Work / Note
Design for translation
Research becomes a principle, a principle becomes a tool, a tool becomes a changed practice. Each step loses something, and the last one fails most often — usually for reasons that have nothing to do with the quality of the research.
The situation
A body of evidence is solid, well replicated, and widely cited. Practitioners in the field, asked directly, will often agree with it. And the practice on the ground does not reflect it, sometimes decades later.
The standard explanation is a gap — a distance between research and practice that could be closed with better dissemination. I think that framing is wrong in a way that matters, because it implies the problem is distance, and therefore that the solution is transmission.
The question
What if it is not a gap but a chain, where each link is a translation that costs something and can break independently?
The move
The first translation turns findings into principles. A finding is bounded — it holds for these learners, this content, these conditions. A principle is portable and therefore over-claims by construction. This step is necessary and is always a small betrayal of the evidence, so it should be done deliberately by someone willing to say where the boundary was.
The second turns principles into tools. A principle tells you what to optimise for; it does not tell you what to do on Monday. A tool — a protocol, a checklist, a template, a worked example — makes the principle operable. This is the step most often skipped, because it is unglamorous and reads as a downgrade from the idea.
The third turns tools into changed practice, and this is where it breaks. A tool has to survive contact with an existing routine that already works well enough, in an environment with no spare time, where the person adopting it bears all of the cost up front and receives the benefit later and diffusely. Under those conditions the rational move is usually to keep doing what you were doing.
What it produced
The first consequence is that the last step has to be designed for, not hoped for. That means asking what the tool displaces, not just what it adds. Anything that is purely additional is competing with everything else that is purely additional, and it will lose.
The second is that adoption cost belongs in the design. A tool that takes 40 minutes to learn and saves five minutes a week is a good deal over a year and an impossible sell in the moment. Front-loading the payoff, even at the cost of some fidelity, is often the difference between a tool that is used and one that is admired.
The third is that fidelity is negotiable and should be negotiated openly. Practitioners adapt tools. They always have. A tool designed on the assumption it will be used exactly as specified will be adapted anyway, badly and invisibly. It is better to state which parts are load-bearing and which are free.
The point
What transfers
The reason this page exists, rather than the story that produced it.
- Treat translation as a chain of lossy steps rather than a gap to be closed. It tells you where to look when something has not landed.
- The step from tool to practice is the one that fails, and it fails for structural reasons — competing routines, up-front cost, deferred benefit — not for lack of conviction.
- Ask what your tool displaces. If the answer is nothing, it is additional, and additional things lose.
- Say which parts are load-bearing. Adaptation is going to happen; the only choice is whether it happens with guidance.