Two-Pass Evaluation¶
Staged evaluation protocol — instantiates Evaluation Criteria Suspension During Divergence
Evaluates in two ordered passes — a permissive first pass on potential and learning, then a strict second pass on cost and feasibility — so rough-but-promising ideas are not killed by the harshest filter first.
Two-Pass Evaluation is not about suspending evaluation — it is about restoring it in the right order. It governs the moment convergence begins, insisting that criteria come back in a deliberate sequence rather than all at once. The first pass applies only generative criteria — potential, learning value, frame novelty — and its only power is to advance, not to kill. The harsh implementation filters — cost, feasibility, precedent, staffing — are held back for a second pass that runs only on the survivors. What makes it this mechanism is that its subject matter is the reentry gate and the ordering of criteria across it, not the generative phase at all: it decides which filter a still-rough idea meets first. Its bet is that the sequence of evaluation determines the outcome as much as the criteria themselves — let cost go first and every immature idea dies before its potential is even seen.
Example¶
A research foundation screens grant proposals for an exploratory funding line, and the review committee has a habit of killing unusual proposals in the first five minutes on budget and methodology grounds — exactly the proposals the exploratory line exists to fund. They switch to a two-pass protocol. In the first pass, reviewers may score a proposal only on two questions: is the underlying idea important, and would funding it teach the field something even if it partly fails? Nothing is rejected for cost or feasibility in this pass; a proposal either advances or doesn't, purely on potential. The one thing that can stop a proposal at any pass is the ethics-and-safety guardrail, which is never suspended.
Roughly a third of proposals advance. Only then, in the second pass, do reviewers apply the sharp filters — budget realism, methodological rigor, the applicant's track record — to that shortlist. An unconventional proposal with a shaky budget but a genuinely novel hypothesis, which the old process would have cut in minute three, survives the first pass on potential and gets its budget fixed in the second rather than being killed for it. The gate between the passes is explicit: the first pass formally closes before the second opens, so no reviewer smuggles cost objections into the potential round.
How it works¶
Two-Pass Evaluation is defined by its ordering discipline. First, it splits the criteria into an ordered sequence — a permissive band (potential, learning, frame diversity) before a restrictive band (cost, feasibility, precedent) — and fixes that order in advance so it can't be renegotiated mid-review. Second, the first pass may only promote: it advances candidates on generative merit and is structurally forbidden from rejecting on implementation grounds, which is what protects immature ideas long enough to be improved. Third, an explicit gate closes pass one before pass two opens, marking the true transition from divergence to convergence, and the guardrail set stays live across both passes — safety and legality can stop anything at any time, because those are floors, not criteria. The second pass then applies the harsh filters only to survivors.
Tuning parameters¶
- Criteria partition — which filters go in the permissive band versus the restrictive one. Putting more into the first band protects ideas longer but lets weak ones travel further; a leaner first band converges faster but risks early kills.
- First-pass power — whether pass one may only promote, or may also cut the clearly hopeless. Promote-only maximally protects fragile ideas; allowing gentle culling saves second-pass effort but reopens the early-rejection risk.
- Number of bands — a strict two passes, or a finer staircase (potential → risk → cost → sequencing). More bands smooth the reintroduction but add process weight.
- Gate strictness — how firmly pass one is closed before pass two begins. A hard gate prevents cost objections leaking into the potential round; a soft gate is lighter but lets the harsh filter creep forward.
- Guardrail reach — how much the always-live guardrail set is allowed to stop in pass one. Set too broadly, "safety" becomes a smuggled feasibility veto; too narrowly, real hazards slip through.
When it helps, and when it misleads¶
Its strength is that it names and fixes the thing that usually goes wrong at convergence: not whether you evaluate but the order in which criteria return. By letting potential go first, it stops the harshest, most concrete filter — usually cost — from silently erasing the value the whole divergent phase created. It is the disciplined form of affirmative judgment, the creative-problem-solving principle of appraising an idea's promise before its problems.[1]
Its failure mode is criteria rebound: if the ordering is not enforced, the toughest filters sneak into the first pass anyway — a reviewer who "can't help noticing" the budget — and the protocol collapses into ordinary immediate evaluation with extra steps. The mirror failure is a first pass so permissive that nearly everything advances, dumping an unfiltered flood on the second pass and defeating the point. The classic misuse is letting the permissive first pass masquerade as a real decision ("it passed pass one, so it's approved"), when passing one only means not yet rejected. The guarding discipline is a hard, visible gate between passes and a fixed criteria order agreed before any candidate is seen, so no one can reorder the filters to favor a preferred outcome.
How it implements the components¶
criteria_reintroduction_sequence— the protocol is this sequence: it fixes the ordered return of criteria, permissive band before restrictive band, as its central move.non_suspendable_guardrail_set— safety and legality stay live across both passes as floors that can stop anything at any time, distinct from the criteria being staged.convergence_reentry_gate— the explicit close of pass one is the gate from divergence into staged convergence, marking where evaluation formally resumes.
