When you bring your own provider, deliverability becomes your responsibility and your asset. This guide covers what to get right, in the order it matters.
Authentication comes first
Nothing else helps if authentication is wrong.
SPF authorises the servers permitted to send for your domain. Your provider publishes the record to include; add it and keep it within the DNS lookup limit.
DKIM cryptographically signs your mail. Your provider generates the key pair and gives you the record. Sign with a 2048-bit key where offered.
DMARC tells mailbox providers what to do when SPF or DKIM fails, and — critically — where to send reports. Start at p=none with reporting on so you can see what is happening, then tighten to quarantine and eventually reject once you have confirmed all legitimate mail passes.
Alignment is the part people miss. DMARC requires that the domain in the visible From: address aligns with the domain that passed SPF or DKIM. A sender profile whose from-address does not align with a verified sending domain will authenticate technically and fail DMARC in practice.
In SendNectar, add each domain with an explicit purpose — sending, tracking, return_path, landing_page — and verify each one before use. A sender profile pins a from-address to a verified sending domain, which is how alignment is enforced at configuration time rather than discovered at send time.
Warm up, always
New IPs and new domains have no reputation, and no reputation is treated as suspicious. Sending your normal volume on day one is the single most common self-inflicted deliverability wound.
A workable ramp:
- Start at a few hundred messages a day.
- Roughly double every two to three days, watching bounce and complaint rates at each step.
- Reach full volume over two to four weeks, depending on your target.
- Lead with your most engaged contacts. Positive engagement signals early are what build the reputation the rest of the ramp depends on.
Warmup plans in SendNectar carry daily limits and growth targets so the ramp is enforced rather than remembered.
The metrics that predict trouble
Open rate is a lagging and increasingly unreliable indicator. The leading indicators are:
Bounce rate. Hard bounces above roughly 2% signal a list-quality problem. Above 5% you are actively damaging reputation. Distinguish hard bounces (permanent — suppress immediately) from soft bounces (temporary — retry, then suppress after repeats).
Complaint rate. The threshold that matters is far lower than people expect: 0.1% is a warning, 0.3% is trouble. Complaints are weighted heavily by every major mailbox provider.
Delivery rate. Accepted versus attempted. A gradual slide here, with bounces flat, usually means filtering rather than invalid addresses.
Engagement trend by domain. If Gmail placement slides while Outlook holds, the cause is usually specific to that provider’s filters, not your list overall.
SendNectar scores delivery, bounce, complaint, open, and click rates daily across three separate axes — per domain, per sender profile, and per provider connection. Three axes matter because they let you distinguish a bad list from a bad domain from a bad provider, which are three different fixes.
List hygiene is deliverability
The unglamorous work matters more than the technical configuration once the technical configuration is right.
- Use double opt-in. It costs you some volume at signup and buys you a list that engages. It also produces a confirmation event you can point to later.
- Never buy or rent a list. Aside from being unlawful in most jurisdictions, purchased lists are seeded with spam traps, and a single trap hit can undo months of warmup.
- Sunset the inactive. Contacts who have not engaged in six to twelve months should be moved to a re-engagement flow, then suppressed. Continuing to mail them lowers your engagement average and raises your trap risk.
- Honour unsubscribes instantly. In SendNectar this is not optional — bounce, complaint, and unsubscribe processing cannot be disabled on any plan.
- Cap frequency. Fatigue is the quiet cause of complaints. Layered caps per brand, topic, and message class are enforced across campaigns and automations alike.
Separate your streams
Marketing and transactional mail have different engagement profiles and different tolerance for delay. Mixing them means a promotional campaign’s complaint rate affects the reputation delivering your password resets.
Route them separately: a distinct provider connection and distinct sending routes for the transactional message class, ideally on a subdomain of its own. Transactional mail is also exempt from marketing frequency caps and topic consent gating — that separation only works if the routing reflects it.
When something goes wrong
Work in this order:
- Check authentication. Did a DNS change break SPF or DKIM? This is the most common cause of a sudden drop.
- Check the reputation axes. Is it one domain, one sender, or the provider? That narrows the cause immediately.
- Check what changed in the list. A recent import is the usual suspect — especially one where suppressions were not imported first.
- Check volume. Did a campaign spike well above your established pattern? Sudden volume changes trigger filtering independently of content.
- Check content only last. Content is what people investigate first and is rarely the cause.
A standing checklist
- SPF, DKIM, and DMARC published and aligned, per brand.
- All domains verified before use.
- Warmup plan active on anything new.
- Double opt-in on every capture form.
- Frequency caps defined at brand level.
- Suppressions imported before contacts, every time.
- Daily reputation reviewed weekly, not quarterly.
- Marketing and transactional routed separately.