In our Design for Scale lab, design question #3 asks us: “What’s our optimal fidelity for scale?”
What is fidelity? Very simply, it’s the degree of exactness with which something is copied and reproduced.
Why is it important?
Well, in some fields it’s both easy and critical that that the fidelity of an intervention is high. For example, when you’re sick and you take a pill you’re taking the same pill that others who share your diagnosis will take. The dose, timing and other instructions that you receive must be followed in order to get positive results.
Designing an intervention that should be replicated with high fidelity is both necessary and relatively straightforward for many fields. What about development? As we scale, we know that we can’t simply take our exact model and apply it to new geographies, new demographics and new sectors. We know that it won’t work if we don’t adapt.
Conversely, adapt too much or let the model be watered down and we may not be able to get them same positive impact we did in our pilot or the early stages of our programs.
David Butler, VP for Innovation at Coca-Cola, helps one of the world’s largest companies design for scale using a simple framework of determining what elements are “fixed” and which are “flexible.”
To break it down for us, he uses the analogy of lego bricks. The whole video is great, but for the part on legos start at the 8 minute, 30 second mark and end at 13 minutes, 50 seconds.
If you were able to watch our introduction video, you know that one of our hypotheses for why it takes the development sector a long time to scale is that we’re often using our project cycles to test new adaptations of our innovations. When we launch a “pilot,” that can mean waiting for two-three years to get feedback and course correct at the midterm evaluation.
While we might need to wait until the midterm to start getting hard data on impact, we don’t need to wait several years to get valuable feedback. Why design and launch a full solution, when you can run rapid tests on prototypes first?
So in our human-centered design lab, we’re thinking about how to prototype versus pilot and how this can give us just one new tool to move faster.
GRID explains how all this works in one fantastic graph
One aspect that is taking us some time to get our heads around – how do you prototype anything that’s not a product, such as services or systems? It’s still early days for applying HCD to international development, but here are some examples we love!