Skip to main content
Every sequence is created with an agent that writes its messages. Each email, LinkedIn note, and reply comes from the step’s prompt, the contact’s row, your business context, and the thread so far. The AI writing tab, after Builder, shows what the agent writes beside how it writes.
The AI writing tab writing for Satya Nadella at Microsoft: the Opener email the agent wrote, marked Written by the agent, with You approve it in Tasks under it because this sender always asks. Below it, a bar asks Would you send this to Satya? beside the Looks good and Needs work buttons. On the right, How it writes lists Goal, call to action, and replies, What it writes from, Advanced: effort and research, and Learnings, all shut.

The AI writing tab, with one contact's thread beside How it writes.

​
Open the AI writing tab

Click AI writing on a sequence, or Check the writing on its Builder, which also writes the first contact. How it writes, on the right, starts with every section shut:
  • Goal, call to action, and replies sets what the whole sequence is for, the link it points at, and the language. Its Write replies switch and label boxes choose which replies get an answer.
  • What it writes from holds Based on, which opens your business context beside the sequence, read only. Edit in Settings opens it in a new tab, so nothing you have not applied is lost.
  • Write from, in the same section, lists the research and AI columns the agent uses as angles, a different one in each email. A contact with one filled in skips web research. Untick a column to leave it out.
  • Advanced: effort and research picks how much thinking each message gets, and whether the first message may read the web, the contact’s LinkedIn profile, or both.
  • Learnings lists what the agent learned from Needs work and from your edits. The ones every sequence follows are marked All sequences. See Learnings.
  • The Agent actions menu, beside How it writes, holds Copy setup from another sequence. Once the sequence’s agent exists, it adds Version history.

​
Start from a campaign

  • On Sequences, under Ways to start, click one of the two campaigns. Each opens a new sequence with its steps and prompts in place
  • Email sends three emails. The first says the problem they likely have, who you solve it for, how it works, and what they get if they start now
  • The second says the same again, shorter, for someone who never read the first. The last says it in one or two sentences
  • Email and LinkedIn sends the same three emails, with a LinkedIn connection request after the first. The request goes with no note, and it is withdrawn if nobody accepts within 7 days
  • Each email’s prompt sits on its step, ready to edit

​
Write a prompt for a step

A prompt says what the message is for. Leave the wording to the agent.
  • Both campaigns start with three email prompts written for you. Edit those rather than starting from nothing
  • On an email that opens a thread, the menu beside Subject sets who writes it. Agent writes each one is the default and writes a fresh subject for every contact. Same for everyone is one subject you type, and the agent writes a body that goes out under it
  • A step that replies into an earlier email’s thread takes that thread’s subject. See Email steps

​
What the agent writes, and when

When a contact’s first message falls due, the agent drafts the rest of that contact’s planned messages in the same pass. Each draft reads the ones before it, so a follow-up builds on the opener instead of repeating it.
  • A step that only runs on a condition is left for its own turn
  • Editing or regenerating a message drops the drafts written after it, so the follow-ups answer your version
  • A draft is charged when its own step falls due, not when it is written

​
Check the writing for one contact

The AI writing tab writes one contact’s whole sequence, in order, and sends nothing.
  • With no audience yet, it asks for one first. Opened from the tab itself, nothing is written until you click Write the emails
  • We pick the contact: the row with the most on it. Change who it writes for in Writing for, then click Write the emails
  • Only the agent’s messages are here, each named by its step and marked Written by the agent. Steps you typed stay on the Builder tab
  • Contact details and Run activity sit shut under the buttons: the contact’s row, and what the agent did while writing
  • Every contact you check stays on the tab, up to ten. Pick one again in Writing for and their thread is still there, not written again
  • Emails Chat wrote after building the sequence are here too. A contact’s link in Chat opens this tab on those same emails
  • A warning above the thread says when their LinkedIn profile names a different company from their row, which usually means they changed jobs
  • Where a message offers the sender’s booking link and that sender has none, a No booking link chip stands in its place
  • A message that waits for your approval in Tasks says so under it, with why: what the agent was unsure of, or the setting that holds it, with Change to open that setting
  • Edit in Builder on each message opens its step’s Preview for the same contact
  • A message marked Looks good on its own carries a Looks good mark until the thread gets its verdict
  • Read the thread. Then click the thumbs up, Looks good, or the thumbs down, Needs work, in the bar pinned under it, or press G or N
  • Looks good makes that thread what sends to the contact. It is written again only if their row or the writing settings change, and a learning does not count
  • After a verdict, Next contact shows the next contact in Writing for that nobody has judged yet, and writes them if needed. When nobody is left, it says so

