Appendix D: Methodology and Sources

This guide is built on 908 sources and 4,798 individual insights, collected and analysed through a systematic research process.

Source categories include:

• Industry reports from organisations including Litmus, Mailchimp, HubSpot, Klaviyo, Campaign Monitor, and others
• Independent blogs and publications from email marketing practitioners
• Academic research on consumer behaviour, persuasion psychology, and digital marketing
• Platform documentation and help centres for all major email service providers
• Community discussions from forums, social media, and industry events
• Expert content including podcasts, conference talks, newsletters, and published books
• Case studies and performance data from brands and agencies
• Regulatory documentation from GDPR, CAN-SPAM, CASL, and other frameworks

Research process: Content was collected using an open-source research crawler designed to systematically identify, retrieve, and extract relevant insights from across the email marketing knowledge base. Each insight was categorised by topic, validated against multiple sources where possible, and synthesised into the practical guidance presented in this guide. Where data points appeared in multiple independent sources, they were given higher confidence. Where data points appeared in only a single source, they were either validated through additional research or flagged as single-source estimates.

A note on data: Email marketing benchmarks change over time. The data in this guide reflects the best available information as of early 2026. Industry averages should be used as directional guides rather than absolute standards. Your specific results will depend on your audience, content quality, sending practices, and industry context. Where specific numbers are cited (e.g., ‘461% increase in tweets during Wrapped’), the source is the company’s published data or credible third-party analysis.

The research methodology and crawler code are open-source and available for review. If you find data that contradicts what’s presented here, or if you have more recent numbers, I’d welcome the contribution.

---