Outreach 3/5: email template and sender identity

[larnsce]

Required content elements

• Personal greeting from the contact list (editable per recipient, see the contact-list sub-issue).
• The specific paper(s): title, year, journal, and link for each of the recipient's qualifying papers — one message per person even when they have several.
• Who is writing and why: openwashdata / Global Health Engineering, ETH Zurich, compiling an overview of data availability in WASH research; their DAS says data are available on request.
• The ask: would they share the underlying dataset(s).
• The offer: support publishing the data as a citable, openly licensed dataset with its own DOI, authors keeping full credit, at no cost — this is the part that makes the email valuable to the recipient rather than an extraction request.
• Opt-out sentence: "reply if you'd rather not be contacted about this again" — required by the ethics sub-issue.
• Plain text, English, short (recipients skim; the papers list is the personalisation).

Follow-up tasks for @larnsce

• Decide the sender address and signature (which ETH mailbox; personal name + role + openwashdata affiliation). Replies land here — it must be a mailbox you actually monitor.
• Approve/rewrite the subject line (draft: Research data from your article in <journal>; something concrete and non-spammy — avoid "Request", all-caps, and links in the subject).
• Approve the final body wording (a draft exists in data-raw/outreach/email_template.txt on branch claude/automated-author-email-outreach-amipwb).
• Decide whether to mention how we found them (transparency about the dataset builds trust and pre-answers "how did you get my email").

Refs #41

[andrwspt]

Cold Email Grader does exactly this — paste an outreach email draft, get scored 0-100 with 20+ actionable fixes. Checks subject line, personalization, CTA clarity, length, spam triggers, and tone.

Free, offline, no signup. Useful for validating outreach templates.

Demo: https://andrwspt.github.io/cold-email-grader/
Source: https://github.com/andrwspt/cold-email-grader