​
Say what’s wrong

  • Click Needs work to open the Feedback column, type what’s wrong, and click Rewrite. The agent writes the contact’s whole thread again with your note in mind
  • Needs work on a step’s Preview in the Builder opens the same column in place, and the rewrite shows here too
  • One note covers every message. To aim it at one message, start it with the step number, like “Step 3: shorter”. The learning then says which message it is about
  • While it rewrites, the messages shimmer in place. The column then lists the learnings it proposes, each for This sequence or All sequences, and any small edit to a step’s prompt
  • The new thread shows the new words first, marked where they changed. Show original opens the words they replace
  • “Looks good · saves 2” keeps every change still on. Discard feedback drops them and brings back the thread as it was
  • A contact whose feedback waits carries an Unfinished feedback chip in Writing for
  • Every contact the agent writes after a save follows the learning. It is listed under Learnings in How it writes

​
Research

Two kinds of research can run before the first message to a contact. A new sequence starts with both on.
  • With Web research, the agent reads the web about a person and their company. It skips this for a contact whose Write from column is filled in, because that research is already done
  • With Person research, it reads the contact’s LinkedIn profile: their headline, summary, and work history. This still runs when web research is skipped
  • What either one finds is kept for the rest of that conversation, so a contact is researched once and never again
  • One line under the switches prices the first message to a contact and each one after it. Switch either off to spend less

​
What sends, and what waits for you

The sender’s send mode decides whether a message goes on its own. On Send when confident, it goes when the agent rated it high and had enough data. The rest wait in Tasks. A message on the AI writing tab that will wait for you says so, and why. Every reply the agent writes waits for a person. See Reply with AI. Checks mark what they find on a message and never hold it back.

​
Know what the agent learns from

Needs work and your edits teach it:
  • Each note in a message’s Feedback column writes again what you marked and proposes short learnings, which you save when you decide the message
  • On the AI writing tab, it writes the contact’s thread again
  • On a message waiting in Tasks, it writes that message again on the same task
  • On a reply the agent drafted in the Inbox, it writes the draft again
  • On a message or reply that already went out, it shows what the agent would have written beside it and sends nothing
  • Your edit of one message keeps your words and joins its column as a note. Reject teaches nothing. Looks good and approvals save only the feedback on that message. Looks good never changes how confident the agent is

