AI Email Writing — Personalized Cold Emails at Scale
Mailor writes each cold email from the prospect's own context — company, role and public
information — combined with your product positioning. The result reads as a one-to-one email rather
than a template with merge fields, and it is generated per prospect rather than per campaign.
Why does personalization change reply rates?
Recipients judge relevance in the first line. A message grounded in something true about their
company clears that bar; a mail-merged greeting does not. Personalization at scale is the reason to
automate this step at all — writing one great email by hand is easy, writing four hundred is not.
How does Mailor write the emails?
- Pick a lead. Mailor pulls that prospect's company and role context from the
lead record.
- Generate. The model drafts a subject line and body grounded in that context,
in your chosen tone and language.
- Edit and send. Refine inline, switch tone or language, regenerate for a
different angle, then send.
What controls do you keep?
- Tone — six tones, so the same offer can read direct, value-first,
curiosity-driven, social-proof or story-shaped.
- Language — 18 languages, so a prospect can be written to in their own
language.
- Length guardrails — word-count guidance keeps emails inside the range that
cold-email data consistently favours over long copy.
- Spam pre-check — generated copy is checked for common spam triggers before
it can be sent.
- Regeneration — a different angle for the same prospect, without rewriting
from scratch.
What does a generated email look like?
A subject line for a Shopify Plus CX lead can reference a concrete operational concern — for
example, cutting response time — and close with a light ask rather than a hard pitch. The specific
detail is drawn from the lead, which is what makes it read as researched rather than automated.
Related