--- name: skill-launch-campaign type: skill license: CC BY-NC-SA 4.0 description: >- Reach's procedure for a product drop, a store launch or a relaunch — the one campaign shape a 0→1 retail brand runs repeatedly. Covers the drop arc (tease, waitlist or early access, launch, follow-through, post-launch review) with timings and the channel each phase uses; scarcity handled on-voice (real dated availability and limited runs, never manufactured urgency, per the venture's convention-brand); the asset list per phase handed to Pixel as a skill-creative-brief, copy via skill-content-studio, email via skill-newsletter and skill-lifecycle; the relaunch variant for a venture with an existing list and an old brand to retire (BARA from madebyneud); and KPI targets set before launch and reported after through skill-campaign-report. Use when a venture plans a drop, a launch or a relaunch, or when playbook-store-launch reaches its marketing leg. Anything that spends money or goes public is staged for the operator. Surfaced by APP-1080. --- # launch-campaign Reach's campaign skill for the moment a retail brand puts something new in front of people: a dated drop of a limited line, a store launch, or a relaunch under a new brand. BARA's relaunch and Keroma's launch are both live and both planned by hand; this is the repeatable shape. It plans against the retail calendar from `skill-retail-strategy` and the campaign calendar from `skill-gtm-plan`, uses the channel model in `convention-retail` to pick each phase's channel, and takes its scarcity rules from the venture's discount posture (`skill-retail-strategy` Pass 4, `convention-brand`). It plans and briefs; production is `skill-content-studio`'s and Pixel's, execution is `skill-channel-ops`'s, and `playbook-store-launch` holds the whole store-readiness checklist of which this campaign is the marketing leg. ## Trigger - A limited or maker-made line has a date (`convention-retail`, maker-made small batch: a launch is a dated drop). - A store is launching (Keroma, the first run of `playbook-store-launch`) or relaunching under a new brand (BARA from madebyneud, the second). - `skill-gtm-plan`'s calendar names a drop slot, or the operator asks for a launch plan. ## Behaviour Six steps: set the target, plan the arc, brief the assets, stage the phases, run the follow-through, review. ### Step 1 — the target, before anything is made State what the launch is for in `convention-retail` vocabulary and set the KPI targets, each against the baseline from `convention-shopify`'s queries: units or sell-through of the dropped line, revenue in the launch window, new customers and CAC by channel, list growth from the waitlist, and email revenue share. Mark each known or assumed; a first launch is mostly assumed and says so. A target set after the launch is not a target. ### Step 2 — the drop arc Five phases. The working timings below suit a 0→1 store with a fortnightly newsletter; the Owner can strike them. | Phase | When | Job | Channels | | -- | -- | -- | -- | | **Tease** | T−14 to T−8 days | Make the thing legible and wanted: the maker, the material, the ritual, the number being made. No price yet unless the brand always shows it. | Organic social (Instagram, Pinterest), a newsletter story, the homepage current-story slot. | | **Waitlist / early access** | T−7 to T−1 | Turn interest into a list: a signup for first access, with the promise stated plainly (the date, the time, the run size). Early access for existing customers is the one legitimate advantage a relaunch can offer. | Storefront signup, newsletter, the `skill-lifecycle` welcome flow, a product page in draft. | | **Launch** | T | Sell: the product live, the email out, the post up, within the same hour. State availability as fact — twelve made, shipping from the fourth — and stop when it sells out. | Email to the waitlist first, then the full list; social; the PDP live with the full imagery set. | | **Follow-through** | T+1 to T+14 | Keep the promise: dispatch comms in the voice, the maker story in full, the unboxing, the first reviews requested at arrival. | Order comms (`skill-lifecycle`), social, one newsletter issue. | | **Post-launch review** | T+21 to T+28 | Read the result against Step 1 and decide what the next drop changes. | `skill-campaign-report`. | A drop with no tease sells only to people who happened to be looking; a drop with no follow-through sells once. ### Step 3 — assets and copy Per phase, list every asset and every piece of copy, then brief: - **Pixel** — one `skill-creative-brief` per launch carrying the full asset list by phase: the PDP imagery minimum set (`standard-storefront`), social crops per `skill-channel-ops`'s channel table, the newsletter template slots, any packaging or insert change (`skill-packaging-unboxing`), the homepage