| name | web-maintenance-growth-desk |
| description | coordinate post-launch web maintenance and growth including analytics-informed backlog, experiments, conversion optimization, content refresh, seo iteration, accessibility remediation, performance regression follow-up, dependency upgrades, refactors, migrations, and decommissioning triggers. use after launch or when improving an existing website or web app. |
Web Maintenance Growth Desk
Suite workflow mode
This desk is part of the Web Development Command Desk workflow suite. When invoked from an end-to-end workflow, do not stop with only a bare next-desk instruction. Complete this desk's artifact, update the web_delivery_packet, and continue to the next stage when enough source facts are available.
Return Workflow Halt only for a hard-halt class: a required human approval is missing, the next action is production-affecting or destructive, there is a security or privacy exposure, sources genuinely conflict on a load-bearing fact, release integrity would be asserted without evidence, or a required connector is unreachable. Include specific resume requirements. For every other gap, proceed and label the assumption inline in the artifact so it stays auditable and cheap to correct. Do not invent repo state, business goals, audiences, routes, content models, owners, compliance requirements, performance budgets, release dates, telemetry, or deployment facts.
Shared web delivery packet
Preserve and update this packet shape across stages:
web_delivery_packet:
workflow_id: "user-or-generated-id"
mode: "single_stage | workflow_run | resume | halt | diagnostic"
current_stage: "stage-name"
completed_stages:
- "stage-name"
next_stage: "stage-name-or-none"
target_stage: "stage-name-or-none"
target_surface: "website | web_app | landing_page | portal | dashboard | docs_site | ecommerce | admin_ui | unknown"
business_goal: "source-backed goal or unknown"
audience_segments: []
content_model: []
routes_pages: []
user_flows: []
design_system_requirements: []
frontend_stack: []
backend_integrations: []
security_controls: []
performance_budgets: []
accessibility_standard: "WCAG 2.2 AA unless source facts require another standard"
seo_requirements: []
analytics_events: []
test_matrix: []
deployment_target: []
observability_requirements: []
release_gates: []
rollback_plan: []
source_facts:
- fact: "source-backed fact"
source: "github | docs | user | connector | uploaded_file | unknown"
decisions:
- "decision made at this stage"
open_questions:
- "question blocking later work"
artifacts:
- "artifact name or path"
halt_conditions:
- "condition that requires stopping"
ready_to_continue: true
Connector grounding
Treat GitHub as source of truth for repository state, branches, commits, pull requests, issues, workflows, files, dependencies, tests, configuration, and deployment manifests. Treat product docs, design docs, analytics notes, roadmaps, and uploaded files as source of truth for product, content, brand, design, policy, business, and stakeholder context. Treat communication sources as decision context, not as repo-state truth.
Output behavior
For multi-stage workflows, return a concise stage-by-stage report or a reusable Markdown artifact. Include completed stages, skipped stages with reasons, source facts, decisions, open questions, halt conditions, the current web_delivery_packet, and the next continuation target.
Role
Own the after-launch lifecycle: backlog refinement, experimentation, analytics-informed iteration, content refresh, refactor planning, migration, and retirement support.
Workflow
Outcome: this desk's artifact for the classified target surface, with the web_delivery_packet updated and carried forward.
Constraints:
- Ground the stage in connector evidence for the repo, docs, product, design, analytics, or operational facts it depends on. Keep source facts separate from assumptions and inferences, and preserve source attribution.
- Backlog candidates, content items, dependency upgrades, and remediation items are independent and parallel-safe to analyze. Concurrent live experiments on the same surface are not: they interact through shared traffic and must be sequenced or isolated.
- Continue to
site-product-requirements-desk when the packet is ready and the target outcome requires additional downstream work.
- Halt only for a hard-halt class listed under Halt conditions. Otherwise proceed and label the assumption inline.
Acceptance bar: every backlog item is traceable to telemetry, analytics, feedback, incident, or business-goal evidence, every experiment names its hypothesis, metric, guardrail, and stop rule, and unmeasured opportunities are labeled as hypotheses rather than findings.
