// execute
Product Launch
Use when the user is planning a launch - a new product, a new market, a major feature, a pricing change - or is unsure how big a launch something deserves. Also use when the user mentions a launch plan, launch date, go-to-market for a release, launch tiers, go/no-go, launch checklist, "should we announce this", or a launch that is slipping. Tiers the launch first, scores readiness on the same twelve-item gate as the AAJ Launch Readiness Planner, and builds the dated work-back plan and launch-day go/no-go.
npx skills add sarojkjha/aaj-marketing-skills --skill product-launchClaude Code
/plugin marketplace add sarojkjha/aaj-marketing-skills/plugin install aaj-marketing-skills@aajOpenAI Codex
codex plugin marketplace add sarojkjha/aaj-marketing-skillscodex plugin add aaj-marketing-skills@aajClaude or ChatGPT — no install
Add this as a custom connector:
https://aaj-engines-mcp.sarojkjha10.workers.dev/mcp// when to use it
Use when the user is planning a launch - a new product, a new market, a major feature, a pricing change - or is unsure how big a launch something deserves. Also use when the user mentions a launch plan, launch date, go-to-market for a release, launch tiers, go/no-go, launch checklist, "should we announce this", or a launch that is slipping. Tiers the launch first, scores readiness on the same twelve-item gate as the AAJ Launch Readiness Planner, and builds the dated work-back plan and launch-day go/no-go.
// what you give
What is being launched and to whom, whether it changes pricing or how customers work, a 0-2 score for each of the twelve readiness items, and the launch date if one is set
// what you get
The launch tier with its reasons, a readiness score and verdict with the gaps ranked, a dated work-back plan filtered to the tier, and the launch-day go/no-go gates
Engine spec
// decision it supports
How big a launch does this change deserve, is it ready for a date, and what has to happen on which day?
// inputs
What is being launched and to whom, whether it changes pricing or how customers work, a 0-2 score for each of the twelve readiness items, and the launch date if one is set
// output
The launch tier with its reasons, a readiness score and verdict with the gaps ranked, a dated work-back plan filtered to the tier, and the launch-day go/no-go gates
// acts or analyzes
Analyzes onlyMakes no network calls, writes no files and changes nothing in your tools.
// where your information goes
| How you run it | Where your inputs go |
|---|---|
| Skill installed in your agent, or the local MCP server | The engine runs on your machine and AAJ receives nothing. Your AI agent still sends what you give it to its model provider (Anthropic, OpenAI and so on), under your own account's terms. |
| Hosted MCP server | Your agent's model provider, as above, plus AAJ's server on Cloudflare, which receives the engine inputs over HTTPS, runs them in memory and keeps no copy. Cloudflare keeps error logs (no inputs) for up to 7 days. |
No engine needs customer names, emails or other personal data; leave them out.
// limitations
- Readiness is the team's own 0-2 scores on twelve items, and a missing item scores 0. The engine cannot verify them. The 85% and 60% thresholds are AAJ's, not an industry standard.
- The tier comes from the launch fields you choose. It flags a mismatch with what the team intends only if you give plannedScale.
- Past-due tasks are marked "due date passed, not marked done" unless you list them in "done". The plan is not linked to the readiness gate scores.
- The four launch-day go/no-go gates are listed, not checked. Someone has to confirm each one on the morning.
// review before acting
// setup
npx skills add sarojkjha/aaj-marketing-skills --skill product-launchhttps://aaj-engines-mcp.sarojkjha10.workers.dev/mcpAgents listed: Claude Code, Cursor, OpenAI Codex, Windsurf, Cline. Tested and recorded: Claude Code.
Product Launch
Launches fail in two opposite ways. Some get a date before they are ready, and the work that was not done gets squeezed into the last week. Others are not launched at all: a real change ships as a line in the changelog and nobody notices.
Both come from skipping the same first question - how big a launch does this deserve? - and then setting a date before checking whether the work behind it exists.
When to use
The user is planning a launch, choosing a launch date, deciding whether something is worth announcing, or has a launch that keeps slipping.
Do not use this to make every release into an event. A launch is a claim on your audience's attention, and spending it on small changes trains people to ignore the one that matters.
Before you start
- Read the brand/product context (
.agents/product-marketing.md) for positioning and audience. - Say what is being launched in one line, and who it is for.
