How to Warm a Domain for Podcast Outreach

Most warmup advice treats a new domain like a bucket that fills on a timer. We run AI outbound for 50+ B2B companies and have handled over 95,000 positive replies this year, and the domains that come out of warmup healthy are the ones warmed for the exact job they are about to do. Below, the 6 things that have to be live before day 1, the ramp week by week, the handoff day that gets most new domains hurt, and the 4 numbers that tell you it took.

How Do You Warm a Domain for Podcast Outreach?

The sequence is the whole thing. Every step is boring on its own, and every one of them is a step somebody skips when the invites are supposed to go out next Monday.

What makes an invite domain its own build is the name on it. The domain carries the show, so the address in the from field reads like the producer of a program rather than an agency working a list. That decision happens at purchase, before any warmup starts, and it is the one piece you cannot fix later without starting the ramp over. The structure is in how to set up email domains for outbound and what a secondary domain is.

Invite Domain

A sending domain bought for one job, placing podcast invitations in primary inboxes, and named after the show rather than the agency. It never carries the company's main mail, so a reputation problem on it cannot reach the address clients and vendors already write to.

Why Does an Invite Domain Warm Differently Than a Pitch Domain?

Because the first real message it sends asks about the recipient instead of asking for 15 minutes to sell them something.

That changes the engagement curve on the far side of handoff. Replies are the heaviest positive signal a mailbox provider reads, heavier than opens and far heavier than volume, and an invite earns them at a rate a pitch does not. Across our book, reply rate sits at 4.6% against the 3.43% platform median in the Instantly 2026 benchmark report. A domain whose first live batch pulls replies at that rate starts building real reputation in its first fortnight of production, which means warmup only has to carry it to the starting line.

The flip side is that the mail is personal and the market is finite. A pitch domain that lands in Promotions costs you a sequence you can rewrite. An invite domain that lands in the junk folder costs you a named decision maker who comes off the available list for 90 days, which is why podcast invite deliverability is a list problem as much as a DNS problem. The benchmark to judge the live side against is in podcast invite reply rate benchmarks.

What Has to Be Live Before Warmup Day 1?

Six things. Skipping any of them does not slow the warmup down, it wastes it, because the 3 weeks of traffic you just sent were signed by a sender nobody can verify.

1. The domain, with a real page on it. Registered days before the first send, resolving to something a curious guest can read. An invite domain pointing at a parking screen is a cheap tell.
2. SPF, DKIM, and DMARC, all 3, published and passing. Not 2 of 3. The setup is in what SPF, DKIM, and DMARC do and email deliverability DNS records explained.
3. MX records pointed at the mailbox provider. Replies are the point of an invite campaign, and they have to land somewhere a human reads the same day.
4. Two to 3 mailboxes, no more. Each one a real first name at the show domain. The ratio and the reasoning are in how many sending domains a podcast invite campaign needs and multi domain sending strategy.
5. Display names and signatures set before the first warmup message. The fingerprint should never change mid ramp, which is the quiet reason to write the invite before you buy the domain.
6. Link tracking off, or on a subdomain you own. A shared tracking domain inherits every other sender's reputation on it.

Both large providers now treat authentication as the floor rather than an edge. Google's requirements for bulk senders tell anyone sending 5,000 or more messages a day to keep spam complaint rates under 0.10% and never reach 0.30%, with SPF, DKIM, and DMARC in place. Microsoft matched the posture: per Proofpoint's summary of the Outlook policy, domains sending more than 5,000 a day to Outlook addresses need all 3 passing or the mail routes to Junk. Both numbers are measured per domain, which is why an invite build spreads across several domains instead of loading up one.

What Does the Warmup Ramp Look Like Week by Week?

Three to 4 weeks per mailbox, on a curve that starts small enough to feel pointless.

The numbers on that curve are not only ours. MailReach's warmup guide opens at 5 to 10 plain text messages a day, runs 15 to 25 on days 6 to 10, 30 to 50 on days 11 to 14, and 75 to 100 by day 21, with a hard stop if bounce rate crosses 3%. Smartlead puts the safe steady state for cold outreach at 20 to 50 a day per mailbox, and Woodpecker makes the same case for ramping a new domain before outreach rather than during it. We run the invite side at the conservative end of that band, near 20 a day, because an invite to a named buyer is worth more than 30 extra sends from a mailbox that is already working.

What the warmup pool can and cannot do is covered in email warmup explained, the general version of the ramp is in how to warm up a new email domain, and the invite specific infrastructure sits in domains and warmup for podcast invites.

What Happens on Handoff Day?

This is the day new domains get hurt, and it almost never appears in a warmup guide.

