Email that lands, and an inbox that tells you what happened.
Email Digit sends your marketing email and your app's email, and manages the authentication that decides whether either one arrives. Every reply that comes back is read, sorted and scored — so you find out what your email actually did, not how many pixels loaded.
Check your domain first
Free, no account. We'll tell you what's broken in your SPF, DKIM and DMARC.
Four free tools. No account, no cap.
Two products. Pick the one that's yours.
They share the same sending stack, the same contacts and the same inbox. They look nothing alike, because you don't. Choose at signup; switch any time.
I send campaigns to a list.
Broadcasts, automations, segments, and an editor that writes on your brand. Contacts are never counted or charged for. Your report leads with replies and revenue instead of open rates.
See the marketing product →For developersI send email from my app.
Password resets, receipts, onboarding sequences. Define the email once, then fire a trigger with one authenticated POST. Test keys sandbox everything. Every send is searchable.
See the developer product →Not sure? Start as a marketer. Nothing is locked, and the other product is one click away in the sidebar.
You bought a builder. You needed infrastructure.
Most email tools are excellent at the part you can see and indifferent to the part that decides everything. Drag-and-drop is solved. Authentication is not.
Gmail and Yahoo have required DMARC from bulk senders since 2024. Microsoft finished enforcing the same rules for Outlook, Hotmail and Live in late 2025, and non-compliant mail is refused outright rather than filed in junk. If SPF, DKIM and DMARC aren't right, nothing else you do matters — the best-designed campaign of your career gets a 550.
Then your reporting tells you 68% of people opened it, which is mostly a mail app opening it for them.
And when someone actually writes back, that reply lands in a shared inbox nobody has read since Tuesday.
> MAIL FROM:<news@yourbrand.com> < 250 OK > RCPT TO:<reader@gmail.com> < 550 5.7.26 Unauthenticated email is not accepted SPF: none · DKIM: none · DMARC: no policy > refused — not filed in junk. Refused.
So we built the authentication layer first, the reply layer second, and the editor on top.
Publish three DNS records once. Then never think about it again.
Add your domain. We give you three records — two CNAMEs and one TXT. You publish them in your DNS, and that's the last time you touch it.
From then on the moving parts are ours. Signing keys rotate on a schedule, with two selectors and a grace window, so mail already in flight still verifies while the new key takes over. Your DMARC policy walks from monitor-only toward enforcement based on what your aggregate reports actually show about alignment, not on a calendar. A brand-new domain gets a graduated daily ceiling that rises only on days its bounce and complaint rates are healthy — and anything over today's ceiling is held for tomorrow, never dropped.
We run our own authentication stack, so no third-party branding appears in your DNS, your message headers, or this dashboard.
How managed deliverability works →- CNAMEed1._domainkeyacme-com-a3f1.dkim.emaildigit.com
- CNAME_dmarcacme-com-a3f1.dmarc.emaildigit.com
- TXT@v=spf1 include:_spf.emaildigit.com ~all
Warm-up as a mechanism, not a checklist.
New domains start at a low daily volume and climb as they earn it. You can't forget it, because it isn't your job.
One-click unsubscribe on every marketing send.
Header-based, the way Gmail and Yahoo require. Opt-outs and hard bounces land on a do-not-send list that only ever gets stricter.
DMARC reports, read for you.
The aggregate XML your domain receives is parsed and turned into a plain answer about who is sending as you.
Hi — we were charged twice this month. Second time this has happened. Can someone sort this out today?
- party_type
- customer
- category
- billing
- intent
- complaint
- sentiment
- negative
- urgency
- high
- intensity
- 72 / 100
Every reply gets read. Including the ones nobody has time for.
When someone replies, we read it before you do. Not for a keyword — for what it is. Six independent readings: who's writing, what it's about, what they want, how they feel, how urgent it is, and how strongly they feel it. “Annoyed customer asking a billing question urgently” and “cheerful stranger pitching us SEO” are different situations, and one label can't tell them apart.
From those, one 0–100 priority number you can sort by, and an explanation of how it got there.
