Outreach Email Deliverability: Daily Limits, Warm-up, SPF, DKIM and DMARC

Cold email deliverability is the share of your outreach emails that reach the recipient’s primary inbox rather than the spam folder, a promotions tab or a silent rejection. It rests on four things you control: the domains and inboxes you send from, DNS authentication (SPF, DKIM and DMARC), how fast you ramp up, and who is on your list. Get them right and good copy gets a fair chance; get them wrong and no subject line will save the campaign.

This lesson belongs to the Outreach Academy. For the full map of channels, start with the pillar guide to outreach marketing. The cold outreach lesson gives a one-screen overview of this topic; here every layer gets the detail: DNS examples, a warm-up schedule, a setup checklist and a troubleshooting table. Our B2B outreach service sets up this foundation before any campaign goes live.

Why deliverability comes first

Every other part of outreach assumes the email was seen. Targeting, personalization, offer and follow-ups all multiply a number that starts with inbox placement. If half of your emails sit in junk, the best copy works on half the audience, and the reply rate you measure understates the real one.

Delivery rate vs inbox placement

Sending tools report a “delivered” figure. It only says the receiving server accepted the email. Mail routed straight to junk also counts as accepted, so a dashboard can show 98% delivered while most of a campaign sits unread. Inbox placement is the metric that actually matters: the share of emails that reach the primary inbox. Aim for high placement at every major provider; you only learn the figure by testing, which the monitoring section covers.

Why cold email starts at a disadvantage

A newsletter has opt-ins, a history of opens and readers who recognize the brand. A cold campaign has none of that. Nobody asked to hear from you, there is no previous thread, and the first impression of you is the email itself. Mailbox providers therefore judge a cold sender almost entirely on infrastructure and behavior: is it authenticated, does volume grow naturally, do recipients reply or hit the spam button. Good deliverability pays back in more replies, a sending identity that stays healthy, less throttling, and the ability to grow outreach predictably.

Sender reputation: what mailbox providers judge

Inbox placement is decided by the receiving side, not by your sending tool. The core input is sender reputation: a score that receivers keep for your domain and for the IP addresses your mail leaves from.

What builds reputation	What damages reputation
Replies and real conversations (the strongest positive signal)	Spam complaints (“Report spam” clicks)
Recipients pulling mail out of junk or marking it important	Hard bounces to mailboxes that do not exist
SPF, DKIM and DMARC passing with alignment	Spam-trap hits (dormant or seeded inboxes)
Steady daily volume	Sudden spikes, above all on a new account
Age and a clean history	Blocklist entries such as Spamhaus or Barracuda
Emails people open and read	Emails deleted unread by people with no reason to receive them

Three properties shape everything below. Reputation cannot be transferred: a fresh name starts with no history, and a second-hand one may carry someone else’s bad record. Reputation decays: an account that stops sending, or sends badly for a few weeks, loses standing. And B2B gets no exception. Company inboxes run on Google Workspace, Microsoft 365 and similar services, so outreach to companies is filtered by the same rules as consumer mail. Recipient behavior is what the filters trust.

On a shared IP pool (common with sending platforms), other customers also affect the IP side of your score. Domain reputation is the part you fully own, which is why the decisions below matter so much.

Gmail, Yahoo and Outlook.com sender requirements

Since 2024 the three largest consumer mailbox providers have published formal rules. They target bulk senders, but the baseline covers everyone, and the thresholds work as guardrails for cold email too.

Provider	Applies to	What is required	Source
Gmail	All senders to personal accounts	SPF or DKIM; valid forward and reverse DNS (PTR); TLS; RFC 5322 formatting; no impersonation of its addresses in From:; spam rate below 0.3%	Google sender guidelines
Gmail	Bulk senders, roughly 5,000 or more emails in 24 hours to personal accounts	SPF and DKIM; a DMARC record (policy may be p=none); From: domain aligned with the SPF or DKIM domain; one-click unsubscribe for marketing and subscribed mail plus a visible unsubscribe link	Google sender FAQ
Yahoo	All senders	SPF or DKIM; spam rate below 0.3%; valid forward and reverse DNS; RFC 5321/5322 compliance	Yahoo Sender Hub
Yahoo	Bulk senders (no numeric threshold published)	SPF and DKIM; DMARC at p=none or stricter that passes; From: alignment; one-click List-Unsubscribe (RFC 8058 POST highly recommended); visible unsubscribe link; opt-outs honored within 2 days	same page
Outlook.com, Hotmail, Live	Over 5,000 emails a day to these consumer mailboxes	SPF pass, DKIM pass, DMARC at least p=none aligned with SPF or DKIM; a working unsubscribe link is recommended	Microsoft announcement, April 2025

