What do your customers care about? Running tested features in released products. What do we too often do in organizations? Placating enough people across the organization to making some progress on everything, but finishing nothing. Does that make sense to you?
Instead, we can use agile and lean project portfolio management. We don’t have to rely on dog-and-pony shows, prediction, math, ROI, or estimation. We have much better tools: project value, backlog value and demos
We ask questions that help you differentiate each project's value from each other. How will this project move the organization forward? We start with the zeroth question: Should we do this project at all? We continue with qualitative questions and then ask quantitative questions.
We start with qualitative questions, such as using business value points, which are based not on any predictive number, but might be based on your product's position in the market, or who is waiting for your projects to be complete, or the risk of waiting of waiting for your projects to be done. What's critical is your discussion around the business value, not the points themselves. I'll supply you with some meaty starting questions for the business value discussion.
If you ask quantitative questions first, they devolve into ROI and you never ask the good qualitative questions. If you do need to have some idea of the budget, I’ll address that question, too. But budget is the wrong question. Value is the right question, and we’ll discuss why.