AI Assistant (Leat MCP)
July 20
→
Journey automation
Respond to what a customer does, the moment they do it. Trigger messages, rewards, and actions from a purchase, a lapse, a tier change, or an expiring balance, with no calendar to wait for. Because a journey step can hand its decision to Decisioning, the same automation can run on a fixed rule or on a per-customer judgment, and you choose which.
A purchase, a visit, a signup, a tier change, a stamp card completed, a referral converted, a form submitted, a reward redeemed, points approaching expiry, a segment entered or left. Absence too: no visit in longer than the customer's usual gap, a balance untouched for months. Any event on the record can start a journey, and events from your other systems can start one through a webhook.
A journey can send an email or a wallet notification. It can issue points, a voucher, a reward, or credit. It can add the customer to a segment, update an attribute, flag them for review, prompt staff at their next visit, or fire a webhook to another system. Or, instead of specifying the action, it can hand the step to Decisioning, which chooses the best action for that customer from everything permitted, including none.
A journey is as simple or as structured as it needs to be. Trigger, wait three days, check whether they've visited, and if not, act. Branch by tier, by segment, by what they did in the last step. Exit a customer the moment they do the thing the journey was trying to cause, so nobody gets a win-back offer the day after they came back. Set a goal for the journey and it reports against it.
Every action a journey takes passes through Governance first, so consent, frequency caps, budgets, and eligibility apply on every step. Test a journey against recent history before it goes live and see who would have entered and what it would have cost. Once live, see entries, drop-offs, completions, and conversions per step, and measure the journey's effect against a holdout under Optimization, so a journey that runs forever is one that has earned it.
Lapsed customer recovery
Spot fading visit patterns across locations before a customer quietly stops coming back
Same-visit rewards
Reward a purchase while the customer is still in the store
Tier progression management
Move customers through tiers on complete spend rather than what one system saw
Post-purchase feedback
Ask how it went while the visit is still fresh, and act on what comes back
New customer acquisition offers
Reach first-time buyers with an offer that can't leak to existing customers
