# Cold email guides

> Cold email guides from the first DNS record to the booked meeting: warmup, SPF, DKIM and DMARC, sending limits, sequences, replies and AI agents.

Source: https://docs.ayinbox.com/learn/

These guides explain how cold email works, from the DNS records that let a receiver trust your domain to the reply that turns into a booked meeting. They cover deliverability, campaigns, replies and meetings, and running outbound with an AI agent, and most of them also say what AY Inbox does at that step.

Deliverability [#deliverability]

* [Email warmup](/learn/email-warmup/): what warmup does for a new mailbox, how long it takes, and why it keeps running after campaigns launch.
* [Warmup pools](/learn/warmup-pools/): how the mailboxes in a pool warm each other, and why a vetted pool is worth more than a large one.
* [SPF, DKIM and DMARC](/learn/spf-dkim-dmarc/): what each record checks, what alignment means, and how to move DMARC from `p=none` to `p=reject` in steps.
* [Cold email deliverability handbook](/learn/deliverability/): the signals that decide whether accepted mail reaches the inbox, the thresholds to stay under, and what to measure.
* [Inbox placement vs delivery rate](/learn/inbox-placement/): why a high delivery rate can hide mail landing in spam, how placement is measured, and the rate to aim for.
* [How many cold emails to send per mailbox](/learn/cold-email-sending-limits/): AY Inbox's default daily cap and gap between sends, why a new mailbox starts lower, and why more volume should come from more mailboxes.
* [Cold email laws by country](/learn/cold-email-rules/): what CAN-SPAM, GDPR and PECR, CASL and Australia's Spam Act ask of a sender, and where consent comes first.
* [Recovering sender reputation](/learn/reputation-recovery/): how to decide whether a mailbox landing in spam is worth saving, the ramp back over about six weeks, and when to retire it.

Campaigns [#campaigns]

* [Personalization with merge tags](/learn/personalization/): variables, fallbacks, conditionals, functions and spintax, and one complete template built from them.
* [Cold email sequences](/learn/cold-email-sequences/): how many follow-ups to send and how far apart, when to reply in the same thread, and how stop on reply ends a sequence.
* [Using your own leads and mailboxes](/learn/bring-your-own-leads-and-mailboxes/): importing a list you built, keeping the verification results you already paid for, and connecting the mailboxes you own.

Replies and meetings [#replies-and-meetings]

* [Speed to lead](/learn/speed-to-lead/): why the first answer to a reply should come within minutes, where the delay usually comes from, and how to measure it.
* [Categorizing cold email replies](/learn/reply-classification/): setting aside bounces and auto-replies first, then labelling each human reply by what the person wants.
* [Opt-out replies](/learn/handling-opt-out-replies/): how to tell a request to stop from a plain no, and why a request to stop ends mail from every mailbox at once.
* [Out-of-office replies in cold email sequences](/learn/out-of-office-replies/): why an away message is not a reply, and how a lead is held until the person is back.
* [AI replies to cold email](/learn/ai-replies-to-cold-email/): what an AI draft needs to read to answer well, how to keep it from inventing facts, and where a person approves it.
* [Safe AI auto-replies](/learn/safe-automatic-replies/): what to fix before an AI answers prospects with no one approving, from the links and figures it may use to the replies it must hand to a person.
* [Knowledge base for AI sales replies](/learn/campaign-knowledge-base/): what to write down for each offer so drafted replies quote the right price, links and rules, and what never goes in.
* [Booking meetings from replies](/learn/book-meetings-from-cold-email-replies/): when to offer named times instead of a scheduling link, getting the prospect's timezone right, and booking the time they pick.
* [Following up on a meeting request](/learn/follow-up-until-booked/): how often to follow up when a prospect asks for a call and then goes quiet, how far apart, and when to stop.

Agents [#agents]

* [Cold email for AI agents: MCP, API and CLI](/learn/cold-email-for-ai-agents/): what an agent can run over MCP, the REST API or the CLI, and which actions stay behind a permission a person grants.
* [Cold email from Claude Code](/learn/outbound-from-claude-code/): connecting Claude Code through the MCP server or the `ayinbox` CLI, and a first session from importing leads to drafting replies.

Reference [#reference]

* [Cold email glossary](/learn/glossary/): short definitions of the terms used across these docs, from alignment and complaint rate to spintax and warmup pools.
