Seven cold outreach myths that cost you the inbox

Most bad cold outreach advice is harmless. The beliefs below are not: each one leads directly to a decision that raises complaints, burns a sending domain, or sends you optimising a number that stopped meaning anything years ago.

Myth 1: cold outreach is spam

Legally, the word "spam" has definitions, and relevant B2B outreach can sit inside them. Under CAN-SPAM in the US, commercial email is lawful if the headers and subject line are not deceptive, there is a clear opt-out, you include a valid physical postal address, and you honour opt-outs within 10 business days. Under GDPR in the EU and UK, you need a lawful basis — legitimate interest is often available for B2B, provided you actually perform and document the balancing assessment, tell people where their data came from, and act on objections. National laws implementing the ePrivacy Directive add further rules on unsolicited electronic marketing, and they differ meaningfully by country.

What makes mail spam in the eyes of the only judge that matters — the inbox provider — is simpler: recipients report it. You can be fully compliant and still be filtered, because filtering is driven by engagement and complaints, not by legal status. Both things have to be true: lawful, and wanted.

Myth 2: more volume fixes a bad reply rate

If 500 emails produced two replies, 5,000 will produce twenty replies and a deliverability problem. Volume multiplies whatever your message already is. A low reply rate is a targeting or copy signal, and the correct response is a narrower segment, not a bigger one.

Volume also carries you across thresholds you cannot come back from quickly. Google asks bulk senders to keep reported spam below 0.3%; at ten times the volume, the same complaint *rate* produces ten times the complaints, and the domain damage arrives faster than your pipeline does.

Myth 3: warm-up is something you finish

Warm-up is not an onboarding step. Reputation is a rolling judgement about recent behaviour, and it decays. A mailbox that was landing cleanly in March can be filtered in June after a bad list, a volume jump, or two weeks of silence followed by a burst.

Run warm-up continuously alongside live sending, keep volume growth gradual, and watch per-mailbox reputation rather than assuming a mailbox that was fine last quarter still is.

Myth 4: open rate tells you whether people are interested

It does not, and has not for years. Apple's Mail Privacy Protection pre-fetches tracking images for Apple Mail users, and corporate security gateways do the same, so a meaningful share of recorded opens are machines. A subject line that wins on opens may just have reached a more Apple-heavy audience.

Measure replies. Watch bounces per domain and complaints per provider. And watch inbox placement, because a campaign with no replies and no bounces is usually a campaign in the spam folder — which your sequencer will report as 100% delivered.

Myth 5: spam filters are looking for spam words

Avoiding "free", "guarantee" and exclamation marks is cargo cult. Modern filtering decides on authentication, sending history, domain and IP reputation, list quality, recipient engagement and complaint rates. Content is a small input, and content *patterns* matter more than vocabulary — an image-only message with six links and no plain text is a worse signal than any single word.

The real content rules are dull: real text, few links, no attachments on cold mail, a signature a human would have, and a plain-text part that matches the HTML.

Myth 6: cold outreach needs a team

It needs a narrow segment and a repeatable process. One person working a 200-account list with a real trigger per account will out-reply a team working 20,000 generic addresses, and will not burn a domain doing it. Headcount buys list size, and list size is the input most likely to hurt you.

Myth 7: we are on a dedicated IP, so we are fine

A dedicated IP means your reputation is yours alone. That is a double-edged benefit: nobody else's complaints hurt you, and nobody else's good behaviour helps you. A fresh dedicated IP has no history at all, which is worse than a well-run shared pool until you have warmed it.

More importantly, for Gmail and Microsoft the domain is at least as significant as the IP. You can move IPs and keep your problem, because the problem is attached to the domain that signs the DKIM signature.

The one nobody calls a myth: "delivered" means delivered

Every sequencer reports "delivered" when the receiving server accepted the message with a 250. That tells you nothing about the folder. Gmail's Promotions tab, for instance, is an inbox tab — mail there is delivered and categorised, and it reads as a clean hit in most tooling while replies quietly do not arrive. Spam folder delivery looks identical to inbox delivery from the sender's side. If you are not measuring placement independently, you are not measuring deliverability.

The myths, scored

What to do instead

1. Keep lists small and verified; bounce rate is the earliest warning you get.
2. Send from separate outreach domains, never the domain your invoices leave from.
3. Publish SPF, DKIM and DMARC on every sending domain, and read the DMARC reports.
4. Add List-Unsubscribe with an HTTPS URL and a mailto, plus List-Unsubscribe-Post: List-Unsubscribe=One-Click per RFC 8058, once you are sending at bulk volumes. A working unsubscribe is always cheaper than a complaint.
5. Stop after two follow-ups.
6. Warm continuously and measure placement per provider, not in aggregate.

Where MailPilot fits

MailPilot replaces the guesswork these myths fill in. It warms mailboxes continuously over OAuth on Gmail and Outlook or over SMTP/IMAP for SendGrid, Amazon SES, Mailgun, Postmark and custom SMTP, checks SPF, DKIM and DMARC for each sending domain, measures real inbox placement per provider with a seed-inbox network, and tracks per-mailbox reputation daily — so "delivered" stops being the number you steer by.