Work Prioritisation

How do you decide what to build next — and when to stop?

Ship in fixed cycles. Test assumptions at each stage. Increase investment only as risk decreases. Fixed time, variable scope. Expect to be done.

Shaping Up

Disagree then commit on quarterly objectives
Two cycles per quarter's strategic mission
Identify the hardest challenge and attack that first
Six-week mark: tweak mission, review tactics

Stage Progression

Stage	Time	Question	Output
Napkin	Minutes	Does this have obvious fatal flaws?	Core assumptions listed
Mock-up	1-2 days	Does this solve the customer's problem?	Single use-case prototype
Market	3-4 days	How big can this be? Why now?	Defensible rhetoric
Build	Weeks	Are there unseen fatal flaws?	Working prototype, cost estimates
Demand	Ongoing	Does retention exceed benchmarks?	Go-to-market strategy

The Algorithm

Deterministic. Any agent can run node scripts/prioritise-prds.mjs --diff and get the same table order. Scores (judgment) live in PRD frontmatter. Math lives in the script.

INPUT:
  prds[]        <- read from /phygital-mycelium/index.md Active table
  commissioning <- read from /phygital-mycelium/commissioning/index.md

FOR EACH prd IN prds:

  STEP 1 - SCORE PRIORITY (what deserves attention)
    Read the PRD. Score each dimension 1-5 using rubrics below.
    priority = (pain * demand * edge * trend * conversion) ^ (1/5)

  STEP 2 - SCORE READINESS (can we build it now)
    Read commissioning status. Score each dimension 1-5 using rubrics below.
    readiness = (0.20 * principles) + (0.25 * performance)
              + (0.20 * platform) + (0.15 * protocols) + (0.20 * players)

  STEP 3 - CHECK HARD GATES
    gate_pass = ALL of:
      [] Benchmarks defined (thresholds exist before build)
      [] Feature/Function/Outcome table exists (commissioning checklist)
      [] Named builder != named commissioner
      [] Kill date + kill criteria written (if build started)
      [] Baseline measurement exists (if claiming improvement)
    IF NOT gate_pass: status = HOLD

  STEP 4 - COMPUTE COMBINED SCORE
    combined = priority * readiness / 5

  STEP 5 - CHECK DEPENDENCIES
    blocked_by[] <- read PRD dependency declarations
    IF any blocked_by[].status == BLOCKED: promote blocker above this PRD

  STEP 6 - CHECK KILL DATE
    IF kill_date exists AND kill_date - today < 30 days:
      Apply time_pressure bonus = +0.5 to combined

SORT prds BY combined DESC, applying:
  1. HOLD items sink to bottom of their maturity tier
  2. Blockers promote above the items they block
  3. Kill date items get +0.5 bonus

OUTPUT:
  Sorted table -> becomes Active table order in /phygital-mycelium/index.md

This follows the Tight Five sequence for business execution:

Principles: is value transformation and distribution explicit?
Performance: are benchmarks and drift triggers defined?
Platform: can we ship on existing operational base?
Protocols: does this create reusable know-how?
Players: is demand real and are the right people onboard?

Priority Rubrics

Priority: (Pain x Demand x Edge x Trend x Conversion) ^ (1/5)

Score	Pain	Demand	Edge	Trend	Conversion
5	System blocked or user locked out	Every downstream capability depends on this	No competitor has this; patentable	Category-defining shift happening now	Paying customers exist
4	>10hrs/week wasted, validated by user	JTBD validated by interviews or usage data	Competitors exist but miss key insight	Strong tailwind, 2-3yr window	Demo-ready, prospects engaged
3	Pain acknowledged but workarounds exist	Internal hypothesis with supporting signals	Some differentiation, not unique	Growing category, crowded	Prototype exists, no revenue signal
2	Mild friction, low urgency	Assumed need, no evidence	Commodity with cosmetic differences	Stable or slow growth	Spec only, no build
1	No evidence of pain	No evidence of demand	No differentiation	Declining or unknown	No evidence anyone would use this

Readiness Rubrics

Readiness: 0.20(Principles) + 0.25(Performance) + 0.20(Platform) + 0.15(Protocols) + 0.20(Players)

Performance is highest weight because unmeasured work drifts. Principles/Platform/Players are equal core constraints. Protocols still matter, but after value, measurement, and demand are clear.

