You need to estimate for each of the different business models individually. Then you will know what your revenue will be like for each avenue and it will be easy to estimate for combinations too.
A good way to improve the estimates withing your business model is to look at similar competitors publicly available information. A wealth of information is available on publicly traded companies like Zynga and Facebook by looking at their government filings that are required for publicly traded companies.
Hope this helps.
Estimations, so many factors to consider... It's almost impossible to predict what your business will do. Otherwise we all would be rich and happy!
Why not focusing on targets and milestones? If you set a target of X at the end of a period of time, then your attitude will be working towards an objective rather than I estimated wrongly.
To usefully estimate revenue for an unproven startup that hasn't even built a product yet is impossible. To estimate expenses with pretty good accuracy is often possible.
That's part of why Guy Kawasaki suggested to make a MAT (Milestone, Assumptions, Tasks) instead of a budget in his book "The Art of the Start ".
Don't write a business plan with revenue forecasts that aren't based on hard data from existing users -- that will be unconvincing for an experienced investor.
Especially don't write a business plan which gives the appearance that you don't know which business model is right, or that you will haphazardly mix business models as you go along -- that sends all the wrong signals.