Our innovation framework serves as a fundamental guideline for everyone involved in the design, implementation, or evaluation of a GIZ Data Lab experiment. It is intended to facilitate systematic, goal-oriented, and responsible data innovation without being prescriptive and limiting. 

The framework builds on existing innovation models - including the Double Diamond model - and is constantly adjusted and refined based on discoveries from the Lab’s pilot phase. With regular check-ups, we intend to make the process, expectations, roles, and criteria for decision-making transparent to everyone from the very beginning.

Small A GIZ DATA LAB Innovation Framework

Key Dimensions of each Phase

P0 Initiate

The “Initiate” phase serves as a thinking and pre-selection stage to determine if an issue, idea, or problem lends itself to systematic exploration and innovation.

  • P1 Understand

    The “Understand” phase is intended to provide a high-level understanding of the problem at hand. It lays the foundation for the subsequent “Define” phase while seeking to narrow down a specific issue (or issues) that can be addressed through an experiment.

  • P2 Define

    In the “Define” phase, the experiment team reviews and further distills the insights collected in the “Understand” phase, working out the particular issue to be addressed while preparing all necessary information required for the prototyping process.

  • P3 Prototype

    During the “Prototype” phase, the teams develop one or several prototypes - along with a set of hypotheses - that can be tested. A prototype is not confined to a technological solution; it can also include methodologies, processes, and frameworks in various forms.

  • P4 Testing

    The “Testing” phase requires strict planning in order to obtain as much information as possible relative to the costs. During this phase, each prototype undergoes a real-world test, after which each is likely to require adaptation and refinement through a further series of tests.

  • P5 Scale

    Based on its outcome(s), not every experiment will prove suitable for scaling. However, it is important to include the “Scale” phase during planning, as it may impact the overall approach, choice of partners, and/ or prototype development.