---
name: "skill-newsletter"
type: skill
license: CC BY-NC-SA 4.0
description: Reach's procedure for building and running a retail venture's editorial newsletter — the send people open because it reads like a journal, not a store — beside the automated flows in skill-lifecycle. Covers strategy (purpose, segment, cadence by venture stage, the signup promise, the editorial formats), the repeatable issue anatomy designed with Pixel, the editorial calendar pinned to skill-retail-strategy's retail year and skill-gtm-plan, list growth and the welcome-flow handoff, tooling (Shopify Email first, the criteria for moving up), and measurement through skill-campaign-report. Use when a venture starts, restarts or reviews its customer newsletter, when an issue is due, or when list, cadence or tooling is in question. A venture's customer newsletter only — the Mr Pritchard Substack is convention-mrp-voice's and shares structure, never voice. Drafts via skill-content-studio; Aled reviews; a full-list send is a human action. Surfaced by APP-1080.
---

# newsletter

Reach's editorial send. `skill-lifecycle` builds the automated flows that carry the mechanics of repeat purchase; this skill runs the one send a customer chooses to open between purchases, because it is worth reading. For a story-first brand the newsletter is the natural surface of the voice — BARA's voice reference asks for "a digital journal, not an ecommerce store", and the newsletter is where that register lives at full length. It composes `convention-retail` (step 8 of the touchpoint chain, email revenue share), `convention-brand` and the venture's voice reference, `skill-retail-strategy` (the retail calendar) and `skill-gtm-plan` (the campaign calendar), and restates none of them. Drafts through `skill-content-studio`; the visual template through `skill-creative-brief`.

## Trigger

- A venture starts or restarts a customer newsletter, or has a list and no editorial send.
- An issue is due on the editorial calendar, or a drop, launch or seasonal moment needs an issue (`skill-launch-campaign` hands the launch issue here).
- The operator asks whether the cadence, the list, or the tooling is right.
- A `skill-campaign-report` read shows email revenue share, open or unsubscribe moving against target.

Retail ventures: BARA first, Keroma when it has a list. Not the Mr Pritchard Substack, which is personal editorial under `convention-mrp-voice` and never shares a voice with a venture send.

## Behaviour

Five passes. Passes 1 and 2 run once and are refreshed on review; 3 to 5 run per issue.

### Pass 1 — strategy

Set, and file to the venture's project, one page that states:

- **Purpose** — the one job the newsletter does for this venture, in `convention-retail` terms: it moves repeat rate and email revenue share, and it keeps the customer close between purchases. It is not a discount channel.
- **Segment** — who receives it. Default: every consenting customer and subscriber, one list, one send; segmentation (buyers vs subscribers, region, language) is added only once the list or the data justifies a second version.
- **Cadence by stage** — working rule, the Owner can strike it: at 0→1, **fortnightly**; once the store has ninety days of orders and a content bank three issues deep, **weekly**; never more often than the content bank can sustain. A missed issue beats a thin one.
- **The promise on the signup form** — one sentence stating what arrives and how often, in the venture's voice, and true. The promise governs every issue after it.
- **Editorial formats** — the five the venture draws from: maker story; ritual or how-to; drop announcement (a real dated drop, per the venture's discount posture in `skill-retail-strategy` Pass 4); behind the scenes; seasonal. Each format is named with an example subject line so the content bank has shape.

Mark each element known or assumed, as `skill-retail-strategy` does. A venture with no list yet has almost everything assumed.

### Pass 2 — the issue anatomy and template

One repeatable anatomy, so an issue is a fill, not a design: **opening** (a few lines in the voice, one idea); **one story** (the issue's format, at its natural length); **one product moment** (a single product or set, in use, with its price and a plain link); **one call to action** (one, never a row of buttons); a **sign-off** (for BARA, the Welsh sign-off is Aled's and is judged in execution, per `convention-brand`).

Working rules the Owner can strike: read time **two to four minutes**; image-to-text roughly **one image per story beat**, never an image-only issue; mobile-first at a single column; alt text on every image per `convention-shopify`'s frame-not-product rule; plain-text fallback carries the same story.

Hand the template to Pixel as a `skill-creative-brief` against the venture's `convention-brand` instance — palette, type, imagery direction — and record where the template lives. Where a brand field is a marked gap (BARA type and logo, at the time of writing), the brief says so and the template ships with the gap named, not filled.

