email-marketing

SKILL.md

Email Marketing Expert Skill

When to Use This Skill

Invoke this skill for any task involving:

  • Planning or auditing an email marketing programme
  • Writing email copy, subject lines, CTAs, or multi-email sequences
  • Advising on list building (forms, popups, lead magnets, double opt-in)
  • Explaining or setting up email authentication (SPF, DKIM, DMARC)
  • Building segmentation strategies or explaining segment types
  • Designing automation flows (welcome sequences, drip campaigns, re-engagement)
  • Interpreting or explaining email metrics and analytics
  • Comparing or recommending ESP platforms
  • Advising on data collection, personalisation, or merge tags
  • Explaining transactional vs marketing email distinctions
  • Troubleshooting deliverability problems

Expert Principles: How to Think About Email

These are the practitioner-level stances that should inform all advice given. Do not give generic or hedged recommendations that contradict these.

  1. List quality beats list size, always. An engaged list of 500 outperforms a disengaged list of 10,000. Optimise for attracting the right subscriber.

  2. Email metrics are directionally useful, not precisely meaningful. Open rates especially. Never treat them as gospel. Never make major decisions based on small metric differences without proper statistical rigour.

  3. Start simple, then add complexity. Most email programmes are better served by well-executed basics than poorly-executed sophistication. Crawl, then walk, then run.

  4. Set a goal before you build anything. Every campaign, every automation, every sequence needs a clear, measurable objective. Without one, it's theatre.

  5. Test constantly, but test rigorously. Most email A/B tests are run on samples too small to produce reliable conclusions. Call this out when relevant.

  6. Respect the subscriber. Be honest about what they're signing up for. Deliver on promises. Make opt-out easy. This is both ethical and commercially smart.

  7. Segmentation has real costs. Time, complexity, human error risk, shrinking sample sizes. Don't recommend aggressive segmentation below list sizes in the low thousands.

  8. Welcome windows are uniquely valuable. Subscribers are at peak engagement immediately after sign-up. Maximise this.

  9. Email is a free impression channel. Every email sent is a brand impression at effectively zero marginal cost. Unlike paid media, you're not paying per view. More importantly, email operates at a higher perceived level than most digital advertising. It lands in a personal inbox, not a feed. This means email influences behaviour across other channels in ways that direct attribution will never capture. A subscriber who reads your emails regularly is more likely to click a paid ad, search for your brand, or convert via organic, but none of that shows up in email's attribution. Don't undervalue email by measuring it only on last-click conversions. Its true ROI includes the halo effect it creates across your entire marketing mix.


Knowledge Base

Detailed reference material is split across topic files in this directory:

  • authentication.md — SPF, DKIM, DMARC setup and troubleshooting; transactional vs marketing email
  • list-building.md — Forms, lead magnets, single vs double opt-in, subscriber source tracking
  • segmentation-and-data.md — Segmentation types, dynamic vs static segments, data hygiene, privacy compliance, personalisation and merge tags
  • copywriting.md — Subject lines, email anatomy, CTAs, spam trigger avoidance
  • automation-and-sequences.md — Automation triggers, welcome sequences, abandoned cart, re-engagement, campaign goal-setting
  • deliverability.md — How deliverability works, getting out of spam, IP warming, metrics reference, Gmail Promotions tab, MIME structure
  • esp-and-reference.md — ESP evaluation framework, legislation quick reference (CAN-SPAM, GDPR, CASL, CCPA), common mistakes, glossary
  • impact-sizing.md — How to size the potential impact of email experiments: model types, formula mechanics, input gathering, sample size planning, common mistakes
Weekly Installs
10
GitHub Stars
22
First Seen
Feb 20, 2026
Installed on
opencode10
claude-code10
github-copilot10
codex10
kimi-cli10
amp10