Score	Principles	Performance	Platform	Protocols	Players
5	Value loop + margin logic explicit	Kill criteria + benchmarks + review cadence all defined	Ships on existing stack with zero new infra	Creates reusable SOP adopted by 2+ capabilities	5+ real interviews or equivalent evidence
4	Value loop clear, margin estimated	Kill criteria + benchmarks defined, review TBD	Ships with minor additions to existing stack	Creates reusable SOP, not yet adopted	2-4 interviews or strong usage signals
3	Value proposition stated, loop unclear	Some benchmarks defined, gaps remain	Needs 1 new service or integration	Follows existing SOP, doesn't create new	Internal user validated
2	Problem stated, no value loop	Kill criteria exist, no benchmarks	Needs multiple new services	No SOP connection	Hypothesis only, no validation
1	Problem unclear	No measurement defined	Needs net-new platform	No protocol value	No evidence of demand

Hard Gates

If any gate fails, priority is HOLD regardless of score.

Gate	Required Evidence	Result if Missing
Benchmark Gate	Thresholds defined from Benchmark Standards	HOLD
Commissioning Gate	Feature/Function/Outcome table with pass/fail checks	HOLD
Owner Gate	Named builder + named commissioner (different people)	HOLD
Kill Gate	Explicit kill date and kill criteria	HOLD
Data Gate	Baseline measurement exists before change	HOLD

Decision Bands

Score Band	Decision	Action
>= 4.0	Ship Now	Allocate current cycle
3.0-3.9	Prepare	Close missing dependencies, then rescore
2.0-2.9	Incubate	Keep in backlog, gather evidence
< 2.0	Stop	Drop or rewrite problem framing

Dependencies

Identity & Access <-- BLOCKS EVERYTHING
  |
  +-- Agent Platform <-- blocks all agent instances
  |     +-- Sales Dev Agent
  |     +-- Content Amplifier
  |     +-- [future agents]
  |
  +-- Sales CRM & RFP <-- blocks sales chain
  |     -> Sales Dev Agent -> Sales Process Opt -> Trust Commerce
  |
  +-- ETL Data Tool <-- blocks data access
        -> Data Interface

Data flow between sales chain:

From	Provides	To
CRM	contacts, deals, activities, pipeline state	Dev Agent
Dev Agent	scored prospects, outreach history, channel metrics	SPO
SPO	qualification proof, satisfaction outcomes, relationship data	Trust Commerce
Trust Commerce	on-chain reputation, referral conversion	--

Duplication Check

FOR EACH pair (prd_a, prd_b) IN prds:
  overlap = count(shared components in composition inventory)
  total   = max(components_a, components_b)
  IF overlap / total > 0.50:
    CHECK: do they declare Parent/Component/Peer relationship?
    CHECK: do shared components use different verbs? (own/use/compose/orchestrate)
    IF no relationship declared: FLAG as hierarchy signal
    IF same verb for same component: FLAG as duplication

Current Run (2026-02-24)

Sales trio overlap: 9 of 13 components shared (69%). Not duplication — three zoom levels with distinct verbs:

Component	CRM	SDA	SPO
CRM Contacts + Deals	own	use	use
CRM Activities + Tasks	own	use	use
Sales Forecasting algo	wire	compose	orchestrate
Explore-Exploit algo	wire	compose	orchestrate
SPCL Scoring algo	wire	compose	orchestrate
Compound Rate Tracking	wire	compose	orchestrate
Email Adapter	--	own	orchestrate
LinkedIn Adapter	--	own	orchestrate
Document Intelligence	own	compose	orchestrate

Flags:

Issue	PRDs	Fix
Identical switch trigger (verbatim)	SDA + SPO	SPO trigger should derive from SDA output metrics, not copy SDA's input trigger
Ambiguous workflow ownership	Agent Platform + Content Amplifier	Content Amplifier owns marketing workflow wiring; Agent Platform owns orchestration infra
Unidirectional relationship declarations	Sales chain	Each PRD should declare upstream AND downstream

Result: No duplication. Hierarchy correctly structured.