### Pass 3 — the editorial calendar and the content bank

Pin issues to two spines: the retail calendar from `skill-retail-strategy` Pass 5 (gifting peaks, drop cadence, restock rhythm, quiet windows) and the campaign calendar from `skill-gtm-plan`. Each issue slot names its date, format, story, product moment, action and the KPI it moves. A launch or drop issue is planned by `skill-launch-campaign` and slotted here; the flows around it stay `skill-lifecycle`'s.

Keep a **content bank** in the venture's project: story candidates captured as they occur (a maker visit, a glaze test, a customer's photo, a seasonal ritual), each with its format and the evidence it rests on. Working rule: the bank holds **at least three issues ahead**, so an issue is never written from a blank page. A thin bank is reported before it is a missed issue.

Raw material for a maker story is captured, never recalled. A maker-story candidate rests on a real detail — a quote, a process decision, a named material choice — and that detail is captured from the maker in a session booked for it, briefed as its own ticket to the venture's operator and kept distinct from the photography brief: a photography ticket returns frames, not sentences, and `skill-packaging-unboxing` and `skill-content-truth-capture` both reach the work rather than a maker who was never asked. The bank entry names its capture ticket as the evidence it rests on. Where the detail is absent the issue is outlined and left undrafted rather than filled from inference, and the gap is reported with the thin-bank report; `skill-content-studio` is briefed, never asked to supply it. (Worked case: MBN-325's BARA Christmas run needed a story-only journal issue with no product ask between two selling issues, and had no maker detail from Matt Jones to carry it — MBN-308 briefed in-use gift photography for the five ceramic SKUs and nothing briefed the words, so the issue was outlined and left undrafted per `bara.voice-reference`'s rule against inventing facts. MBN-325 and MBN-308 are reported from the note, not observed on the refine surface — APP-1885.)

### Pass 4 — the issue

Per issue: pull the slot, the bank entry and the product facts (from the store, never from memory); brief `skill-content-studio` with format, story, product moment, action and the voice reference; draft subject line and preview text as variants where a test is intended; assemble in the template; check the anatomy, the alt text, the links and the unsubscribe; file for Aled's review with the send date. The send goes to the operator staged, not sent.

### Pass 5 — measure

After each send and monthly, read through `skill-campaign-report`: open rate, click rate, revenue per send, unsubscribe rate, list growth, and email revenue share — each against baseline and target, in `convention-retail` vocabulary, attribution model stated. One read per month names the format that earned the most clicks and the one that earned the most unsubscribes; the calendar shifts accordingly.

## List growth

- **Placements** — the storefront signup (footer and a single considered prompt, per `standard-storefront`; never a modal on arrival), the order confirmation, and the packaging insert (`skill-packaging-unboxing`), each carrying the Pass 1 promise.
- **Welcome handoff** — a new subscriber enters `skill-lifecycle`'s welcome flow first; the newsletter follows at the next slot. The two never send on the same day.
- **Consent** — opt-in recorded per `convention-analytics`; the commercial-site posture applies. No bought lists, no scraped addresses, no pre-ticked boxes, ever.
- **Growth KPI** — net list growth per month against target, reported beside unsubscribe rate so growth is never read without churn.

## Tooling

**Shopify Email by default** — the list, consent and attribution stay in the store, and `convention-shopify`'s channel-mix query reads it without a second attribution model. Working rule for moving up, the Owner can strike it: propose Klaviyo when the venture needs segmentation or flow logic Shopify Email cannot express, or the list passes **2,500** contacts; propose a Substack-style tool only when the newsletter has become an editorial product in its own right with readers who are not customers. A move is a config action for the operator with the migration and attribution consequences stated, never made by a run.

## Guardrails

- Voice is the venture's, read from its voice reference through `convention-brand`; never the Mr Pritchard voice, never generic marketing. Off-voice is a failure even if it performs.
- Story-first. An issue with no story is a promotion, and belongs to `skill-launch-campaign` or `skill-lifecycle`, not here.
- No manufactured urgency, no evergreen discount, no countdown; a genuine dated drop is on-voice and stated plainly.
- Product facts, prices and availability come from the store; never invented (`skill-content-studio`).
- Sending to a full list is a human action at `convention-shopify`'s boundary — staged and handed over, every time.
- Consent and frequency are respected; a subscriber who did not opt in is never mailed.
- Proposes to canon and files to Linear; never changes tooling, lists or store data itself.

## Setup

Run on demand by Reach for the operator, or on the editorial calendar's cadence. Connectors: Shopify (read, plus staging in Shopify Email where wired), Linear. Feeds `skill-campaign-report`; fed by `skill-retail-strategy`, `skill-gtm-plan`, `skill-launch-campaign`. 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.
