Coarse To Fine Search¶
Search broadly at a coarse level first, then refine only the most promising regions in more detail.
The Diagnostic Story¶
Symptom: The team applies high-resolution effort to a large number of candidates before doing any broad screening, exhausting review capacity before the most promising regions are reached. Alternatively, a coarse dashboard is used as a final judgment when it was only meant to guide deeper search. Either way, high-value options are missed, premature commitment locks in early choices, and nobody can explain why certain candidates received detailed attention and others were ignored.
Pivot: Define a coarse representation of the full space, use it to identify which regions deserve refinement, concentrate detailed evaluation there, and build in an audit mechanism to detect and recover from false negatives before the screen becomes permanent.
Resolution: Search cost drops because detailed evaluation is reserved for promising regions, not the entire space. The original search boundary stays visible so attention never becomes unfairly narrow. Final choices rest on high-resolution evidence where stakes require it, not only on the coarse screen.
Reach for this when you hear…¶
[drug discovery] “We ran full assays on three hundred compounds before screening — we burned the budget before we even got to the likely hits.”
[investigative audit] “I need a quick scan of all the accounts before I commit my forensic team to a handful of suspects, or we will miss the real outliers.”
[urban planning] “The site-selection criteria knocked out eighty percent of parcels in a day — now we do the detailed environmental review on the ten that actually qualify.”
When This Archetype Applies¶
Partial catalog groundingSome structural conditions are represented by existing abstractions, but no sufficient condition set is fully represented.
Diagnostic problem
The solution, diagnosis, design, or search space is too large for detailed evaluation of every possibility, but crude screening alone risks missing important options.
What this problem means
The structural problem is the tension between **coverage** and **depth**. A broad search protects against tunnel vision but is too shallow to decide well. A fine-grained search gives richer evidence but cannot be applied to every possibility. Without a staged resolution strategy, teams either waste detailed effort on low-value options or use coarse screens as if they were final judgments.
The common failure is not just inefficiency. It is also invisibility: once a region is filtered out, nobody can tell whether it was genuinely low value, merely poorly represented by the coarse model, or never considered seriously.
Show the applicability expression
Applicability expression3 distinct conditions
groundedpartly groundedopen
3 conditions, all required.
3Required in every casenumbered 1–3
These hold no matter which pattern applies.
Exhaustive search infeasible · open
The option, region, hypothesis, design, or parameter space is too large for exhaustive fine-grained evaluation.
The source archetype describes the situation as follows: The number of possible options, regions, hypotheses, designs, or parameter settings is too large for full fine-grained evaluation. The normalized requirement above isolates the load-bearing portion used in this condition set.
Detailed evaluation costly · grounded
Authoritative detailed evaluation is expensive, slow, scarce, risky, or attention-intensive.
The source archetype describes the situation as follows: Detailed evaluation is expensive, slow, scarce, risky, or attention-intensive. The normalized requirement above isolates the load-bearing portion used in this condition set.
Refinable search structure · open
The search space has neighborhoods, categories, scales, branches, stages, or families that permit selective refinement.
The source archetype describes the situation as follows: The search space has structure: neighborhoods, categories, scales, branches, stages, or candidate families that can be refined selectively. The normalized requirement above isolates the load-bearing portion used in this condition set.
Other requirements and context (3)
Why these sit outside the expression
Solution feasibility — it describes whether the intervention can work, not whether the diagnostic problem exists.
Supporting context — it may accompany or help interpret the situation, but it is not a load-bearing condition in a sufficient diagnostic set.
Deployment constraint — it constrains how the intervention must be deployed, not the situation that calls for it.
Solution feasibilityThere is a lower-resolution representation that can cheaply preserve at least some signal about where detail is worth adding.
Use this archetype when the search space is too large for detailed review of everything, but a cheaper representation can still reveal where deeper investigation should go. In this archetype, the relevant feasibility condition is: There is a lower-resolution representation that can cheaply preserve at least some signal about where detail is worth adding. It identifies something that must be possible or available for the intervention to be workable.
Supporting contextEarly precise analysis would create false precision, sunk-cost bias, or premature commitment to a few cases.
It is weak when the early representation is arbitrary, when the cost of missing an option is unacceptable, or when rules require full detailed review before any exclusion. In this archetype, the relevant contextual consideration is: Early precise analysis would create false precision, sunk-cost bias, or premature commitment to a few cases. It helps interpret the situation or strengthens the practical case for examining the archetype.
Deployment constraintMissing a promising region is possible, so the coarse filter needs an audit, exception, or backtracking mechanism.
It is also invisibility: once a region is filtered out, nobody can tell whether it was genuinely low value, merely poorly represented by the coarse model, or never considered seriously. In this archetype, the relevant deployment constraint is: Missing a promising region is possible, so the coarse filter needs an audit, exception, or backtracking mechanism. It identifies a boundary that responsible implementation must respect.