Details that are easy to get wrong:

• The bulk threshold is not a precise line. Google’s own pages say “more than 5,000”, “5,000 or more” and “close to 5,000” in 24 hours. Mail from the same primary domain is counted together, so sales.example.com adds to example.com. Bulk status, once reached, has no expiration date.
• Enforcement is stricter now. From November 2025 Google is ramping up enforcement on non-compliant traffic, with temporary and permanent rejections.
• Outlook.com: non-compliant mail may be rejected. The April 2025 update says rejection with 550; 5.7.515 Access denied from 5 May 2025, while the Outlook postmaster policy page still describes junk-folder routing first. Plan as if rejection is the outcome.

Spam-rate thresholds: 0.10% and 0.30%

Google asks senders to keep the user-reported spam rate in Postmaster Tools below 0.10% and never to reach 0.30%. The rate is calculated daily. A bulk sender above 0.3% loses eligibility for mitigation and regains it only after seven consecutive days below that line (Google FAQ). Yahoo uses the same 0.3% ceiling.

For cold outreach the arithmetic is harsh: three spam flags per thousand emails put you at the ceiling. A list that annoys even a small slice of recipients gets there fast, which is why targeting belongs in a deliverability guide at all.

One-click unsubscribe (RFC 8058) and when it is needed

One-click unsubscribe is defined in RFC 8058. The message carries two headers, and both must be covered by a valid DKIM signature:

List-Unsubscribe: <https://example.com/unsubscribe?id=8f3a2c>
List-Unsubscribe-Post: List-Unsubscribe=One-Click

Gmail and Yahoo show an unsubscribe button next to the sender name; one click sends a POST request to that HTTPS URL without opening a page.

When is it mandatory? Gmail demands it from bulk senders for marketing and subscribed messages, and Yahoo from its bulk senders. Senders that already had unsubscribe links in promotional mail had to add one-click by 1 June 2024. A small cold campaign spread over dedicated domains usually stays far below the bulk threshold, so the header is not formally needed there. It still helps: a recipient who can leave in one click has no reason to press the spam button, and every avoided complaint protects the 0.3% ceiling. Separately, the law asks for an opt-out in every commercial message. In the US, CAN-SPAM gives you 10 business days to honor it (FTC guide); Google recommends processing unsubscribes within 48 hours, and Yahoo asks for 2 days.

Outreach email limit per day: provider caps vs safe cold volume

Daily limits exist because volume is itself a spam signal. A brand-new account that sends 200 emails on day one looks exactly like a compromised or throwaway one. A fresh domain has no reputation to absorb the spike, so mail goes to junk or gets throttled.

Provider caps are hard ceilings, not targets

Account type	Published limit	Source
Google Workspace	2,000 messages per user per day, rolling 24 hours (trial accounts: 500)	Workspace sending limits
Google Workspace	10,000 total recipients a day, 3,000 external; 1,500 a day via mail merge; 100 recipients per message via SMTP	same page
Consumer Gmail	Limit triggered by over 500 recipients in one email or over 500 emails in a day; sending resumes within 1–24 hours	Gmail help

Workspace limits can change without notice, and trial accounts get higher limits only after $100 in cumulative payments, which can take up to 75 days. Other platforms publish their own caps; take current figures from their documentation, not from a blog post. A custom SMTP relay removes the platform cap but not the reputation problem.

None of these figures is a cold email recommendation. They protect the platform’s infrastructure and count all outgoing mail. An outreach inbox sending anywhere near 2,000 a day will be filtered long before it hits the cap, and accounts that exceed quotas get temporarily blocked.

Safe cold emails per inbox per day (common practice)

There is no official safe number. Practitioners converge on volumes well below the caps. Treat the table as common practice, not an official rule, and adjust it to your own data.

Inbox state	Cold emails per inbox per day (common practice)
Brand-new inbox on a fresh domain, weeks 1–2	0, warm-up only
End of warm-up, weeks 3–4	10–20, best-fit prospects only
Fully warmed, clean metrics	20–30
Aged account with a long clean history	30–50, raised slowly while metrics stay clean

Set a hard daily cap in your sending tool so a misconfigured campaign cannot overshoot, and space emails minutes apart instead of in bursts. If you write fewer than about ten emails a day by hand to known contacts from an established company inbox, separate outreach infrastructure is less critical. Anything that runs as a campaign deserves its own infrastructure.