current-story treatment (`skill-storefront-merchandising`). Against the venture's `convention-brand`; a marked brand gap is named in the brief, not filled. - **`skill-content-studio`** — the copy per phase and channel, in the voice, from the product facts in the store: the tease story, the waitlist promise, the launch email and posts, the PDP copy, the follow-through story. - **`skill-newsletter`** — the tease and follow-through issues slotted on the editorial calendar; **`skill-lifecycle`** — the waitlist welcome and the order comms. - **`skill-seo-discovery`** — the PDP and collection search surface ready before the tease, so the product is findable the moment it is searched for. ### Step 4 — stage the phases Hand each phase to `skill-channel-ops` as a set of dated actions, each mapped to its slot and KPI. Everything is staged: the PDP unpublished, the email scheduled but unsent, the posts queued, any code or price set but inactive. The operator releases the launch; the run reports what will happen when the release is made, in one line per action. ### Step 5 — follow-through Watch the launch window daily through the store's live figures: units, sessions by source, conversion, the waitlist's conversion. Where a line sells out, the PDP states it with the next date if one exists (`standard-storefront`); where it does not, the follow-through carries the story further rather than reaching for a discount. Delays are told before they are asked about (`convention-retail` step 5). ### Step 6 — review At T+21 to T+28, `skill-campaign-report` reads every Step 1 target against actual, attributes what the data supports, and states what the next drop changes: the run size, the tease length, the channel that earned its place, the one that did not. The review is filed to the venture's project and fed to `skill-gtm-plan` and `skill-retail-strategy`. ## Scarcity, on-voice Scarcity is stated, never staged. A dated drop, a limited run with its number, an end date that is true, early access for existing customers — all on-voice for BARA (`bara.voice-reference`, approved 2026-08-10, via `convention-brand`) and the default for any premium-craft venture here. Countdown timers, "only 3 left" that is not true, evergreen sales, a launch discount that becomes the price — off-voice and excluded. Where a venture's discount posture is unset (Keroma at the time of writing), the plan proposes one, marks it assumed, and never defaults to urgency. ## The relaunch variant A relaunch has what a launch lacks — an existing customer list and an old brand — and each changes the plan: - **The list is the launch.** Existing customers get the story first and early access; the tease phase tells them what is changing and what is not (the makers, the products, the care), so the relaunch reads as a chapter, not a replacement. - **Retire the old brand deliberately.** One announcement, in both names, with the reason; redirects from the old domain and handles (Forge, via `skill-seo-discovery`); the old name kept on the store's legal entity and order comms until the operator confirms otherwise (`convention-brand`, BARA instance). Search equity moves with the redirects, not with a fresh domain and hope. - **Two baselines.** Report the relaunch against the old brand's ninety days, and start the new brand's baseline at T so the next campaign has one. - **Do not launch the brand and a new line on the same day** — working rule — unless the operator chooses to; one story at a time reads clearly. ## Guardrails - Everything that spends money, sets a price, issues a code, sends to a full list, publishes or goes live is staged and released by the operator (`convention-shopify`, `agent-reach`). - No manufactured urgency, ever; availability is stated as fact and in the voice. - KPI targets are set before launch and reported after; a launch without a Step 1 target is not run. - Product facts, dates and run sizes come from the operator and the store; never invented (`convention-ticket`). - Plans and briefs; produces no asset and no copy, and operates no channel itself. - Reads the venture's `convention-brand` through the voice reference; never lets one venture's launch register bleed into another's. - Propose-only to canon. ## Setup Run on demand by Reach when a drop, launch or relaunch is on the calendar, and from `playbook-store-launch`'s marketing leg. Connectors: Shopify (read), Linear. Fed by `skill-retail-strategy`, `skill-gtm-plan`, `skill-audience-icp`; feeds `skill-creative-brief`, `skill-content-studio`, `skill-newsletter`, `skill-lifecycle`, `skill-seo-discovery`, `skill-channel-ops`, `skill-campaign-report`. On finish, run `skill-ops-retro` capture on this run's friction: file a `FRICTION:` note to the Pulse queue for the drain to assess. Owned by Reach (R3-ACH), run for the venture's operator.