Cold Email Reply Rate Dropped: Infra Checklist

When a cold email reply rate drops, the cause is usually one of three things: what you sent changed (list, offer, or copy), where it landed changed (placement), or how much you sent changed (volume per mailbox or per domain). Before you buy a new batch of domains, work out which one it is. Check recent list and copy changes, bounce rate, authentication, blacklists, sending volume, tracking settings, and mailbox connection health, in that order. The cheapest checks come first, so start at the top.

This checklist is built for teams running Google Workspace mailboxes under a sequencer such as Instantly or Smartlead, but most of it applies to any setup.

Look at human replies, not opens. Opens are unreliable because of image proxies, privacy features, and security scanners, so an open rate change can appear with no real change in behaviour. See Why Open Rate Misleads Cold Email Teams.

Then compare like with like:

• Same segment, same offer, same sequence step, over at least a couple of weeks.
• Your own baseline, not someone else's benchmark.
• Replies from humans only. Remove auto replies and out of office messages.

If the drop holds up, work through the checklist.

This is the most common cause and the cheapest to check.

• New data source or a broader segment?
• New offer, new subject lines, new first lines?
• A new step added to the sequence, or the call to action changed?
• Links or attachments added?

If the drop started the day a new list or new copy went live, and it shows up across every domain, start here. Roll back one variable and compare.

Check bounce rate per campaign and per domain. Rising hard bounces usually mean list quality slipped, and bounces can damage mailbox and domain reputation quickly. Verify lists before sending and suppress anything that bounced.

DNS changes, a new tool added to SPF, or a lapsed domain can break authentication quietly.

1. Send a plain test from an affected mailbox to a personal Gmail address.
2. Open it, choose More, then "Show original", and confirm SPF, DKIM, and DMARC all show PASS.
3. Look up the domain's DNS: exactly one SPF record, DKIM present at the expected host, DMARC present at _dmarc, MX pointing to Google.

Google's sender guidelines require SPF or DKIM from everyone sending to Gmail accounts, and SPF, DKIM, and DMARC from senders above 5,000 messages a day. SPF DKIM DMARC Setup for Cold Email has the full record list.

Run every sending domain through a public blocklist lookup tool. If domains are listed, you have a placement problem on those domains, and resting them while you fix the cause makes sense.

If they are clean, the drop is probably somewhere else: volume, the list, or the copy. A new domain does not fix a send that was already missing the inbox, and the new domains will inherit the same habits. See Why Cold Email Domains Burn and How to Slow It.

This split tells you a lot.

• Every domain dropped together: a shared cause. List, copy, sequencer settings, a tracking change, or shared infrastructure if your inboxes sit on a shared pool.
• Only some domains dropped: something specific to those domains. Check their age, mailbox count, recent volume, and authentication.

Sort reply rate by domain, not just by campaign. Rotation can hide a weak domain inside a healthy average.

Replies can fall a few weeks after a team scales.

• Did per mailbox daily limits go up?
• Were more mailboxes added to the same domains, raising density?
• Did follow up steps increase total daily sends?
• Were new mailboxes put into campaigns before warmup finished?

Common operator practice is roughly 20 to 40 cold sends per mailbox per day once warmed, and two or three mailboxes per domain. If you went above that recently, step back down before changing anything else. See How Many Cold Emails Per Mailbox Is Safe.

For domains sending meaningful volume to Gmail users, Google Postmaster Tools shows spam rate and domain reputation. Data can be sparse at low volume. Google's guidance is to keep the reported spam rate below 0.3 percent. If complaints are up, tighten targeting and make opting out easy.

Turning on open tracking adds a pixel; click tracking rewrites links. Both are common, and both add something a receiving filter may weigh. If the drop lines up with a tracking change, test a campaign with tracking off. If you need link tracking, set up the custom tracking domain your sequencer documents (a CNAME on your own domain).

Check the sequencer for disconnected accounts, authentication errors, or mailboxes that silently stopped sending. In the Admin console, check for suspended or restricted users and for changes to 2FA or app passwords. A mailbox that stopped sending lowers reply volume without changing anything about placement.

Warmup activity can make a mailbox look busy and healthy while campaign mail struggles. Judge health by campaign replies and bounces, not warmup dashboards.

• Buy a fresh batch of domains before you know the cause.
• Switch sequencers. Domains and mailboxes come with you, and so does the problem.
• Raise volume to make up the numbers.
• Change list, copy, and volume all at once, which makes the cause impossible to find.

When the checklist points at infrastructure, ColdMail is the layer to look at. It provisions official Google Workspace mailboxes on dedicated domains and publishes SPF, DKIM, DMARC, and MX automatically, which removes a common source of silent authentication breaks. Provider health indicators for Gmail, Outlook, and Yahoo help you see when one domain is drifting, and an official admin panel per domain with 2FA keeps account status visible. AI warmup is available as an optional add on; see coldmail.app for current add on pricing.

ColdMail is not a sequencer. Mailboxes are ready to connect in Instantly and Smartlead using the methods those tools support, and they work with n8n and other tools.

If your replies dropped and you want a second pair of eyes on the infrastructure, Book a Strategy Call at coldmail.app.