How many inboxes and domains you need

Once the per-inbox limit is fixed, scale comes only from adding inboxes and domains, never from raising the per-inbox ceiling.

Inboxes needed = daily cold volume ÷ cold emails per inbox
Domains needed = inboxes ÷ inboxes per domain
Then add a 20–30% buffer for warm-up, rotation and domains that need rest.

A worked example: a team wants 300 cold emails a day. At 25 per inbox that is 12 inboxes. At three per domain that is 4 domains. With a 25% buffer, plan for five of them and 15 inboxes, so one can rest or be replaced without stopping the campaign. Teams usually keep 2–5 inboxes per domain; more than that concentrates exposure again. Doing this math first helps turn “more pipeline” into a concrete infrastructure plan with a known cost and a lead time of several weeks.

Dedicated sending domains: why, how many and how to set them up

Nothing in the inbox providers’ rules says cold email must come from a separate domain. It is practice, not law or a provider rule, but nearly every serious outbound team follows it for one reason: isolation.

Why cold outreach moves off the main domain

Your primary domain carries invoices, support tickets, contracts, password resets and partner conversations. If cold outreach damages it, all of that starts landing in junk, and recovery can take weeks or months. A secondary domain absorbs the damage: if it burns, you pause it and the company keeps working.

Subdomains (sales.example.com, outreach.example.com) are a middle path. They separate part of the reputation but share the organizational identity of the root, and Google counts their mail together with the root for the bulk threshold. Separate root names isolate more cleanly. Either way, each one needs its own authentication, warm-up and monitoring.

A portfolio of several sending names adds three benefits: load is spread so none of them carries much; one can pause while the others keep going; and campaigns can be split by product, region or language, which makes the data easier to read.

Naming, history and redirects

• Stay close to the brand. getexample.com, tryexample.com, example-team.com, examplehq.com. A name unrelated to the company reads like phishing.
• Prefer established TLDs such as .com, .net, .io or your country code. Cheap extensions with a history of abuse attract extra filtering.
• Avoid numbered names like example1.com, example2.com; they seem disposable to filters and to people.
• Research the history. A second-hand name may bring old problems. Check it on major blocklists and in web archives before buying.
• Redirect with a 301 to your main website, so anyone who types it in reaches the real company site.

Inboxes, DNS and the order of setup

1. Connect the new name to an email platform (Google Workspace, Microsoft 365, Zoho or a specialized cold email infrastructure service), verify ownership and publish MX records so replies and warm-up mail are received.
2. Create the inboxes under real team members’ names, with a photo and a plain signature. Role names like sales@ seem automated.
3. Publish SPF, DKIM and DMARC, as described in the next section.
4. Add a custom tracking domain (a CNAME on your own subdomain) if your tool tracks clicks, so links do not point to a tracking host shared by thousands of other companies.
5. Audit before launch: blocklist status, DNS validity, a sample email with headers reviewed.

Different platforms have different trade-offs. Inboxes on the big hosted suites are widely accepted but need more manual setup and have acceptable use policies; read them, because bulk cold sending can conflict with platform terms. Specialized services automate DNS and allow more inboxes per domain, with less transparency about the shared infrastructure behind them.

SPF, DKIM and DMARC for cold email

Authentication is the first technical gate. It proves the email really comes from the domain in the From: line. The examples below use example.com; adapt them to your platform.

SPF: which servers may send

SPF (Sender Policy Framework) is a TXT record naming the servers allowed to send mail for your domain. A Workspace user would publish:

example.com. TXT "v=spf1 include:_spf.google.com ~all"

If a sequencing tool sends through its own servers, its include: goes into the same record:

example.com. TXT "v=spf1 include:_spf.google.com include:spf.sendingtool.example ~all"

Rules that break SPF silently:

• One SPF record per domain. Two v=spf1 records make SPF fail. Merge them.
• Ten DNS lookups maximum. Every include:, a, mx and redirect costs lookups, nested ones too. Over ten, SPF returns a permanent error (RFC 7208).
• ~all vs -all. ~all (softfail) marks unlisted servers as suspicious; -all (hardfail) says refuse them. Many teams use ~all while still adding tools and switch to -all once the set of senders is final.

