Collection
Search for evidence as a reproducible process, not a lucky query
Find stronger evidence by decomposing concepts, testing retrieval, chaining citations, documenting searches and using AI as a search assistant without letting generated summaries become the evidence.
12 elements · Explore in any order.
2 checklists · 2 principles · 8 protocols
12 elements
Split the question into a few search concepts
Search engines need concepts, not your entire internal monologue.
Read the idea ProtocolBuild a terminology map before optimizing the query
You cannot search a vocabulary you have not discovered yet.
Read the idea ProtocolUse known-good papers as a search test
A search that cannot find the paper on your desk deserves suspicion.
Read the idea ProtocolWalk backward from a strong paper
A good paper contains a map of the work that convinced its authors.
Read the idea ProtocolWalk forward to see what changed after a key paper
The paper's reference list knows the past. Its citing papers know part of the future.
Read the idea ProtocolIterate citation chaining only when it keeps adding value
Snowballing is useful until it becomes shoveling.
Read the idea ChecklistDeduplicate before spending attention on screening
Reading the same paper three times is not triangulation.
Read the idea ProtocolKeep the exact query that produced the evidence
A conclusion without its search trail ages badly.
Read the idea PrincipleSeparate discovery tools from evidence sources
The thing that found the evidence is not necessarily the evidence.
Read the idea ProtocolUse AI to expand search language, then verify the vocabulary
AI is useful for widening the vocabulary; it is not the authority on which terms the field actually uses.
Read the idea ChecklistScreen against explicit inclusion criteria, not relevance vibes
Interesting is not the same thing as eligible.
Read the idea PrincipleStop the search when the next evidence is unlikely to change the decision
A search can be methodical without being infinite.
Read the idea