AI marketing team
An AI marketing team that asks before it acts
It works from the same guest record, consent and results as the rest of Engage, so it spots what a single-channel tool cannot. By default nothing reaches a guest or spends money until a person approves it.
What can I ask the AI team to do?
Type a request in plain English, such as "build me a win-back email for guests who have not been in since July". The AI team stages a plan of at most 10 steps in plain words, and shows any step you are not allowed to run as left out. Its drafts land in the normal builders, where you edit them like your own work. It acts with your permissions through the same actions as the screens, so its sends meet the same checks as yours.
- Requests in plain English, turned into a plan of at most 10 steps
- Drafts staged in the normal builders, and editable there
- Runs with your own permissions, never more
- Many actions undo in one click; sends and publishes cannot, and the preview says so
How do approvals work?
By default, anything that reaches a guest or spends money stops for a person to approve: a send, a form publish, a webhook, a review reply or an ad launch. The preview shows exactly what will happen, and the approval belongs to that exact action. A failed send is never retried on its own.
- A preview with the recipient count after consent, named sample recipients, the evidence, what cannot be undone and side effects such as a tag that starts a journey
- Approvals are single-use, void if a word changes, expire after 24 hours and need a signed-in person, never an API key or an AI tool
- Ads always stop for a person: the exact campaign and its spend are approved every time
How does Engage find regulars who have gone quiet?
Every night Engage looks for guests with three or more visits in 90 days and none in the last 30, and raises them with the rows behind them. That is one of nine fixed checks, plus a tenth once a week, each comparing your account with its own baseline. They are rules, not guesses, and a model never decides whether something happened. Accounts above 50,000 contacts are checked venue by venue. Checks that read visits need a connected WiFi system. A venue without one still gets campaign, deliverability, journey and calendar ideas.
- Lapsed regulars: three or more WiFi visits in 90 days, none in the last 30
- Welcome gap: new WiFi sign-ups with no welcome message within seven days
- Return-rate drop: the 45-day return rate well below its normal range (needs about 10 months of history)
- Campaign outlier: human opens or clicks far from usual within 72 hours of a send
- Send-time mismatch: recent sends against the hours this audience opens
- Journey leak: completion drops sharply at one step
- Deliverability risk: bounces above 5%, spam complaints above 0.1% or a worsening trend
- Quiet segment: grew 20% or more in 60 days with no sends
- Calendar gap: an event within 14 days and nothing planned
- Weekly, at most once per venue: a lapsed-guest Meta ad draft for a person to approve, where a Meta account is linked and at least 1,000 lapsed guests can be uploaded after the holdout
How does the AI team raise ideas?
As ranked idea cards, each with the rows behind it. Cards rank by contacts affected, confidence and a category weight (deliverability weighs most), adjusted by what has worked for your account. The system writes every figure; a model may add one plain sentence with no number in it. A daily digest lists the top three, and a knowledge panel shows what the AI team knows about your venues, offers, dates and rules, each fact marked worked out, confirmed or your answer.
- Ranked cards you work on, snooze for seven days or archive
- The usual value, what happened, how unusual it is and the sample size, all written by the system
- A daily digest of the top three
- Staged email keeps a 5% held-back group, and what held up re-ranks similar ideas
What writing help does Engage give?
Subject lines and rewrites in your brand voice, which you accept or discard. The writing tools see only your words, brand voice and never-say list, never a contact, a segment or a count. Each account has a default ceiling of 20 million AI tokens a month, and reaching it stops drafting, never the nightly ideas. The scoring, voice and checks shared by every email, form and page are on the Creative Studio page.
- Subject-line suggestions and paragraph rewrites you accept or discard
- The writing tools see only your words, brand voice and never-say list, never guest records, with typed emails and phone numbers masked and the never-say list hard-checked
How much control can I hand over?
You choose how much the AI team does on its own, from working only when asked to staging a draft behind every idea. Each channel has its own setting, no channel inherits another's, and only a person who manages your Engage account can change one, never the AI team. A new account starts at Watch.
- Ask first: it works only when you ask
- Watch: idea cards and a digest, with nothing staged
- Draft ready: a draft staged behind each card, waiting for your approval
- Set per channel, and changed only by a person who manages your Engage account
How does autopilot get earned?
Earned, never switched on for you. A channel moves beyond Draft ready only after a long recorded run of drafts that people approved without editing, counting only approvals made with the preview open. Even then it runs only inside a playbook an owner or admin approved, and each run is announced and held before anything sends. Review replies follow a narrower rule of their own, set out on the Reviews page.
- Earned on the record: a long run of drafts approved unedited with the preview opened, where a rejection counts against it, never a setting alone
- A playbook is a standing approval with bounds (audience, content, schedule, contact caps and a hold window) that an owner or admin approves once; editing a bound voids it
- Each run is announced to the owner and every admin with a cancel link, then held before it sends; if nobody can be reached, nothing sends
- Ads never qualify, and an outside AI tool can never use a playbook
Questions
Can the AI team do more than I can?
No. It runs as the signed-in person with no extra permissions, so it can only do what that person can do in Engage. Approving a send always needs a signed-in person, never the AI team itself.
Can the AI team make numbers up?
No. The system computes every figure on a card from your own data. A model may add one plain sentence, labelled as the AI team's reading, and a checker rejects it if it carries a number, a direction or a cause.
What guest data does the AI see?
Engage's writing tools see only your words, brand voice and never-say list, with typed emails and phone numbers masked. Text that guests wrote is masked and treated as data, never as instructions. Calls go to one provider, Anthropic, through one gateway. An assistant you connect over MCP reads what the key's owner can read, including contact details, so connect only tools your data protection policy allows.
Do I still need a marketer?
Yes. No human marketer is bundled with Engage. The AI team watches the numbers, raises ideas and prepares the work; your people decide what goes out.
Start free with Purple Engage
The free plan is listed at 500 credits a month, with unlimited* contacts and no card.