6 lessons · about 2 h

Separate Model, Assumptions and Scenarios So Driver Changes Stay Clear

For CFOs, controllers, FP&A teams, accountants and finance operations professionals

You'll structure forecasts so assumptions sit in one documented place, scenarios are named versions of them, and every driver change is logged with its effect.

$47.00

See the full course page, with every lesson →

  • Taught one-on-one by an AI tutor, at your level, typed or out loud
  • Quizzes, flashcards and a certificate when you finish
  • Ready at learn.c-o-u-r-s-e-s.com as soon as you pay; the email tells you how to sign in

A focused 2-hour course for FP&A analysts: keep assumptions in one register, run scenarios as named versions of it, and log every driver change with its effect, including AI-drafted entries you check.

What you will be able to do

  • Give logic, inputs and outputs their own place: Split a forecast workbook into separate input, calculation and output areas. Find driver values buried in formulas, such as =B5*1.04, and move them out where everyone can see them.
  • Keep every driver in one documented register: Record each driver's value, unit, source, owner and effective date. Mark whether it is confirmed or provisional, and link the model's formulas to it so each driver has a single home.
  • Run scenarios without overwriting anything: Set up base, upside and downside cases as named versions of the register values. Switch between them with a selector, and name each one so it is clear what differs from the base case.
  • Log each driver change with its effect: Write entries that state the driver, old value, new value, effect, scenario and date. Measure the effect by comparing a key output before and after, and separate several simultaneous changes by applying them one at a time.
  • Use AI to draft the log, then check it: Give an AI assistant a structured input to draft an entry, then check every number, direction, unit, period and scenario name against the model's outputs before it goes into a management pack.
  • Trace any movement back to its cause: Compare two forecast versions using the selector, the register and the log. Decide whether the movement came from model logic, an assumption or a scenario choice, and spot any part of the change the log doesn't explain.

Taught at learn.c-o-u-r-s-e-s.com: lessons that adapt to your level, an AI tutor you can ask anything, quizzes and flashcards, and a certificate when you finish.

See the full course page, with every lesson →