Coverage
1 of 3 conditions grounded · 2 open.
Mechanisms / Implementations¶
- Multi-Resolution Search (
multi_resolution_search): Implements the archetype by scanning at multiple levels of resolution and escalating detail only where the lower-resolution pass indicates value, uncertainty, or risk. - Coarse Grid Search (
coarse_grid_search): Implements the archetype in parameter or design spaces by testing a rough grid first, then searching more finely near promising grid regions. - Funnel Process (
funnel_process): Implements the archetype in review or product workflows by moving many candidates through cheap early screens before detailed evaluation of a smaller set. - Diagnostic Narrowing (
diagnostic_narrowing): Implements the archetype by starting with broad symptom, signal, or evidence groups and then applying more specific tests to likely diagnostic regions. - Portfolio Screening (
portfolio_screening): Implements the archetype by using coarse financial, strategic, risk, or feasibility filters before intensive due diligence on selected opportunities. - Progressive Candidate Review (
progressive_candidate_review): Implements the archetype by reviewing applications, proposals, designs, or options in stages, with deeper review reserved for candidates that pass earlier screens or uncertainty checks. - Design Downselection (
design_downselection): Implements the archetype in design work by comparing rough concepts first and investing detailed engineering, prototyping, or testing in the most promising concepts. - Search Tree Pruning with Refinement (
search_tree_pruning_with_refinement): Implements the archetype when a tree or hierarchy is explored shallowly first, then expanded more deeply along selected branches while keeping audit checks for pruned branches. - Coarse Grid Search: Evaluates a bounded parameter or design space on a rough regular grid first, then places a finer grid around the most promising cells and repeats until improvement stalls.
- Design Downselection: Implements the archetype in design work by comparing rough concepts first and investing detailed engineering, prototyping, or testing in the most promising concepts.
- Diagnostic Narrowing: Implements the archetype by starting with broad symptom, signal, or evidence groups and then applying more specific tests to likely diagnostic regions.
- Funnel Process: Implements the archetype in review or product workflows by moving many candidates through cheap early screens before detailed evaluation of a smaller set.
- Multi-Resolution Search: Implements the archetype by scanning at multiple levels of resolution and escalating detail only where the lower-resolution pass indicates value, uncertainty, or risk.
- Portfolio Screening: Implements the archetype by using coarse financial, strategic, risk, or feasibility filters before intensive due diligence on selected opportunities.
- Progressive Candidate Review: Implements the archetype by reviewing applications, proposals, designs, or options in stages, with deeper review reserved for candidates that pass earlier screens or uncertainty checks.
- Search Tree Pruning with Refinement: Implements the archetype when a tree or hierarchy is explored shallowly first, then expanded more deeply along selected branches while keeping audit checks for pruned branches.
Related Abstractions¶
Abstractions this archetype builds on — directly (a source ingredient) or as a related pattern. Links follow the typed catalog namespace.
Built directly on (3)
- Approximation: Good-enough representation.
- Progressive Refinement from Core Model: Incremental refinement.
- Search and Retrieval: Locate and extract information.
Also references 7 related abstractions
- Branch and Bound: Systematic search with pruning.
- Constraint: Limits possibilities to guide outcomes.
- Convergence: Movement toward stable state.
- Feedback: Outputs influence inputs.
- Optimization: Finds best solution under constraints.
- Scale: Properties change with size.
- Uncertainty: Incomplete knowledge.
Variants¶
Narrower or domain-specific specializations that share this archetype's core structure. Recognized variants are established; candidate variants are provisional.
Multi-Resolution Search · mechanism family variant · recognized
Search across explicit resolution levels, using each level to decide where the next level of detail should be applied.
Coarse Grid Search · implementation variant · recognized
Evaluate a rough grid of possibilities first, then search finer neighborhoods around promising coarse cells.
Funnel Screening · implementation variant · recognized
Move many candidates through cheap early screens, then reserve deeper review for candidates that pass or remain uncertain.
Diagnostic Narrowing · domain variant · recognized
Start with broad diagnostic categories and selectively apply finer tests to the most plausible or highest-risk possibilities.
Editorial Notes¶
Problem Classification¶
Classification: Decision, Search & Optimization Failure → Hidden, Unbounded & Poorly Pruned Search Space
Problem kernel: flat detailed search is infeasible while crude screening is unsafe
Rationale: The candidate space needs staged narrowing so low-cost coarse evidence reduces scope without discarding promising fine-grained options.
Independent corroboration: The earliest necessary condition in the frozen evidence is: The solution, diagnosis, design, or search space is too large for detailed evaluation of every possibility, but crude screening alone risks missing important options. That is a search space discovery and reduction problem because A large or falsely bounded option space cannot be navigated because viable regions are hidden, exploration is undirected, or narrowing and pruning lack safe justification.
Review outcome: Independent reviewer agreement; high confidence.