DKIM: a signature that proves integrity

DKIM (DomainKeys Identified Mail) signs each outgoing email with a private key. The public key sits in DNS under a selector, so receivers can confirm the email was not altered and came from you:

google._domainkey.example.com. TXT "v=DKIM1; k=rsa; p=MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEA..."

Here google is the selector; your platform tells you which one to use. Gmail accepts keys of at least 1024 bits and recommends 2048 (Google). Use 2048 where your DNS host allows it and rotate keys periodically. A common trap: publishing the key does not start signing. On the big hosted suites you also switch signing on in the admin console.

DMARC: alignment and policy

DMARC ties the two together. It confirms that SPF or DKIM passes and that the passing domain is aligned with the visible From: domain, tells receivers what to do on failure, and says where to send reports. The record lives at _dmarc:

_dmarc.example.com. TXT "v=DMARC1; p=none; rua=mailto:[email protected]; adkim=r; aspf=r"

• p=none: take no action, only send data. The minimum accepted from bulk senders by all three big inbox services.
• p=quarantine: treat failing mail as suspicious (usually the junk folder).
• p=reject: refuse failing mail.
• rua=: where daily aggregate reports go. They reveal every service sending as your domain, including the ones you forgot.

Alignment is where outreach setups usually break. If a sequencing tool sends with its own envelope address and signs with its own key, SPF and DKIM pass for the tool while DMARC fails for you. The fix is custom DKIM inside the tool, so the signature matches the From: line.

Choosing a DMARC policy for outreach domains

Do not jump to p=reject. A strict policy with one misconfigured legitimate source blocks your own mail. Move gradually:

Stage	Record	When
Start	v=DMARC1; p=none; rua=mailto:[email protected]	New setup, warm-up, adding tools
Tighten	v=DMARC1; p=quarantine; pct=25; rua=mailto:[email protected]	Reports show all legitimate mail passing and aligned for a few weeks
Enforce	v=DMARC1; p=quarantine; rua=mailto:[email protected]	Placement stable, no failing legitimate sources
Optional	v=DMARC1; p=reject; rua=mailto:[email protected]	Every source passes and you accept that any error will block mail

Treat p=none as the floor, not the goal. Google’s guidelines state that Gmail “will begin using a DMARC quarantine enforcement policy”, without a date, so moving toward quarantine on your own schedule makes sense (Google). Today’s bulk rule still sets the floor at p=none, not quarantine or reject. An enforced policy is also the precondition for BIMI, which shows a brand logo next to authenticated mail.

MX, PTR and confirming that it all passes

• MX records must point to your email platform so replies and warm-up mail are received, not rejected.
• Reverse DNS (PTR) matters if you control the sending IP; the big hosted suites handle it for you.
• Confirm with a real email. Send a test to a personal Gmail account, choose “Show original” and look for SPF: PASS, DKIM: PASS and DMARC: PASS. Raw headers show the same as Authentication-Results: spf=pass dkim=pass dmarc=pass.

What authentication does not solve

SPF, DKIM and DMARC are table stakes: unauthenticated bulk mail is filtered or refused outright. But authentication only proves who sent the email, not that anyone wants it. Mail that passes all three can still go to junk because of complaints, dead mailboxes among recipients, heavy tracking, a sudden jump in volume or recipients who delete without reading. Receivers evaluate the whole system: infrastructure, behavior, content and reactions. The next sections cover what authentication cannot.

Email warm-up: what it is and how it works

Email warm-up (also spelled warmup) is the gradual, consistent increase of daily volume on a new or inactive inbox, so receivers see a natural pattern instead of a spike. It gives a fresh account an initial reputation before real outreach starts.

How warmup networks work

Most teams automate it. A warm-up tool connects your inbox to a network of other inboxes. The network writes to you, you write back, and its accounts open those emails, reply, mark them important and pull any that land in junk back to the inbox. These are system-generated conversations, not your campaigns. Typical settings let you choose a starting volume, a daily increment (2 emails on day one, 4 on day two, 6 on day three, and so on), a ceiling, a reply target and randomized numbers so activity does not look scripted. Network size and quality matter: a large pool of real company inboxes is worth more than a small pool of free accounts.

Warm-up schedule (illustrative, common practice)

There is no official schedule from any mailbox provider. This plan reflects common practice for a fresh setup with a few inboxes; adjust it to your own data.