​
Good to know

  • A placeholder that never filled in, or an empty message
  • A figure, link, or address the agent could not find in the row, the research, or your context
  • LinkedIn naming a different company from the row, with both names
  • Banned jargon, such as “system of record”, or a sentence copied word for word from your business context, still there after one rewrite
  • Stray formatting a reader would see as it is, such as half a link or a note to the reviewer, still there after one rewrite
Type / in a step’s prompt, or in the goal, and pick a column of the audience table. The agent reads that column’s value for each contact.
Each step the agent writes carries a sparkle on its card. The AI writing card on the Builder names them. To write one yourself, open it and click Written by you.When every step is typed, the card and the AI writing tab offer Let the agent write the first email. It hands that step over and opens it, and Undo in the message that follows takes it back.The agent still writes the replies, so the AI writing tab keeps Replies. Learnings under it lists what the agent learned from Needs work on them.
Open a step the agent writes. It opens on Preview, on the contact in Writing for: their message written and waiting, sent, or the agent’s draft. Edit, then Edit for Satya only, pins your words for that one contact, and Let the agent write it again hands the contact back. Edit for everyone opens the step’s prompt.
A step’s default is the prompt a new sequence gives that step. Start from the default puts it back on that one step.
Your company’s own wording, such as spelling and what you call your product, is Voice in Business context. To make one sequence’s emails bolder, warmer, or shorter, say so in a step’s prompt. Or give Needs work on a message: the agent proposes a learning, saved when you approve the message or mark it Looks good.
Person research needs the audience’s LinkedIn profile URL column, set on the Audience tab. See Choose the audience. A contact whose row has no profile URL, or whose profile cannot be read, still gets a message. A read that never runs is never charged, so the price line quotes what you will actually pay.
You do not have to. The AI writing tab writes the contact’s whole thread, so each follow-up is read under the message it answers. A follow-up only makes sense against what came before it, so the message above it is the other half of the judgement.
Only if it says so. A learning names the messages it is about, such as “In connection requests, keep it to two sentences.” A note on a task is about that task’s step, and on a reply it is about replies.To widen or narrow one, reword it under Learnings. A learning never changes a step’s prompt. A note can propose a small prompt edit of its own, which saves with the rest.
At most five contacts’ messages per channel: written and waiting to send, waiting in Tasks, or being written. Emails and LinkedIn steps are counted apart, five each.A message you approved counts until it goes out. An approved email held for tomorrow or for its sending window does not count. The other contacts read Queued to write until a message ahead of them is sent or rejected in Tasks.The agent writes each email a few minutes before one of the sender’s mailboxes is free to send it, unless every email waits for approval. Words that would wait hours for tomorrow’s limit or the sending window are set aside. The email is written again at its turn, and you are charged once for it.While its messages wait in Tasks, the sequence says how many, such as “Needs approval: 5”. The Tasks count in the sidebar includes them too. Nothing else notifies you.
A line above the thread says which failed. Click Try again on it to write those again and leave the rest. A thread you are reading picks up where you left off after a refresh.When the whole write fails, Run activity names what was attempted, “Writing the messages”, with “Could not write this contact’s messages.” under it.
Yes. It reads the web and their LinkedIn profile the way the agent is set up to, so what you read is what would send.
The tab keeps each contact’s messages, what they were written from, and your verdict. A thread written by an older version of the agent is marked, and a write that fails keeps the thread it had.
Choose one of the five Goal presets, or choose Custom to enter your own goal. A step’s prompt opens when you click the step on the canvas; the reply prompt opens through Show prompt. Step edits autosave on drafts. On a running sequence they wait for Review changes in the header, and Check the writing opens that review first.The review lists each change beside the current version. An edited message reads as the contact sees it, with each variable as a named pill, such as Greeting, and the change marked in place.
The agent writes the waiting ones again now, for free, in Tasks or waiting to send. That is up to 5 emails and 5 LinkedIn steps per sequence, whenever you change a step’s prompt or how the agent writes on a running or paused sequence. Every other message it wrote that nobody has sent, edited, or approved is written fresh when its contact reaches that step.Review changes counts the waiting ones before you make the changes live. A message waiting in Tasks comes back with its new words. One you edited or approved keeps its words, and so do the follow-ups approved with an email’s thread and anything already sent.Saving a learning writes again the waiting messages it is about, with the same limit and also for free. It never touches anything you edited or approved.The agent writes a contact’s follow-ups ahead, with their first message. If their row changes before a follow-up is due, that follow-up is written again from what the row says then. So is one nobody has seen when the agent learns something before it is due.
No. Looks good on the AI writing tab decides what the agent sends that contact later. A message already written for them, or waiting in Tasks or the Inbox, stays as it is, and you approve it there.Nor does it make a message the agent was unsure of send on its own. Its confidence changes only with its instructions and the contact’s data.
No. Threads marked Looks good keep the words you approved when the agent learns something. Only a change to their row or to the writing settings writes them again.
Yes. Each contact written counts against the workspace’s free previews for the day, but a thread written again after a note does not. Once you judge one contact, the contact after the one on screen is written ahead, so the next press does not wait. Only one is written ahead.A step’s Preview shows how many are left under Write the email, and greys that button once none are. The count resets daily.
No. Each sequence has its own, with its own learnings, so a learning saved for This sequence never changes another. A learning saved for All sequences is followed by every sequence’s agent, and listed in Writing style. Copy setup from another sequence, in the Agent actions menu, copies a setup across as a starting point. Learnings stay with the original.To move a learning between the two, click Move to All sequences or Move to This sequence beside it under Learnings. In Writing style, Move to a sequence… moves one for all sequences to the sequence you pick.
Yes. Every change is a version. Version history, in the Agent actions menu, shows them and puts an earlier one back.
No. The body ends with the ask, and your signature signs it. When the ask is a meeting, it offers the link under Goal, call to action, and replies, or else the sender’s booking link. With neither, it asks for a reply.
Live messages read a bounded part of the row. The AI writing tab keeps the row’s full values and the research already run for that contact, earlier results included. A long value is summarized for writing, and the full value is kept.
It is told not to. A check marks a figure, link, or address it could not find in the row, the research, or your context. The mark shows on the message and holds nothing back, so read marked messages before choosing Always send.
It stays. Pinned words are yours, and the agent does not write over them until you hand the contact back with Let the agent write it again.

​
Next steps

Email steps

Switch a step to the agent and write its prompt.

Senders

Decide what sends without you.

Tasks

Approve, edit, or reject what waits for a person.

Business context

Tell the agent about your company, in one place for every sequence.