Protocol
Search analogies by relation, not by topic
A useful analogy can look unrelated on the surface and still solve the same relational problem.
When it fits
- The obvious solutions in your own domain are exhausted and you want outside analogies that transfer a mechanism rather than a surface style.
When to avoid it
- Distant analogy can widen search but does not establish feasibility. Similarity at the relational level can still hide different incentives, physics, regulation or failure costs.
Why it matters
Abstract the problem before searching. State the relation or mechanism you need—buffering variability, separating conflicting flows, revealing hidden state, allocating scarce capacity—then look for other domains that solve that relation. Research on knowledge-rich creative problem solving and distant analogical transfer suggests that relational processing can make structurally relevant but superficially different analogs easier to retrieve. Bring back the mechanism, then test its boundaries in your domain.
Steps
- Rewrite the problem as a relation or mechanism without domain nouns.
- Search for other domains where the same relation appears.
- Map source roles and relations to the target explicitly.
- List what does not transfer before using the analogy as a solution.
An example
Instead of searching only for 'better support queue UI,' abstract the problem as 'how does a system prevent one noisy source from consuming shared capacity?' That opens analogies to network rate limiting, operating-system scheduling or physical traffic control.
Check your result
The analogy map names a shared relation, not merely a similar-looking object, and at least one nontransferable difference is documented.
Keep this limit in mind
- Distant analogy can widen search but does not establish feasibility. Similarity at the relational level can still hide different incentives, physics, regulation or failure costs.
Connected ideas
Useful withInvite one deliberately different discipline into ideation
Evidence and sources
A 2022 review argues that real-world creative problem solving in knowledge-rich domains often depends on reorganizing existing knowledge and on analogical transfer across situations that share deeper structure.
The review supports analogical transfer as an important research direction but does not validate a single operational analogy-search method.
Creative problem solving in knowledge-rich contexts · Abstract
Experiments on cross-domain analogy found that abstract relational processing and active generation can improve retrieval or transfer of distant analogies that lack obvious surface similarity.
The experimental tasks are simplified relative to professional problem solving; a retrieved analogy still requires domain-specific feasibility and boundary checks.
Promoting interdomain analogical transfer: When creating a problem helps to solve a problem · Abstract
In experiments reported in 2014, generating solutions to semantically distant analogies increased relational mapping on a later unrelated task, whereas merely evaluating analogies did not show the same transfer.
The result supports active relational generation, not the claim that farther analogies are always better or more creative in applied work.
Far-out thinking: generating solutions to distant analogies promotes relational thinking · Abstract