Drift Control

Run this every week for active PRDs:

Recompute Priority Score
Evaluate benchmark state: Pass / Warn / Fail
Trigger action:
- Pass: continue or promote standard
- Warn: corrective loop and retest in 7 days
- Fail: freeze rollout, rollback, or kill
Log the decision and evidence in the commissioning dashboard

Rescore Triggers

Trigger	Action
PRD moves maturity level (L0->L1, etc.)	Rescore that PRD
Kill date within 14 days	Force rescore + apply +0.5 bonus
Blocker clears	Rescore everything it blocked
Weekly (active PRDs only)	Rescore + verify index order
Combined score drops below 0.5	Candidate for kill
Kill date passes with zero north star progress	Kill

Frontmatter Schema

Every PRD stores these fields. The script reads them; the score-prds skill writes them.

# Priority (judgment - assigned by LLM/human using calibration examples)
priority_pain: 0          # 1-5
priority_demand: 0        # 1-5
priority_edge: 0          # 1-5
priority_trend: 0         # 1-5
priority_conversion: 0    # 1-5

# Readiness (judgment - assigned by LLM/human using calibration examples)
readiness_principles: 0   # 1-5
readiness_performance: 0  # 1-5
readiness_platform: 0     # 1-5
readiness_protocols: 0    # 1-5
readiness_players: 0      # 1-5

# Metadata
kill_date: ""             # YYYY-MM-DD if applicable
blocked_by: ""            # comma-separated prd slugs
gate_status: ""           # BLOCKED if hard-blocked
gate_reason: ""           # why blocked
last_scored: ""           # YYYY-MM-DD

Computed Outputs

Output	Formula
`priority`	(pain x demand x edge x trend x conversion) ^ (1/5)
`readiness`	0.20(principles) + 0.25(performance) + 0.20(platform) + 0.15(protocols) + 0.20(players)
`combined`	priority x readiness / 5
`kill_bonus`	+0.5 if kill_date within 30 days
`dep_bonus`	+0.1 per PRD that declares blocked_by this PRD
`final`	combined + kill_bonus + dep_bonus
Gate checks	BLOCKED, NO PLAYERS (players=1), KILL DATE PASSED
Sort	final descending, scored before unscored

node scripts/prioritise-prds.mjs        # ranked table
node scripts/prioritise-prds.mjs --diff  # compare against index order
node scripts/prioritise-prds.mjs --json  # machine-readable output

Success Metrics

Type	What to Measure
Quantitative	Retention, unit economics, CAC/LTV
Qualitative	How users feel if product disappeared
Ethical	Does it serve genuine needs?

Stress plus rest creates growth. Judge challenges to manage oscillations for peak performance over time.

PRD with AI

A Product Requirements Document translates the job to be done into what gets built. When working with AI:

Share the PRD and tackle one step at a time — don't overload context
Test each feature before moving to the next
Use TDD — let the AI test its own code
Ask AI to explain the architecture so you understand what was built

Be less specific in some areas to allow creative solutions. Start over when things go wrong rather than patching complexity.

Context

Phygital Mycelium — The capability catalogue (table order = build order)
Commissioning Dashboard — What's specified, built, proven
Commissioning Protocol — L0-L4 maturity model, gates, process
Jobs To Be Done — Discover what to build
Validate Demand — Prove people will pay
Priorities — The priority algorithm (Pain x Demand x Edge x Trend x Conversion)
Tight Five Platform — The 5P readiness framework
Benchmark Standards — Trigger-based benchmark protocol
Business Development — From idea to plan
Business Instruments — Control loops at each stage
Manufacturing — Production systems thinking
Flow Engineering — Mapping value streams

Questions

How do you decide what to build next — and when to stop?

When the algorithm says "Ship Now" but your gut says "Not ready" — which do you trust, and what signal would change your mind?
Is the 5P readiness weighting right — should Performance really outweigh Principles?
What happens when two PRDs have identical scores but competing dependencies?
At what point does drift control become overhead that slows down the thing it's meant to protect?

Shaping Up​

Stage Progression​

The Algorithm​

Priority Rubrics​

Readiness Rubrics​

Hard Gates​

Decision Bands​

Dependencies​

Duplication Check​

Current Run (2026-02-24)​

Drift Control​

Rescore Triggers​

Frontmatter Schema​

Computed Outputs​

Success Metrics​

PRD with AI​

Context​

Links​

Questions​