- Find out whether it changes pricing or how existing customers work. Either one means customers have to hear it from you, whatever else happens.
- Score the twelve readiness items honestly. 0 not started, 1 partly, 2 done. "Partly" is the honest answer more often than teams want it to be.
Method
Tier first. The tier decides how much launch the change gets.
| Tier | When | What it gets |
|---|---|---|
| 1 - full launch | A new product or a new market, or a major feature for the whole market that opens a new revenue line or changes pricing | An external announcement with a date, the full work-back plan, and the go/no-go gates |
| 2 - customer launch | A major feature, any pricing change, or a change every customer has to adapt to | Email, in-app, a post and a sales brief to customers and the relevant segment. The Must and Should tasks, and the gates |
| 3 - release note | Everything else | A changelog entry, an in-app note where it is used, and one adoption metric. No date, no gate |
Then the gate. Twelve items, two per dimension - positioning, product, audience, assets, team, measurement - each scored 0, 1 or 2. 85% or above is ready to set a date. 60% to 84% is close. (These are AAJ's thresholds, not an industry standard.) Below 60% is the building phase: this is not a launch-planning problem yet, and setting a date would move work that has not been done into a week that does not exist.
These are the same items and thresholds as the AAJ Launch Readiness Planner and the Launch Plan Checklist, so a founder who fills in the checklist and an agent running this skill reach the same verdict.
Then the plan. Working back from the date: foundation at T-30, assets from T-20, pre-launch from T-10, launch day, amplification after it, and a retro at T+7. Tier 1 gets every task; Tier 2 drops the ones marked Nice. Any Must task already past its due date is reported, because a launch date with overdue foundations is the building phase with a deadline attached.
On the day, four go/no-go gates. The feature works end to end, tracking is firing, the landing page works on desktop and mobile, and the owners are online and know the runbook. Any one of them false is a no-go.
Workflow
Step 1 - Tier it. Run the engine with the launch description. If the team's planned scale does not match the tier, the engine says so; resolve that before anything else.
Step 2 - Score the gate. Twelve scores. The engine ranks the gaps, zeros before partials.
Step 3 - Decide the date. Only in the "ready to set a date" band. In "close", fix the zeros first. In "building", do not set one.
Step 4 - Build the plan. Give the engine the date. It returns every task with its due date for the tier and flags anything already overdue.
Step 5 - Run the day. Check the four gates on the morning. Then the T+7 retro against the goal metrics set in the gate.
Run the tool
node resources/launch.js --demo
run(config) returns the tier with its reasons, the readiness score and verdict with the gaps, the
dated plan, and the go/no-go gates.
Use the engine rather than an ad hoc checklist. The same gate every time is what stops enthusiasm reading as readiness.
Present the result
- The tier and why, and any mismatch with what the team planned.
- The readiness verdict - ready, close or building - with the score.
- The gaps, zeros first, each with why it matters.
- The date decision: set it, fix the zeros first, or do not set one yet.
- The plan, with anything overdue called out.
- The go/no-go gates for the morning of the launch.
Guardrails & common mistakes
- Do not set a date in the building phase. Every launch that slips was dated before the gate was clear.
- Do not launch a Tier 3 change as an event. Ship it with a note and move on.
- Do not launch a Tier 1 change quietly. A new product rarely gets a second first impression.
- A pricing change is always at least Tier 2. Customers must hear it from you before they see it on an invoice.
- Set the goal metrics before the date, not after. Without a target, a launch cannot be told apart from noise, and the retro learns nothing.
- Name one owner. Diffuse ownership drops balls exactly when timing matters.
Related AAJ resources
- The same gate as a free tool: https://aajconsult.com/tools/launch-readiness-planner
- The spreadsheet version: https://aajconsult.com/resources/launch-plan-checklist
- The full method: https://aajconsult.com/playbooks/product-launch-playbook
Related skills
positioning-statement— the story every launch asset inheritsaaj-copywriting— the launch page and emailspr-and-earned-media— the external half of a Tier 1 launch
// faq
// install
npx skills add sarojkjha/aaj-marketing-skills --skill product-launch// worked example
Get the worked example for Product Launch
One email. The filled-in config, the output, and how to read it.
// human in the loop
Need a human in the loop?
AAJ builds the strategies, tools, and playbooks these skills are made of — with engagements for growth-stage and enterprise teams.