1. recognize you need a model- mental model(semiconsciously in mind)
- mathematical model (more precision)
2. acknowledge your models limits- incorrect(logic or assumptions are wrong, do not use this)
- incomplete (characteristic shared by all models, add to this)
3. expect the unexpected - some factors will be overlooked (changes in environment and interaction can create factors that are no longer relevant)
4. understand the use and user -models are evaluated as a triplet (model, application, user)
- workings of a model are transparent, consistently applied, results can be reproduces and verified by others
5. check the infrastructure - benefits are determined how innovation is introduced
- pace is different, can cause imbalance