It seems like EAs have around six different common ways of comparing interventions, each with their own pros and cons:

1. Cost-benefit Analyses
   1. Pros
      1. Allow us to compare interventions within and across cause areas while using a consistent metric
   2. Cons
      1. Makes it hard to compare interventions that have different effects
      2. Can result in interventions with many different positive effects being under-utilized
         1. I think citizens initiatives are a good example of this
            1. They create momentum for the farm animal activism movement, create social awareness, and produce legal change, but cost-benefit analyses usually only focus on one of these effects.
2. The ITN Frawework
   1. Pros
      1. Allows us to compare different cause areas at a very macro level
   2. Cons
      1. It can de-emphasize cause areas which may not be neglected but still possess high leverage interventions
3. Models
   1. Pros
      1. Allows us to map out complex problems and compare the effects of different interventions at a more granular level
   2. Cons
      1. We often lack the proper data to create models that are really worth using
      2. It takes a very long time to create models that are very valuable
4. Projections (A type of model)
   1. Pros
      1. Allow us to see how different interventions would result in change if implemented over a long time period
   2. Cons
      1. Often make overly simplistic assumptions about how to fix the problem at hand
5. Theories of change (Another type of model)
   1. Pros
      1. Give us a sense of how an intervention could help to create large-scale change
   2. Cons
      1. Often don't enable us to strictly compare the effectiveness of different interventions
6. Public discussion
   1. Pros
      1. Enable us to find the very best arguments for an against different interventions and cause areas
   2. Cons
      1. Fundamentally limited by the information available to those involved in the discussion

I'm curious. What are there other ways of comparing interventions and cause areas? I think it's important that we avoid measurability bias, but it also seems like relying on intuition is probably a very bad approach to use instead.