Precise Prediction application represents the most advanced level of ChurchSalary’s compensation analysis tools. Available exclusively to Pro+ Members, this application uses artificial intelligence (AI) and statistical modeling to forecast salary ranges and analyze compensation trends with exceptional accuracy.
Precise Prediction is designed for ministries who want a deeper understanding of how various factors—such as budget, experience, and education—affect pay. It allows unlimited analyses for any number of employees and locations.
Note: The “Beta” label indicates that this first iteration of ChurchSalary’s AI model is still in development. Please treat its predictions as a good first start or as a second opinion for your own analysis. They are by no means definitive or authoritative at this point. As the model improves and goes through retraining cycles, the accuracy of its predictions will increase.
How to Use Precise Prediction
Note: The Precise Prediction application is available only to Pro+ Members.
Log in to your ChurchSalary account.
Click on Launch Dashboard.
Select an Employee Profile, then click the Precise Prediction button in the Applications section.
You can use Precise Prediction across multiple campuses or churches, as well as different saved employees—all without leaving the page.
Inside the Prediction app, you can:
Select a different employee, church, or campus saved in your profiles or
Select one of the 18 available position options from the drop-down.
You can fine-tune key factors such as Location, Budget Size, Employment Status, Weekly Hours, and Church Size before generating predictions.
💡 Tip: The accuracy of your prediction depends on how complete and up to date your Employee Profile is. Be sure to review position title, church size, budget, and other relevant fields before generating a report.
How does the ChurchSalary AI Model work?
The Director of ChurchSalary, Aaron Hill, worked closely with Dr. Marcus Schwarting (PhD, Univ. of Chicago) to create an AI model that would output a total salary prediction based on a set of inputs.
To do this, the entirety of ChurchSalary’s 2.0 dataset was combined with location data in order to train an internal AI model — essentially a mathematical model — that weighs a wide array of inputs in order to generate a single output or prediction. The final ChurchSalary AI model was trained to predict total salary using the following variables as input.
Employee
Position
Status (FT/PT)
Experience (relevant)
Education
Ordination status
Church
Budget
Size
Location → four key metrics are sourced at two levels: (1) County and/or (2) Metro/Micropolitan Statistical Area or (3) PUMA (Public Use Microdata Area).
Cost of Living Index
Median Household Income
Per Capita Income
Median Age