Cold email deliverability is broken - here's how to actually fix it
You're sending 500–2,000 cold emails a week, opens look 'fine' on paper, and reply volume is a fraction of what your tool's benchmarks promised. The problem is almost never the message - it's that Google and Microsoft never showed it. Deliverability is a compounding, invisible tax on every outbound program, and most teams only notice it once the pipeline has already dried up.
You'll recognise this if…
Root causes, in order of impact
Sending from the main domain
One spam complaint on cold outreach can tank the domain your CEO uses for real email. Cold email must run on separate secondary domains.
DNS misconfigured
Missing SPF, DKIM, or DMARC - or DMARC set to p=none forever. Google and Microsoft now enforce these for bulk senders.
No warm-up or ramped too fast
Fresh domains blasted at 50/day on day one hit the spam folder immediately. Warm-up is 4–8 weeks, not a switch you flip.
List quality
Catch-all domains, role addresses (info@, sales@), and stale scraped data drive bounces above 3% - a hard deliverability killer.
Copy triggers filters
Links stacked in signature, tracking pixels, HTML formatting, 'free', 'guarantee', all-caps subject lines. Plain-text-looking beats designed every time.
Bombarding one seed inbox
Warm-up services that only route to Gmail miss half the story - Outlook/Microsoft has completely different filters.
Inbox placement by setup quality
% landing in primary inbox
Before vs after
Before - the status quo
- ✗Sending from yourcompany.com to 400 contacts a day
- ✗DMARC set to p=none, DKIM never checked
- ✗One warm-up tool, Gmail only
- ✗Open rates 20%, reply rate 0.4%
- ✗Nobody owns 'why did open rates drop'
After - with one accountable operator
- ✓Three secondary domains, 15–30 sends/day each, ramped over 6 weeks
- ✓SPF/DKIM/DMARC verified, DMARC on quarantine with reporting
- ✓Warm-up covers Gmail + Outlook, spam-rate tracked in Postmaster
- ✓Open rates 55–70%, reply rate 3–8%
- ✓Weekly deliverability report: sends, bounces, complaints, Postmaster status
The playbook
- 1
Buy 2–3 secondary sending domains
Alt spellings of your brand (getbrand.com, brandhq.io). Never send outbound from the money domain.
- 2
Set SPF, DKIM, DMARC on each
SPF include your ESP; DKIM 2048-bit; DMARC start at p=none with rua reporting, move to quarantine after 30 clean days.
- 3
Warm up for 4–8 weeks
Start at 5–10/day, ramp 5–10/day per week to a ceiling of 30–40/day per inbox. Never go above.
- 4
Scrub the list
Verify with two tools (NeverBounce + ZeroBounce), drop catch-alls, drop role addresses, keep bounce rate under 2%.
- 5
Plain-text-looking copy
No HTML templates, one link max, break signature into two lines, kill tracking pixels for the first touch.
- 6
Monitor weekly
Google Postmaster + Microsoft SNDS. Bounce, complaint, spam rate, IP reputation - one dashboard, reviewed every Monday.
One accountable operator, not a stack of vendors
- I've run deliverability weekly for U.S., U.K., UAE, and EU teams - not once as a one-off audit.
- I own the Postmaster dashboard, the DNS records, the warm-up ceilings, and the weekly report - you don't need a specialist tool subscription on top of a specialist agency.
- When a domain gets flagged, I diagnose and swap in a warm backup within 48 hours - not next quarter.
FAQs
How long does it take to fix deliverability?
The DNS and infrastructure side is a week. Reputation recovery on a burned domain is 6–12 weeks - often it's cheaper to retire the domain and warm up new ones in parallel.
Do I need a deliverability tool?
You need Google Postmaster (free), Microsoft SNDS (free), and a warm-up service. Paid 'deliverability platforms' are useful at scale but not a substitute for someone actually reading the reports weekly.
Can you do this without replacing my current agency?
Yes. I've run the deliverability layer while an agency owns copy and targeting - most agencies are relieved to hand it off.
What tools do you standardise on?
Instantly, Smartlead, or Apollo for sending; NeverBounce + ZeroBounce for verification; Google Postmaster + Microsoft SNDS for monitoring.
Related pain points
Email warm-up: what actually works, what's a waste of money
Every outbound tool advertises 'automated warm-up' as a solved problem. It isn't. Warm-up is a 4–8 week discipline - schedule, volume ramp, reply simulation, seed inbox mix - and skipping any step means new domains start their life in the spam folder and stay there.
DeliverabilityWhy cold emails go to spam - 11 reasons ranked by impact
Cold email lands in spam for a specific, diagnosable reason - never mystery. The problem is teams treat 'spam' as one thing when it's actually 11 different failure modes stacked. Fix the wrong one and nothing changes. Rank them by impact and fix them in order.
OutboundCold email best practices for 2026 - what changed, what still works
Every 'best practices' article from 2022 is now actively harmful. Google's sender rules, AI-generated inbox fatigue, and Microsoft's tightening in 2024–2025 quietly killed the tactics that used to work. The teams still growing pipeline through cold email in 2026 are running a very specific, much narrower playbook.
OutboundUnqualified leads: why your pipeline is full of tire-kickers
Sales keeps saying 'the leads are bad.' Marketing keeps saying 'sales isn't working them.' Both are half right. Unqualified leads are almost never a data problem alone - they're a targeting-plus-scoring-plus-handoff problem, and the fix has to touch all three.
Diagnose your version in 30 minutes
Every situation is a little different. Book a call - we'll map your specific version of this pain point and I'll tell you the shortest path to a fix.