SendX SMTP Settings and Configuration (2026)

If you came for a host and port to paste into WP Mail SMTP or Postfix, there is nothing to paste and no upgrade that creates one. What follows is the job you actually had: authenticating your domain so SendX mail passes DKIM and DMARC, sending transactional email the way SendX supports it, and picking a relay to sit beside SendX when your stack needs SMTP AUTH. Brevo, Postmark, Amazon SES, Resend and SMTP2GO all publish relay credentials. Among SendX’s peers, Brevo and Sender ship a relay; MailerLite and EmailOctopus are campaign-first like SendX.

Quick reference

Everything below comes from SendX’s own documentation. Where SendX publishes no value, the row says so rather than inventing one.

Why SendX has no SMTP relay

SendX is a marketing-first ESP priced on contacts rather than volume — unlimited sends from $9.99/month for 1,000 contacts. Its transactional capability was kept inside the same REST API rather than standing up a customer-facing MTA. An authenticated relay is a different business: abuse management, IP warming, credential compromise, a support queue full of TLS negotiation failures. One JSON endpoint gives SendX the same job with far less operational surface.

The second half is corporate. Parent company SendWorks already operates a dedicated infrastructure brand, SendPost, positioned verbatim as an Email API + SMTP Relay. SendX’s own navigation links to it through a Switch Products menu, which is exactly why people conclude a relay must exist somewhere in the dashboard. It does not. SendPost is a separate signup with its own docs and, per its FAQ, a demo plus upfront commitment before access. A SendX subscription provisions API keys, never SMTP credentials.

What you actually need instead

Sending transactional email

Use the Transactional Email add-on. Enable it under the profile icon at Settings › Add-ons › Transactional Email and take the Team API key from that add-on’s own Settings panel, not from your general API credentials. Your domain must be verified first. Then post JSON with from, to, subject and htmlBody; replyTo, textBody and custom headers are optional.

curl -X POST https://api.sendx.io/api/v1/rest/send/email \
  -H "X-Team-ApiKey: $SENDX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "from":    { "email": "hello@yourdomain.com", "name": "Your App" },
    "to":      [ { "email": "customer@example.com" } ],
    "subject": "Your receipt",
    "htmlBody": "<p>Your order is confirmed.</p>"
  }'

# 202 {"status":"success","message":"Email queued for delivery"}

Connecting a CMS or an app that only speaks SMTP

WP Mail SMTP, Easy WP SMTP, WooCommerce order mail, Postfix, an office scanner, a legacy ERP — anything whose setup screen asks only for host, port, username and password cannot be pointed at SendX, and no upgrade changes that, since full API access already ships on every paid tier. Two paths: use a plugin or library that speaks a provider’s HTTP API instead of SMTP, or run a real relay alongside SendX. For WordPress it is usually the second — site and WooCommerce mail on Brevo, Postmark or SMTP2GO, SendX for the newsletter.

Routing internal and system notifications

Monitoring alerts, cron failures, CI notifications and form handlers are classic SMTP consumers and the worst fit for a contact-priced ESP: every unique recipient counts against your contact tier though nobody subscribed. Route them through a volume-priced relay — Amazon SES is cheapest at scale, Resend and Mailgun stand up faster — on a subdomain such as alerts.yourdomain.com so reputation stays isolated.

Sending domain authentication

This is the real setup work SendX asks of you, and the reason most people arrive on a page like this. It is short, but SendX inverts two conventions that trip up experienced operators.

SPF and DKIM setup

Start with the part that surprises people: you do not add an SPF record for SendX. The official domain-authentication article states SPF is handled on SendX’s end, because mail leaves through infrastructure that is already authorised. There is no published SendX include: mechanism, and a guessed one is actively harmful: SPF is capped at 10 DNS lookups, and burning one can tip a crowded record into a permerror.

DKIM is where the work is, and SendX inverts the other convention. Most modern ESPs hand you CNAMEs that delegate DKIM; SendX’s docs say the record must be TXT and CNAME must not be used. Selector is sp-dkim. Copy the p= value from the dashboard rather than retyping it — it is per-domain, and one wrong character breaks signing silently.

# DKIM — required. Must be TXT, never CNAME.
Type:  TXT
Host:  sp-dkim._domainkey
       (some DNS providers want the fully qualified form:
        sp-dkim._domainkey.yourdomain.com)
Value: v=DKIM1;k=rsa;s=email;h=sha256;p=MIGfMA0GCSqGSIb3DQEBAQUAA4GNADCBiQKBgQC...
       ^ copy this verbatim from the SendX dashboard, per domain
TTL:   3600

# SPF — no SendX record required. Do not invent an include.
# Leave your existing record for other senders unchanged.

# DMARC — optional but recommended. Start permissive, then tighten.
Type:  TXT
Host:  _dmarc
Value: v=DMARC1; p=none; rua=mailto:dmarc@yourdomain.com; fo=1; pct=100

Propagation is usually minutes but can take up to 48 hours on slow registrars. Verify in the SendX UI once the record resolves — a verified domain is a hard prerequisite for the add-on, so an unverified one shows up as sends that never leave.

DMARC alignment

Because SendX signs with DKIM at sp-dkim._domainkey under your own domain, the signature’s d= value matches your From: domain and alignment passes on its own. That is the whole story here — there is no SPF return-path to align, because there is no customer SPF record. A message passing DKIM alignment satisfies DMARC regardless of SPF.

Publish DMARC at p=none with an rua address first. Read aggregate reports for 2 to 4 weeks, confirm SendX traffic shows as DKIM-aligned pass, then move to p=quarantine and eventually p=reject. SendX mandates no DMARC value of its own. If you run SendX beside a separate relay, tightening to p=reject before that relay is aligned will silently kill password resets while the newsletter keeps arriving.

Sending limits

SendX publishes almost no numeric limits. The table reflects what is documented and says so where nothing is. Do not fill the blanks from third-party comparison sites.

Common mistakes

Pairing SendX with an SMTP provider

The recommended two-provider setup

The standard architecture is boring and it works: SendX handles contact-based marketing, a dedicated relay handles everything the application generates. Split them at DNS level, not just account level — marketing from news.yourdomain.com, application mail from mail.yourdomain.com, each with its own DKIM key. A campaign that draws complaints then cannot drag password resets into the spam folder with it.

Pick the relay on operational fit, not price alone. Postmark when transactional deliverability and readable logs matter most; Amazon SES past a few hundred thousand messages, at the cost of more setup; SMTP2GO and Brevo as the easiest WordPress drop-ins, both pasteable into a plugin in under 5 minutes; Resend for teams building the send path in code. To consolidate on one vendor, Brevo is the closest SendX replacement that also ships a relay.

Keeping the two systems from conflicting

Three failure modes cause nearly every problem in a two-provider setup. First, double-sending: a SendX automation and a lifecycle job in your app both firing the welcome message. Decide which system owns each message type and make the other incapable of sending it rather than merely configured not to.

Second, DMARC. Each provider needs its own aligned authentication before you tighten policy. Add the relay’s records, wait for aggregate reports to show both sources passing, then move from p=none toward p=reject. Tightening with one source unaligned is the classic way a migration kills transactional mail while the newsletter keeps working.

Third, suppression drift. SendX holds marketing unsubscribes and bounces, the relay holds its own list, neither knows about the other — so an address the relay hard-bounced keeps receiving SendX campaigns until something syncs it back. Feed both sets of webhooks into one suppression table in your own database and reconcile both directions on a schedule. With no bulk update in the SendX API, plan that as one call per contact, off-peak.

Frequently asked questions