Email your students actually receive.

This is the part we would lead with even if it were not the part we are best at, because it is the part that is usually broken.

The app reads your actual DNS and tells you what is wrong with it: SPF, DKIM and DMARC, with the record it found shown next to the record it expected. Then it scores the draft against the structural signals that push a message into Promotions, which is a different question from spam and is answered mostly by how the message is built rather than what it says. A preflight check verifies that your unsubscribe links resolve somewhere real before a campaign goes out, because a header pointing at the wrong host looks perfect and fails silently.

Contacts come in by CSV with your own columns mapped on the way, and each one carries a stage, tags, lists, a lead score and an activity timeline.

Segments are saved rules rather than saved lists. Combine stage, tag, list, source, email domain, date added, custom fields, lead score and whether someone clicked a tracked link recently. Each group matches all or any of its rules, and the count is worked out again at the moment a campaign launches.

• Fields you defineAdd your own (course, intake, college, whatever you sort people by) and use them in a segment rule, a template or a report.
• Segments you can see before you sendBuild from rules, preview exactly who matches, then save it to reuse. No sending to a definition you have not looked at.
• A stage for each contactLead, contacted, interested, applied, enrolled, or your own names for them, with the history of how each person moved.
• Suppression that actually holdsChecked when a campaign is queued and again for every message, and an import cannot override it.

Write it, test it against a real inbox, send it, then find out what it did.

Six steps: compose, recipients, senders, review, sending, report. The editor has a live preview, merge fields, reusable approved blocks and a test send. Reports carry what was sent and failed, clicks by link and who clicked, bounces, complaints and unsubscribes, and can resend to anyone who did not click or was not reached.

A campaign can carry a second subject line. A sample goes out first and the rest is held back until somebody picks the winner. Held means held: the remainder costs no worker time and takes no share of your send rate while it waits.

Writing Studio drafts academic and business emails from a short brief, with AI assistance you switch on and a monthly usage limit you set.

What comes back is checked before you see it: against your terminology list, against the links you allow, and for claims that were not in your brief. With the assistant switched off, the rest of the product works exactly the same.

An automation is a trigger and a list of steps. Each step is a template and a delay, and each one sends as an ordinary campaign you can open and read.

People leave a sequence when they unsubscribe, and you can also have them leave when they click or reach a stage, so nobody is still being chased after they have applied.

• Triggers that are easy to reason aboutAdded to a list, tagged, moved into a stage, sent a campaign, clicked, did not click, or added by hand.
• Switching one on does not mail everybodyActivation records a moment, and the sweep only considers what happened after it. Turning on a welcome sequence cannot blast your entire back catalogue.
• Sending survives a closed laptopA queue and a worker handle throttling and retries, so a large campaign does not depend on a browser tab.
• Live operationsEvery campaign sending right now, how fast, and anything stuck, with an activity log behind it.