Period	Warm-up emails a day	Real cold emails a day	What to do
Day 1	0	0	Publish SPF, DKIM, DMARC, MX; set the 301 redirect; review headers
Days 2–7	5–10, rising slowly	0	Switch on warm-up; write a few manual emails to colleagues and partners who answer
Days 8–14	10–25	0	Grow by small steps; run placement tests and blocklist lookups
Days 15–21	20–30	0–10	Test placement at Gmail and Outlook separately; a small batch to verified, best-fit prospects
Days 22–28	20–30	10–20	Watch error rates and replies daily
Day 29+	10–20 (maintenance)	20–30	Keep warm-up on next to campaigns; increase cold volume only on clean metrics

Common estimates of duration range from two to six weeks. Gmail tends to respond faster than Outlook, inboxes on a shared SMTP IP often take longer, and a previously flagged name may need six weeks or more. Two rules hold in every version: never increase volume faster than the metrics justify, and keep activity consistent from day to day.

What to expect: some mail lands in junk or promotions in the early days, Hotmail and Live may return temporary deferrals, and placement swings before it settles. You are ready when test emails consistently reach the inbox at every major provider, no blocklist includes you, and, once volume is large enough to show data, Postmaster Tools shows a low spam rate and good authentication results.

Manual vs automated warm-up

Manual warm-up means writing 5–10 real emails a day to people who answer (colleagues, partners, existing customers), subscribing to a few newsletters and engaging with them, and rescuing anything that lands in junk. It is safe and free but slow, and hard to scale past a handful of inboxes.

Automated warm-up runs in a tool, usually priced by inbox, and can manage dozens of accounts at once. Many teams combine both: manual activity in week one, automated warm-up from week two, real conversations mixed in throughout. Warm-up can be shortened or skipped when a domain is over a year old, has a clean history and sends low volume with good engagement.

The debate about warm-up networks

Warm-up networks are useful but controversial. Their engagement is generated between controlled accounts, not by real prospects. Some practitioners argue that receivers increasingly recognize and discount it, and that automated warm-up may conflict with email platform terms; read your platform’s acceptable use policy before relying on it. Free warmup tools with shared pools of low-quality accounts can drag your reputation down instead of building it. A reasonable position: use warm-up to give a fresh account a starting history, prefer networks of real company inboxes, mix in genuine conversations, and judge success by real reply and spam-rate data rather than the tool’s own health score.

Running warm-up inside your sending tool

• Connect each inbox via OAuth or an app password (two-step verification usually has to be on; Microsoft 365 may need SMTP AUTH enabled).
• Set a minimum period of two weeks, a slow daily ramp, a daily ceiling and a reply rate target.
• Filter warm-up emails into a separate folder instead of deleting them or flagging them as spam. Both send exactly the negative signals you want to avoid.
• Review status regularly. Warm-up pauses if a connection breaks or credits run out, and reconnecting an account may restart the ramp from zero.
• If placement dips on a new inbox, keep warming and check for DNS or blocklist issues. On an established inbox, cut live sending first and pause only if the decline persists.
• Let warm-up finish before the first campaign. Starting early saves a week and can burn the account.

What warm-up can and cannot do

Warm-up sets an initial history. It cannot fix a bad list, broken DNS, a dirty shared IP or an irrelevant offer. Its effect fades fastest once campaigns start, because real recipients then provide the signals that count. Treat it as the foundation, not the strategy.

Sending after warm-up: volume, rotation and follow-ups

The move from warm-up to live outreach is where many setups get burned. Keep it controlled:

• Start below target (10–20 cold emails per inbox per day) and increase in small steps while metrics stay clean.
• Leave warm-up running at maintenance level next to campaigns; many teams keep as many warm-up emails as cold ones on each inbox.
• Rotate across accounts. Let the tool spread each campaign over all inboxes, and rest each domain in turn.
• Send in the recipient’s business hours, spaced out, on a consistent schedule.
• Scale by adding inboxes. Going past 70 emails a day means a second and third account, not one inbox at 150.

Follow-ups and testing without hurting placement

Follow-ups go to the same people from the same inbox, so if the opening email landed in junk, the follow-ups usually do too. Confirm that the opening touch reaches the inbox before adding steps. Keep sequences short (three or four follow-ups at most), space them by several days, bring new information each time and stop on any reply or opt-out. The B2B sales outreach guide shows how to build a multichannel sequence around these limits.