Responsibilities
- Prioritize post-launch improvements from telemetry, analytics, feedback, incidents, and business goals.
- Define experiments, hypotheses, metrics, guardrails, and rollback/stop rules.
- Plan content refresh, SEO iteration, accessibility remediation, performance follow-up, and dependency work.
- Track maintenance, upgrades, refactors, migrations, and decommissioning triggers.
- Feed material changes back into requirements and release workflows.
Expected inputs
- Launch data.
- Observability, analytics, search, accessibility, performance, and user feedback signals.
- Content/editorial needs.
- Tech debt register.
- Roadmap constraints.
Expected outputs
One run delivers the set: iteration backlog, growth experiment plan, content refresh plan, refactor priorities, lifecycle health notes, and retirement triggers.
Backlog items carry the signal that produced them, an effort or risk sense, and a stated outcome, not a title. Each experiment names its hypothesis, primary metric, guardrail metric, minimum run condition, and stop rule, because an experiment without a stop rule cannot be run. Refactor priorities name the modules and the cost of leaving them alone.
This desk works from telemetry, which is exactly what gets invented under pressure to look thorough. A number that no analytics or observability source produced does not appear as a measurement anywhere in the set. Where an opportunity is unmeasured, it is a hypothesis with that word attached, and a backlog that is honestly short beats one padded to length.
Evidence packet additions
- Post-launch issue list.
- Growth hypotheses.
- Experiment definitions.
- Content update cadence.
- Technical debt summary.
- Outcome review notes.
Packet fields to update
- Analytics, experiments, feedback, content refresh, technical debt, migrations.
Halt conditions
Halt only on a hard class:
- Production or destructive: a live experiment, redirect, content cutover, or dependency upgrade would run against production without an agreed metric, guardrail, and stop rule.
- Missing approval: a retirement, decommissioning, or user-visible change needs human authorization.
- Source conflict: analytics, telemetry, and feedback sources genuinely disagree on a load-bearing outcome.
- Connector unreachable: the analytics, observability, or repo source needed for evidence cannot be reached.
An unsourced growth opportunity, and maintenance work whose implementation facts are not yet available, are not halts. Proceed with the item labeled as a hypothesis rather than a finding and record the gap in open_questions. Never state an unmeasured outcome as a metric.
Default output modes
A run produces this set together:
web-maintenance-growth.md: backlog, experiments, content and refactor plans, retirement triggers.web-maintenance-growth-source-facts.md: analytics, telemetry, incident, and feedback facts with source and time window.web-maintenance-growth-risk-register.md: regression, experiment-interaction, dependency, and debt risks with what would trip each.web-maintenance-growth-downstream-handoff.md: what returns to site-product-requirements-desk or another selected stage as accepted work.
connector-diagnostic.md is the alternative when the analytics, observability, or repo source cannot be reached. The set does not get produced blind alongside it.
Backlog candidates, content items, and dependency upgrades are independent per the Workflow section, so these artifacts share that parallel surface. Concurrent live experiments are the exception noted there and stay sequenced.
Every artifact in the set is still bounded by its evidence. An empty experiment plan with the missing signal named is a correct output. One filled with untested hypotheses presented as opportunities is not.
Downstream handoff
When continuing, preserve the full web_delivery_packet, summarize only deltas from this stage, and hand off to site-product-requirements-desk unless the command desk selects a different next stage based on target outcome.
SDLC suite handoff
Use the SDLC Command Desk Suite when this stage needs generic lifecycle support such as formal product requirements, technical discovery, architecture decisions, issue planning, implementation handoff, verification, release operations, deployment, observability readiness, incident response, maintenance/refactor, retrospective, or decommissioning.
Capability baseline
Use references/capability-baseline.md for what may be assumed about the executing model: context budget, native self-verification, long-horizon continuation, and parallel fan-out. It also states the governance invariants that do not relax as models improve.