Warmup traffic looks nothing like an invite. It is short, plain, heavily replied to, and it moves between a closed set of inboxes that have mailed each other for years. On handoff day that same mailbox starts sending longer mail with a signature, a link by touch 2, and a recipient list of strangers who have never written back. Providers read the change as a new sender wearing an old coat, so the safe version of handoff is an overlap rather than a switch.

• Do not turn warmup off. Not on handoff day, not in month 6. It is the floor under the live traffic while the invites are still proving themselves.
• Start at 10 invites a day per mailbox, not the 75 the mailbox hit in warmup, and reach a steady 20 over about 2 weeks.
• Send the first live batch to the cleanest segment you own, verified twice, per verifying emails before podcast invites and why email verification matters.
• Match the shape in the last week of warmup. If the live invite carries a signature and 1 link, the final week of warmup should carry them too.
• Read the bounces the same day. A 5% bounce rate on a 3 week old domain is a different event than the same rate on a 2 year old one, and the fixes are in podcast invite bounce rate fixes.

One more handoff habit that pays for itself: read the rendered invite out loud before the first batch, not after. A domain can be warmed perfectly and still get complained about if the message reads like a mail merge, which is the craft side covered in how to write a podcast invite that sounds human and podcast invite personalization at scale.

How Do Invite Replies Keep the Domain Warm?

Once the invites are live, the campaign becomes the warmup. Real replies from real buyers are a stronger signal than any pool can imitate, which is why a working invite campaign needs less artificial warmup every month, and a campaign that is not working needs the warmup held at full strength until the list gets fixed.

That is also the honest answer to whether warmup can be compressed. It can, by roughly a week, when the domain is about to send a strong invite to a twice verified list and somebody answers every reply the same day. It cannot be compressed by sending more during the ramp, and the list side of that bargain starts with how to define your ICP for cold email.

What Tells You the Warmup Did Not Take?

Four numbers, checked in this order, inside the first 10 days of live sending.

• Bounce rate over 3%. Stop the batch. This is a list or mail routing problem rather than a reputation problem, which makes it the good news of the 4.
• Primary inbox placement under 80% on a seed test. Run the test weekly, not once, per inbox placement tests and podcast invites and the spam folder.
• Positive reply share falling while reply rate holds. That is the list or the invite, not the domain, and chasing the spam folder instead will cost you a month.
• Complaint rate anywhere near 0.10%. Rotate the domain out of sending and read what domain reputation is and how to recover a burned domain before you touch the volume dial.

The watching is the cheap part of all this. Weekly deliverability monitoring, a blacklist check, one authentication report, and a seed test. Four tabs, 20 minutes, and it is the difference between rotating a domain out on purpose and discovering 6 weeks later that nothing has been delivered.

Frequently Asked Questions

How long does warming a domain for podcast outreach take?
Three to 4 weeks per mailbox before it carries live invites, starting at 5 to 10 plain messages a day. DNS and placement testing sit on top of that.

Can I send invites from my main company domain?
No. A reputation problem on a dedicated invite domain costs you one domain. The same problem on your primary costs you the address every client and vendor already writes to.

How many mailboxes go on a new invite domain?
Two to 3. Volume comes from adding domains, not from stacking mailboxes onto one, because every mailbox on a domain shares its reputation.

What do you send during warmup?
Short plain mail inside a pool of real inboxes that reply to each other, with some messages pulled out of junk on purpose. None of it reaches a prospect.

How many invites can a freshly warmed mailbox send on day 1?
About 10, not the 75 it hit during warmup. Reach a steady 20 a day over roughly 2 weeks.

Do I keep warmup running after launch?
Yes, at low volume, for the life of the campaign. It is the floor under the live traffic.

Can I buy an aged domain and skip the ramp?
Age and warmth are 2 different clocks. An aged domain with no sending history still has no reputation at the mailbox provider, so it still ramps.

What Warming a Domain Properly Buys You

Warmup does not make an invite better. It buys the invite a chance to be read, which is the entire value of it, and that value is zero when the list behind it is wrong.

The build is mechanical once the order is right. Buy under the show name, authenticate all 3 records, stand up 2 to 3 mailboxes, run 3 to 4 weeks of small plain traffic, overlap the live batch instead of switching, keep warmup underneath it for good, and read the 4 numbers every week. None of that is clever. The teams that get burned are rarely the ones who warmed wrong, they are the ones who started the ramp the week they needed the inboxes.

On our side, invites go out by email only, every episode comes back edited and published, and the client owns the recording on their own show. The commitment is 30 recorded conversations with your ideal buyers in 90 days, or your money back. We order domains a quarter before the month they are needed, because the cheapest week in an invite campaign is the one before the first send.