Deferred Scoring Queue¶
Process queue — instantiates Evaluation Criteria Suspension During Divergence
Routes every raw idea into a holding queue that is scored later in one batch, so capture is never interrupted by evaluation.
The Deferred Scoring Queue suspends evaluation by decoupling it from capture in time. Every idea, as it arrives, is dropped into a queue and marked unscored; the scoring itself happens later, in a dedicated batch pass over everything the queue has accumulated. What makes it this mechanism is that it is an artifact-and-workflow, not a facilitation rule: the queue is a durable holding buffer with an explicit "not yet judged" state, and the suspension is enforced by the pipeline rather than by anyone's self-restraint. It does not tell people how to generate ideas, nor which criteria come back in what order — it only guarantees that no idea is scored at the moment of capture. Its bet is that if scoring can never interrupt intake, intake never has to slow down to defend itself.
Example¶
A software team runs a standing intake for product ideas. In the old flow, every idea posted to the channel drew an immediate reply — "out of scope," "we can't staff that," "duplicate" — and people stopped posting anything ambitious. They install a Deferred Scoring Queue: ideas go into a board column literally named Unscored, with a posted rule that no one attaches a score, a priority, or a verdict there. The criteria that will apply later — reach, effort, strategic fit — are written down on the board so everyone knows scoring is coming, just not now.
For two weeks the queue fills with 90 ideas, from the obvious to the far-fetched, none of them ranked. Then, in a single scoring session, the team runs the whole queue against the posted criteria at once. Because the ideas are judged as a batch and against a stable rubric, the scoring is more consistent than the old reflexive replies ever were — and the queue, now annotated with scores, becomes a record of what was proposed and how it fared, useful when the same idea resurfaces next quarter. Several ambitious ideas that would have been killed on arrival survive to the batch, and one clears the bar.
How it works¶
The queue works by three structural commitments. First, an intake with a suspended-scoring state: every entry lands as unscored, and the workflow forbids writing a score, rank, or verdict in that state. Second, a register of the deferred criteria kept visible beside the queue, so suspension reads as "scoring is scheduled," not "scoring is cancelled" — the difference that keeps it honest. Third, a batch scoring pass: at a set cadence, the accumulated queue is scored all at once against the register, which both restores evaluation and makes it more uniform than piecemeal reaction. The queue keeps the scored results as a running trace. What it deliberately omits is any opinion about the order criteria should return in — it applies them together, in one pass.
Tuning parameters¶
- Batch cadence — how often the queue is drained and scored. Longer intervals accumulate more diverse material but delay decisions and let the backlog balloon; shorter intervals keep things moving but creep back toward scoring-on-arrival.
- Queue visibility — whether the unscored queue is public or private during accumulation. Public queues invite building on others' ideas but tempt onlookers to score covertly; private ones protect suspension but lose cross-pollination.
- State discipline — how strictly the "unscored" state is enforced against stray priorities and side comments. Tight discipline preserves the suspension; loose discipline lets evaluation leak in through the back door.
- Rubric stability — whether the deferred criteria are fixed before intake or set at scoring time. Fixed rubrics make batch scoring consistent; late-bound ones adapt to what actually showed up.
- Trace retention — how long scored queues are kept. Longer retention builds institutional memory of what's been tried; it also costs upkeep and can anchor future judgments.
When it helps, and when it misleads¶
Its strength is twofold: intake never stalls to defend itself, and batch scoring against one rubric is measurably fairer than reflexive, one-at-a-time reaction, where order and mood sway the verdict. The move echoes the nominal group technique, which deliberately separates silent idea generation from the group's later evaluation and ranking.[n1]
Its failure mode is the archetype's signature one: a queue that is never drained. Deferral is only responsible if the batch pass actually happens; a stalled scoring session turns "suspended" into "abandoned," and the queue becomes a graveyard that quietly buries every ambitious idea it was meant to protect. A subtler failure is covert scoring — people can't write a verdict, so they signal it in the ordering or in side comments, and the suspension is only nominal. The classic misuse is letting the unscored queue itself become a de facto ranking ("it's near the top, so it must be good"). The guarding discipline is a hard, scheduled drain-and-score commitment with an owner, and keeping the deferred-criteria register loud enough that everyone can see scoring is coming, not cancelled.
How it implements the components¶
suspended_criteria_register— the queue names the scoring criteria that are deferred and keeps them visible, so suspension is explicitly temporary rather than a quiet cancellation.idea_capture_buffer— the Unscored queue is the holding buffer itself, preserving raw ideas intact until the batch pass.learning_and_trace_record— the scored queue persists as a record of what was proposed and how it fared, reusable when ideas resurface.
It does not implement criteria_reintroduction_sequence — that belongs to its nearest twin, Two-Pass Evaluation, which stages *which criteria return in what order, whereas the queue applies all deferred criteria together in one batch. And it separates capture from scoring across time, not across people — separation-by-person is Separate Generator and Evaluator Roles.*
Related¶
- Instantiates: Evaluation Criteria Suspension During Divergence — supplies the workflow that holds capture and scoring apart in time.
- Sibling mechanisms: No-Critique Rule · Yes-And Extension · Wild-Card Quota · Morphological Matrix · Two-Pass Evaluation · Separate Generator and Evaluator Roles · Timeboxed Blue-Sky Sprint
Editorial Notes¶
Form Classification¶
Form family: Protocol, Workflow & Routine
Rationale: Deferred Scoring Queue operates as a repeatable ordered procedure or handoff sequence that coordinates action because it routes every raw idea into a holding queue that is scored later in one batch, so capture is never interrupted by evaluation.
Independent corroboration: The frozen evidence defines Deferred Scoring Queue as 'Routes every raw idea into a holding queue that is scored later in one batch, so capture is never interrupted by evaluation', so its operative form is Protocol, Workflow & Routine.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Organizational & Management Science
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Structured group-decision methods such as nominal group technique cohered temporal separation of idea capture from later evaluation and ranking.
Related originating lineages:
- Psychology — Creativity research supplied evidence that immediate evaluation inhibits divergent generation.
Review resolution: Structured group-decision methods such as nominal group technique cohered temporal separation of idea capture from later evaluation and ranking. The retained alternate lineages materially shaped the mechanism's form.
Attribution caveat: The persistent queue is a workflow implementation of an established staged ideation method.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; medium confidence.
Notes¶
[n1] The nominal group technique (Delbecq and Van de Ven) structures group decision-making into distinct stages: participants generate ideas silently and independently first, and only afterward does the group discuss, evaluate, and rank them — deliberately preventing evaluation from contaminating generation. ↩