When you A/B test, judge variants by replies and meetings, not opens. Apple Mail Privacy Protection preloads images and many corporate filters fetch tracking pixels automatically, so cold email open rates are inflated and unreliable. Use seed-based placement tests to measure deliverability itself.

List hygiene, verification and bounce rate

Bad data kills reputation faster than volume does. Every email to a mailbox that does not exist is a hard bounce, and a high bounce rate is one of the clearest spammer signals: real senders know their recipients. A soft bounce is temporary (full mailbox, busy server) and can be retried, but repeated soft failures to one contact mean it should go.

• Verify every list before import with an email verification service, and verify again if it is older than about three months. People change jobs constantly.
• Hold the hard bounce rate down. A common rule of thumb among cold email teams is under 2% per campaign, with a pause above that.
• Treat catch-all and role contacts with care. Catch-all servers accept everything at first and may reject later; info@ and sales@ are shared inboxes where spam flags are more likely.
• Never buy scraped lists of unknown origin. They are full of dead contacts and spam traps. In the EU there is a legal problem too: source disclosure duties apply, and in Italy and Germany cold email to companies needs prior consent (the cold outreach lesson linked above lists the rules by country).
• Maintain a suppression list of opt-outs, people who flagged you and invalid contacts, applied across every tool and account.
• Never put many recipients in one email via CC or BCC; it resembles bulk mail and is a common reason for blocks.

Spam complaints come from relevance. An email to someone with a clear reason to care gets ignored at worst; an email to someone with no connection to the offer gets flagged.

Content factors: links, images, tracking and plain text

Content is filtered too, though less than reputation. Cold emails that reach the inbox tend to look like normal one-to-one correspondence:

• Plain text or minimal HTML. No newsletter templates, banners or heavy formatting.
• One link at most, often none in the opening email. No link shorteners and no links to unrelated sites.
• No images or attachments in the opening touch. They add weight and read as marketing.
• Open tracking off, or used sparingly. Tracking pixels are a known filter signal and give unreliable data anyway. Click tracking should run through your own custom tracking domain.
• A short, plain signature: name, role, company. No logo images, no row of social icons.
• Real personalization, not just variable swaps. Identical emails sent to hundreds of people form a fingerprint; a line written for each prospect breaks it.
• No spammy wording: all caps, exclamation marks, “free”, “guarantee”, “act now”.

Short, specific emails also earn more replies, which feeds reputation. The outreach email examples and templates show what that looks like in practice.

Monitoring deliverability: tools by category

Deliverability is a result that decays, not a state you reach once. Review these signals daily during warm-up and launch, and at least weekly afterwards.

Category	What it tells you	Examples
Mailbox provider dashboards	Spam rate, authentication and compliance status at that provider	Google Postmaster Tools (data appears only at meaningful daily volume); Outlook SNDS
Blocklist lookups	Whether your domain or IP appears on Spamhaus, Barracuda, SURBL and others	Multi-blocklist lookup services; each blocklist’s own lookup page
Placement (seed) tests	Where a sample email lands at each major provider: inbox, promotions or junk	Seed-list services, often bundled with sending platforms
DNS and header review	Whether SPF, DKIM and DMARC records are valid and pass	DNS lookup tools, “Show original” in your inbox, spam-score services
DMARC report analyzers	Which services send in your name, and whether they pass and align	DMARC aggregate (rua) parsers
Campaign performance	Bounce, unsubscribe and reply rates per inbox	Your sending tool’s analytics

Automation helps once you manage multiple accounts: per-inbox health alerts, automatic pausing of an inbox whose error rate jumps, caps and send windows set in the tool, time-zone scheduling, a unified inbox for replies and a CRM sync so no answer gets lost. Infrastructure services that automate DNS and pre-warm inboxes save time, but a pre-warmed account is not automatically safe: confirm its authentication and placement yourself before the first campaign.

Troubleshooting: when deliverability drops suddenly

When placement falls, work through a fixed order instead of changing five things at once. Cut or pause volume, then diagnose.