It does not implement idea_capture_buffer or the deferral of scoring against a suspended_criteria_register — those belong to its nearest twin, Deferred Scoring Queue; the queue parks ideas and scores them in one later batch, whereas Two-Pass is about the *order the criteria themselves return in.*
Related¶
- Instantiates: Evaluation Criteria Suspension During Divergence — governs the staged restoration of criteria at the convergence boundary.
- Consumes: Deferred Scoring Queue or Timeboxed Blue-Sky Sprint can supply the captured, unscored option set that this protocol then evaluates in ordered passes.
- Sibling mechanisms: No-Critique Rule · Yes-And Extension · Wild-Card Quota · Morphological Matrix · Deferred Scoring Queue · Separate Generator and Evaluator Roles · Timeboxed Blue-Sky Sprint
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: Two-Pass Evaluation operates as a bounded evaluation of existing evidence or work that produces a finding or disposition because it evaluates in two ordered passes — a permissive first pass on potential and learning, then a strict second pass on cost and feasibility — so rough-but-promising ideas are not killed by the harshest filter first.
Independent corroboration: The frozen evidence defines Two-Pass Evaluation as 'Evaluates in two ordered passes — a permissive first pass on potential and learning, then a strict second pass on cost and feasibility — so rough-but-promising ideas are not killed by the harshest filter first', so its operative form is Assessment, Review & Assurance.
Nearest alternative: Decision, Gate & Allocation — Two-Pass Evaluation includes features of a case-specific gate, selection, routing, prioritization, or resource disposition, but its defining operation is a bounded evaluation of existing evidence or work that produces a finding or disposition.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Innovation & Entrepreneurship
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Universal
Rationale: A permissive first pass for potential and learning followed by a stricter cost and feasibility pass is divergent-then-convergent innovation evaluation. The Design Council's Double Diamond explicitly separates broad exploration from focused selection, testing, rejection, and improvement; economics contributes cost criteria.
Related originating lineages:
- Accounting & Auditing — Ledgers, evidence, internal control, and assurance supplies a distinct formative lineage for the mechanism's two pass evaluation logic.
- Economics & Finance — economics_finance contributes economics, finance, and mechanism-design practice to this mechanism's defining operation—Evaluates in two ordered passes — a permissive first pass on potential and learning, then a strict second pass on cost and feasibility — so rough-but-promising ideas are not killed by the harshest filter first—without displacing the selected primary historical lineage.
- Engineering & Design — engineering_design contributes engineering design, reliability, and systems-safety practice to this mechanism's defining operation—Evaluates in two ordered passes — a permissive first pass on potential and learning, then a strict second pass on cost and feasibility — so rough-but-promising ideas are not killed by the harshest filter first—without displacing the selected primary historical lineage.
- Organizational & Management Science — organizational_management contributes organizational design, management, and operational governance to this mechanism's defining operation—Evaluates in two ordered passes — a permissive first pass on potential and learning, then a strict second pass on cost and feasibility — so rough-but-promising ideas are not killed by the harshest filter first—without displacing the selected primary historical lineage.
- Psychology — Experimental, clinical, and behavioral psychology supplies a parallel or contributing lineage for the mechanism's defining operation: evaluates in two ordered passes — a permissive first pass on potential and learning, then a strict second pass on cost and feasibility — so rough-but-promising ideas are not killed….
- Systems Thinking & Cybernetics — Systems thinking, feedback control, and cybernetics supplies a parallel or contributing lineage for the mechanism's defining operation: evaluates in two ordered passes — a permissive first pass on potential and learning, then a strict second pass on cost and feasibility — so rough-but-promising ideas are not killed….
Review resolution: The blind reviewers disagree on primary lineage (economics_finance versus organizational_management). Authoritative or primary research supports innovation_entrepreneurship as the best historical origin: A permissive first pass for potential and learning followed by a stricter cost and feasibility pass is divergent-then-convergent innovation evaluation. The Design Council's Double Diamond explicitly separates broad exploration from focused selection, testing, rejection, and improvement; economics contributes cost criteria. The cited UK Design Council, Framework for Innovation: Double Diamond directly supports the mechanism's defining operation. All independently supported contributing domains are retained without an arbitrary cap. origin_mode=cross_disciplinary_synthesis records lineage, while domain_reach=universal records later applicability separately from provenance.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Researched adjudication after independent review; high confidence.
Sources consulted:
References¶
[1] Puccio, G. J., Murdock, M. C., & Mance, M. Creative Leadership: Skills That Drive Change. SAGE Publications (2007). Defines affirmative judgment as appraising an idea’s positives and potential before turning to concerns. registry ↩