All use cases
Ideate

Adapt cross-industry analogies to a local challenge

Compare sourced cases from other contexts, extract transferable mechanisms, and turn only verified analogies into locally testable adaptation hypotheses.

Why

The workflow problem

Teams miss mechanisms used elsewhere, while generic web synthesis encourages false transfer and treats different local contexts as interchangeable.

Cross-industry comparison can expand the solution space beyond familiar patterns. AI can accelerate comparison, but source verification and contextual judgement determine whether an analogy is useful rather than misleading.

Concrete output

What you will produce

An analogy matrix containing the original case, source URL, transferable mechanism, contextual mismatch, local adaptation hypothesis, and proposed test.

How

Run the workflow

  1. Specify the local challenge

    Humans define the evidence-backed local need, affected people, setting, non-negotiable constraints, and harms that a borrowed mechanism must not introduce. Keep a short list of what is not transferable.

  2. Gather source-linked cases

    Use a source-aware tool or human research to collect original case material, not summaries alone. A human records publisher, date, context, outcomes claimed, and a stable source URL for each candidate.

  3. Extract mechanisms and mismatches

    Ask AI to compare only the supplied cases and to separate the underlying mechanism from surface features. Every output remains a draft until a human checks the original source.

    Compare cases without claiming transferability

    Compare the source-linked cases below against our local challenge. Use only supplied information.
    
    Local challenge and constraints: [insert]
    Cases: [case ID | original URL | context | observed mechanism | limits]
    
    Return: case ID | transferable mechanism | surface feature to ignore | contextual mismatch | local adaptation hypothesis | assumption requiring local research. Do not invent outcomes, infer equivalence between populations, or recommend implementation.
  4. Verify and score the analogies

    Humans inspect originals, score contextual distance and risk, consult local expertise where needed, and discard cases with weak sourcing or harmful assumptions.

  5. Formulate local experiments

    Convert the strongest mechanisms into small adaptation hypotheses with a target context, owner, and learning question. Test with affected people before committing to delivery.

Human–AI partnership

Who contributes what

AI contribution

Organise supplied cases, compare mechanisms, and identify stated mismatches. It does not establish that an external result is true or transferable.

Human responsibility

Verify sources, understand local conditions, involve affected people and domain experts, reject false analogies, and define responsible experiments.

Stop rule

Stop AI use when a case lacks an original source, an analogy treats vulnerable populations as interchangeable, local constraints are missing, or a proposed adaptation bypasses local-user research.

Check before use

Watch out

Case studies often omit failures and power conditions. Check publication incentives, date, geography, governance, and who was excluded. Never use an analogy to justify a decision without local evidence.

Evidence base

What supports this workflow

One harvested educational-project reference supports the idea of AI-assisted international case research, but it does not evaluate this workflow. The analogy matrix is a method-led reconstruction with limited direct evidence for AI benefit.

Explore all use cases