Symptom	Likely cause	Fix
Replies fall to near zero overnight, no error codes	Emails going to junk after a reputation hit	Run a placement test; check blocklists; halve sending or pause; review what changed (list, template, tool, volume)
Bounces with 550 5.7.1 or similar policy errors	Blocklisting, or the receiver refusing your mail	Find the listing, fix the cause, request delisting; read the full error text
Bounces with 550 5.7.515 at Outlook.com	Failing the SPF, DKIM or DMARC rules for high-volume senders	Fix authentication and alignment; confirm dmarc=pass in headers
Hard bounces above 2%	Old or unverified list, catch-all contacts	Stop the campaign; re-verify; remove catch-all and role contacts
dmarc=fail while SPF and DKIM pass	Alignment broken: the tool signs or sends under its own name	Set up custom DKIM and a custom return-path in the tool
SPF permerror	Two SPF records or more than ten DNS lookups	Merge into one record; remove unused includes
Warm-up stuck at 70–80% inbox	DNS problem or blocklist listing upstream	Recheck every record, the DKIM switch and blocklists before adding warm-up
Gmail fine, Outlook poor	Outlook judges reputation separately and more slowly	Send less to Outlook recipients; keep warming; measure Outlook placement on its own
Spam rate in Postmaster Tools near 0.3%	Irrelevant targeting or no easy opt-out	Narrow the audience; add a clear opt-out line and one-click unsubscribe; slow down

Recovery takes time. Authentication errors can be fixed in a day. Reputation needs several weeks of clean, reduced sending, and delisting can take from days to a manual review. If a domain stays poor after a month of careful sending, retire it or keep it for warm-up only, and shift the volume to fresh, properly warmed replacements. That is exactly why the portfolio has a buffer.

Compliance and deliverability work together

Every commercial email needs a clear sender identity and a working opt-out; in the US, CAN-SPAM also asks for a valid postal address and honoring opt-outs within 10 business days. In Europe the rules depend on the country: Italy and Germany demand prior consent for B2B email, so we run outreach there through LinkedIn and consent-based email only. A compliant program and a deliverable one look the same in practice: relevant recipients, honest identity, easy opt-out. If you would rather not build and run this infrastructure yourself, our B2B outreach and appointment setting service handles domains, authentication, warm-up and monitoring as part of each campaign.

Common mistakes

• Running cold campaigns from the main company domain.
• Rushing warm-up, or jumping from 10 to 200 emails a day.
• Two SPF records, more than ten lookups, DKIM published but not switched on, DMARC on the wrong host.
• Setting p=reject before every legitimate source passes.
• Ignoring DMARC reports and Postmaster Tools.
• Names like brand1.com, brand2.com, or too few domains carrying too much load.
• Unverified or bought lists and no suppression list.
• Heavy HTML, many links, images and open-tracking pixels in an opening email.
• Long sequences that keep going after the opening email landed in junk.
• Cheap warm-up pools, and treating the tool’s health score as the truth.
• No regular review until replies stop.

Cold email domain setup checklist

1. Calculate daily volume, inboxes and domains, plus a 20–30% buffer.
2. Buy brand-related domains on established TLDs; check each one on blocklists and archives.
3. Set a 301 redirect from each domain to the main website.
4. Connect each one to an email platform; publish MX records; create 2–5 inboxes per domain with real names and plain signatures.
5. Publish one SPF record (under ten lookups), DKIM with a 2048-bit key (and switch signing on), and DMARC at p=none with a reporting mailbox.
6. Set up custom DKIM and a custom tracking domain in your sending tool, so everything aligns with the From: line.
7. Send test emails and confirm spf=pass, dkim=pass, dmarc=pass.
8. Run warm-up for at least two to four weeks with a slow ramp; mix in real conversations.
9. Verify the prospect list, build a suppression list and make sure the opt-out works.
10. Start at 10–20 cold emails per inbox per day with a hard cap in the tool; leave warm-up on.
11. Track invalid-contact rate, spam rate, replies and placement daily at launch, then weekly; move DMARC toward p=quarantine once reports are clean.
12. Rotate and rest domains; retire damaged ones and replace them from the buffer.

Key takeaways: inbox placement matters more than volume; scale with more inboxes, not higher per-inbox limits; authentication is the entry ticket, not a guarantee; warm-up builds a foundation, not a strategy; and deliverability needs ongoing maintenance to improve and hold.

Next steps

Deliverability is the foundation; the rest of the outreach program sits on top of it. Continue with:

• Outreach metrics and KPIs to track reply and meeting rates per campaign.
• LinkedIn outreach for the channel that carries outreach in markets where cold email needs consent.
• Appointment setting for turning replies into booked meetings.

If you want the infrastructure, warm-up and campaigns run for you, talk to our B2B outreach team about lawful markets and channels for your offer.