Then it drafts the reply. You pick the situation from a dropdown — there's no prompt to write — and get an editable draft in your brand's voice. Anything flagged for review or carrying a risk flag is draft-only: it will not send for you, however confident it looks.
And when it gets one wrong, you correct it once. That correction becomes a rule for that sender or that domain, inside your workspace. Nothing you correct trains a model shared with anyone else.
How reply intelligence works →We don't score open rates. Here's why that's the honest answer.
Apple Mail preloads images for most of its users. That fires your open pixel whether or not a human ever looked at the message. A large share of reported opens now come from software, not people.
So the industry's headline metric is mostly machines. We still record opens, and we still show them — de-emphasised and labelled unreliable. What we don't do is let them count. Not toward a contact's score, not into a segment, not as a branch in an automation, and not as the headline number on a report.
What we score is what took a decision: a click, a reply, a conversion. Weighted, and decayed by recency, so a contact who went quiet six months ago stops looking valuable.
This means our reports sometimes show smaller numbers than the tool you're leaving. Those are the numbers that were always true.
- Clicks0scored
- Replies0scored
- Conversions0scored
- Opens0unreliable · not scored
- Text
- Heading
- List
- Button
- Columns
- Divider
- Spacer
- Image
- Social
- Quote
- Menu
- Table
- Video
- HTML
Write it, or describe it and edit what comes back.
A block editor with the fourteen block types you expect — text, image, button, columns, tables, menus, video, quote — plus per-block spacing, backgrounds and a hide-on-mobile switch. Merge fields autocomplete as you type. Paste an existing HTML email and it comes in intact.
Or describe the email and get a draft. Point us at your website once and we read your colours, fonts and tone of voice, then apply them to everything we generate, including the email your app sends at 3am. If a site blocks automated access we say so and give you editable defaults, rather than inventing a brand and labelling it yours.
Output is email-safe HTML — Outlook, dark mode, mobile — and right-to-left languages render properly, not as broken left-aligned text.
Automations you can read at a glance.
A canvas of triggers, waits, branches and sends — on a segment, a tag, a reply or an API event. Goals count a conversion or a reply, never an open. Every run keeps a log that says what it did and what it skipped.
What we do that most sending tools don't.
Resend has the better developer experience and Mailchimp has more templates. We're not going to pretend otherwise. We win on deliverability you don't manage, replies you don't lose, and metrics that mean something.
Managed DKIM · SPF · DMARC
We rotate the keys and walk your DMARC policy toward enforcement for you, not just monitor it.
Reply classification
Six readings of every reply, an explainable 0–100 priority, and a drafted answer. Risky replies are never auto-sent.
Engagement without open rate
The score counts clicks, replies and conversions only. Opens are shown, never scored.
Right-to-left email
Arabic, Urdu and Hebrew in generated email and previews.
Generous free tier. Predictable beyond.
Paid plans are invoiced directly. Prices in USD. See every plan →
We're new. Here's the honest inventory.
We're in private beta and we don't have customer logos to show you, so here's the list instead.
- managed SPF, DKIM and DMARC with automatic key rotation and policy progression
- domain warm-up
- marketing broadcasts, segments and automations
- app-triggered email with test-mode keys
- reply reading, scoring and drafting
- a brand-aware editor and AI generation
- signed webhooks with retries
- the four free tools above
- client libraries in any language — the REST API is the API
- Zapier, n8n, Make and Salesforce connectors
- card checkout, so paid plans start with a short conversation
- a public status page
- no SOC 2 or ISO certification, and no audit underway
- moving from Mailchimp, Klaviyo, ActiveCampaign or HubSpot. We do it with you on a call and you approve it before anything lands.
- CSV import you can run yourself any time.
If something on this page turns out to be wrong, tell us and we'll fix the page.
Start with your domain, or start sending.
The free plan needs no card, and there's nothing to cancel. If your authentication is broken, the domain check will find it in about ten seconds and tell you exactly what to publish.
No card · unlimited contacts on every plan · managed authentication included