This question can be split to two:

1. What should I fill so that the model would be meaningful for other users?

It is hard to draw any specific guidelines here. Basically: the more historic data, the better. However, often the company has changed its structure so dramatically that long history data loses its meaning.

If any dramatic changes has not happened, we would recommend that every company would have at least two years history data as complete: divisional (if company reports divisions) and quarterly net sales and EBIT figures. Furthermore more rough (income statement and balance sheet) historic data should be for additional two years. So, altogether 4-5 years of historic data makes a decent model.

2. What should I fill in order to get model work fluently?

This means that you need a certain amount of data in order to get model work. This is what you must fill any case:

I-divQ sheet
  • Division names - At least one division (names as 'Group' for example)
  • Latest actualized current year quarter - Mark with 'x' to row 4
  • Latest actualized income statement to cumulative values section
  • Fill at least cumulative Q4
  • Latest actualized current year quarter to cumulative values section - Not needed if Q1 not actualized
  • Current year (Y+0) estimates in quarter level
  • Net sales and EBIT (either relative or absolute)
  • Year Y+1...Y+3 estimates in full-year level
  • Net sales and EBIT (either relative or absolute)
I-main sheet
  • The last estimate year (as terminal values)
  • Net sales growth and EBIT-%
  • Latest actualized balance sheet
  • No of shares for current year